MLP Pytorch
Published:
Library
import warnings
warnings.filterwarnings("ignore")
try:
from pytorch_model_summary import summary
from torchviz import make_dot
from einops.layers.torch import Rearrange, Reduce
import talos
except:
!pip install talos -q
!pip install torchviz -q
!pip install pytorch-model-summary -q
!pip install einops -q
!pip install torchmetrics -q
from pytorch_model_summary import summary
from torchviz import make_dot
from einops.layers.torch import Rearrange, Reduce
import talos
from torchmetrics import HingeLoss
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import torch.utils.data as data
from torch import nn, einsum
import torchvision.transforms as transforms
import torchvision.datasets as datasets
from torch.utils.data import Dataset, DataLoader
from sklearn import metrics,decomposition,manifold
from sklearn.model_selection import train_test_split
from tqdm.notebook import trange, tqdm
import matplotlib.pyplot as plt
import numpy as np
import os
import copy
import random
import time
import pickle
from functools import partial
import pandas as pd
#fix seed
SEED = 42
random.seed(SEED)
np.random.seed(SEED)
torch.manual_seed(SEED)
torch.cuda.manual_seed(SEED)
torch.backends.cudnn.deterministic = True
Prepare Data
#data root folder
ROOT = 'data/assignment_2'
# read train pickle file
with open(f'{ROOT}/train', 'rb') as file:
train_dict = pickle.load(file, encoding='bytes')
# read test pickle file
with open(f'{ROOT}/test', 'rb') as file:
test_dict = pickle.load(file, encoding='bytes')
# get train and test feature, label
X_train, y_train = train_dict[b'data'],train_dict[b'coarse_labels']
X_test, y_test = test_dict[b'data'], test_dict[b'coarse_labels']
print("Training sample: ",X_train.shape)
print("Testing sample: ",X_test.shape)
#number of labels
number_classes = len(set(y_train))
print("Number of classes: ",number_classes)
Training sample: (50000, 3072)
Testing sample: (10000, 3072)
Number of classes: 20
#reshape feature to image present
X_train = np.array(X_train.reshape(-1 ,3, 32, 32),dtype=np.uint8).transpose(0, 2, 3, 1)
X_test = np.array(X_test.reshape(-1 ,3, 32, 32),dtype=np.uint8).transpose(0, 2, 3, 1)
#normalize data
mean_rgb, std_rgb = [], []
for i in range(3):
mean = X_train[:,:,:,i].mean() / 255
mean_rgb.append(mean)
std = X_train[:,:,:,i].std() / 255
std_rgb.append(std)
print(f'Calculated mean: {mean_rgb}')
print(f'Calculated std: {std_rgb}')
Calculated mean: [0.5070751592371323, 0.48654887331495095, 0.4409178433670343]
Calculated std: [0.2673342858792409, 0.25643846291708816, 0.2761504713256834]
# transform data
train_transforms = transforms.Compose([
transforms.ToPILImage(),
transforms.RandomRotation(5),
transforms.RandomCrop(32, padding=2),
transforms.ToTensor(),
transforms.Normalize(mean=mean_rgb, std=std_rgb)
])
test_transforms = transforms.Compose([
transforms.ToPILImage(),
transforms.ToTensor(),
transforms.Normalize(mean=mean_rgb, std=std_rgb)
])
#dataset class
class Cifar100(Dataset):
def __init__(self, X, y , transform):
super().__init__()
self.X = X
self.y = torch.from_numpy(np.array(y)).long()
self.transform = transform
def __len__(self):
return len(self.X)
def __getitem__(self, idx):
X_batch = self.X[idx]
# augmentations
if self.transform:
X_batch = self.transform(X_batch)
X_batch = X_batch.permute(1, 2, 0)
if self.y is not None:
y_batch = self.y[idx]
return X_batch, y_batch
return X_batch
#create dataset
train_data = Cifar100(X = X_train, y = y_train,transform = train_transforms)
test_data = Cifar100(X = X_test, y = y_test, transform = test_transforms)
#print information data
print(f'Number of training examples: {len(train_data)}')
print(f'Number of testing examples: {len(test_data)}')
Number of training examples: 50000
Number of testing examples: 10000
#plot image
def plot_images(images):
n_images = len(images)
rows = int(np.sqrt(n_images))
cols = int(np.sqrt(n_images))
fig = plt.figure(figsize=(20, 15))
for i in range(rows*cols):
ax = fig.add_subplot(rows, cols, i+1)
ax.imshow(images[i].cpu().numpy(), cmap='bone')
ax.axis('off')
#sample plot
N_IMAGES = 9
images = [image for image, label in [train_data[i] for i in range(N_IMAGES)]]
plot_images(images)
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
#valid ratio
VALID_RATIO = 0.9
n_train_examples = int(len(train_data) * VALID_RATIO)
n_valid_examples = len(train_data) - n_train_examples
#train val split
train_data, valid_data = data.random_split(train_data,
[n_train_examples, n_valid_examples])
Xtrain, Xval, ytrain, yval = train_test_split(X_train, y_train, test_size=VALID_RATIO, stratify=y_train,random_state=SEED)
print(f'Number of training examples: {len(train_data)}')
print(f'Number of validation examples: {len(valid_data)}')
print(f'Number of testing examples: {len(test_data)}')
Number of training examples: 45000
Number of validation examples: 5000
Number of testing examples: 10000
#data loader for train/val/test
BATCH_SIZE = 64
train_iterator = data.DataLoader(train_data,
shuffle=True,
batch_size=BATCH_SIZE)
valid_iterator = data.DataLoader(valid_data,
batch_size=BATCH_SIZE)
test_iterator = data.DataLoader(test_data,
batch_size=BATCH_SIZE)
#Build Deep Learning Model
MLP Base
# create MLP model for RGB Image
class MLPRGB(nn.Module):
def __init__(self, input_dim, output_dim, units=64, act_fun=F.relu):
super().__init__()
self.act_fun = act_fun
self.input_fc = nn.Linear(input_dim, units)
self.hidden1 = nn.Linear(units, units)
self.hidden2 = nn.Linear(units, units)
self.hidden3 = nn.Linear(units, units)
self.hidden4 = nn.Linear(units, units)
self.hidden5 = nn.Linear(units, units)
self.output_fc = nn.Linear(units, output_dim)
def forward(self, x):
# x = [batch size, height, width,chanel]
batch_size = x.shape[0]
x = x.view(batch_size, -1)
# input layer
h_1 = self.input_fc(x)
h_1 = self.act_fun(h_1)
# 1st hidden layer
h_2 = self.hidden1(h_1)
h_2 = self.act_fun(h_2)
# 2nd hidden layer
h_3 = self.hidden1(h_2)
h_3 = self.act_fun(h_3)
# 3rd hidden layer
h_4 = self.hidden1(h_3)
h_4 = self.act_fun(h_4)
# 4th hidden layer
h_5 = self.hidden1(h_4)
h_5 = self.act_fun(h_5)
# 5th hidden layer
h_6 = self.hidden1(h_5)
h_6 = self.act_fun(h_6)
# output
y_pred = self.output_fc(h_6)
return y_pred
# create MLP model using MLP + Dropout
class MLPGRAY(nn.Module):
def __init__(self, input_dim, output_dim, units=64, act_fun=F.relu):
super().__init__()
self.act_fun = act_fun
self.dropout = nn.Dropout(p=0.19)
self.input_fc = nn.Linear(input_dim, units)
self.hidden1 = nn.Linear(units, units)
self.hidden2 = nn.Linear(units, units)
self.hidden3 = nn.Linear(units, units)
self.hidden4 = nn.Linear(units, units)
self.hidden5 = nn.Linear(units, units)
self.output_fc = nn.Linear(units, output_dim)
def forward(self, x):
# x = [batch size, height, width, chanel]
batch_size = x.shape[0]
x = x.view(batch_size, -1)
# input layer
h_1 = self.input_fc(x)
h_1 = self.act_fun(h_1)
# 1st hidden layer
h_2 = self.hidden1(h_1)
h_2 = self.act_fun(h_2)
h_2 = self.dropout(h_2)
# 2nd hidden layer
h_3 = self.hidden1(h_2)
h_3 = self.act_fun(h_3)
h_3 = self.dropout(h_3)
# 3rd hidden layer
h_4 = self.hidden1(h_3)
h_4 = self.act_fun(h_4)
h_4 = self.dropout(h_4)
# 4th hidden layer
h_5 = self.hidden1(h_4)
h_5 = self.act_fun(h_5)
h_5 = self.dropout(h_5)
# 5th hidden layer
h_6 = self.hidden1(h_5)
h_6 = self.act_fun(h_6)
h_6 = self.dropout(h_6)
# output
y_pred = self.output_fc(h_6)
return y_pred
ResMLP
#support classes
def pair(val):
return (val, val) if not isinstance(val, tuple) else val
# classes
class Affine(nn.Module):
def __init__(self, dim):
super().__init__()
self.g = nn.Parameter(torch.ones(1, 1, dim))
self.b = nn.Parameter(torch.zeros(1, 1, dim))
def forward(self, x):
return x * self.g + self.b
class PreAffinePostLayerScale(nn.Module): # https://arxiv.org/abs/2103.17239
def __init__(self, dim, depth, fn):
super().__init__()
if depth <= 18:
init_eps = 0.1
elif depth > 18 and depth <= 24:
init_eps = 1e-5
else:
init_eps = 1e-6
scale = torch.zeros(1, 1, dim).fill_(init_eps)
self.scale = nn.Parameter(scale)
self.affine = Affine(dim)
self.fn = fn
def forward(self, x):
return self.fn(self.affine(x)) * self.scale + x
#MLP model with idea from paper ResMLP: Feedforward networks for image classification with data-efficient training: https://arxiv.org/pdf/2105.03404.pdf
class ResMLP(nn.Module):
def __init__(self, input_dim, output_dim, act_fun=nn.GELU()):
super().__init__()
self.patch_size = 4
self.num_patches = (input_dim // self.patch_size) * (input_dim // self.patch_size)
self.dim = 128
self.wrapper = lambda i, fn: PreAffinePostLayerScale(self.dim, i + 1, fn)
self.rearrange = Rearrange('b (h p1) (w p2) c -> b (h w) (p1 p2 c)', p1 = self.patch_size, p2 = self.patch_size)
self.ln1 = nn.Linear((self.patch_size ** 2) * 3, self.dim)
self.patching = self.wrapper(0, nn.Conv1d(self.num_patches, self.num_patches, 1))
self.ln2 = self.wrapper(i, nn.Sequential(
nn.Linear(self.dim, self.dim*2),
act_fun,
nn.Linear(self.dim*2, self.dim)))
self.ln3 = nn.Sequential(Affine(self.dim),Reduce('b n c -> b c', 'mean'),nn.Linear(self.dim, output_dim))
def forward(self, x):
x = self.rearrange(x)
h1 = self.ln1(x)
h2 = self.patching(h1)
h3 = self.ln2(h2)
pred = self.ln3(h3)
return pred
# create model
INPUT_DIM = 32
OUTPUT_DIM = number_classes
model = ResMLP(INPUT_DIM, OUTPUT_DIM)
#number of params pytorch model
input_sample = torch.zeros((1, 32, 32, 3))
print(summary(model, input_sample, show_input=True))
---------------------------------------------------------------------------------
Layer (type) Input Shape Param # Tr. Param #
=================================================================================
Rearrange-1 [1, 32, 32, 3] 0 0
Linear-2 [1, 64, 48] 6,272 6,272
PreAffinePostLayerScale-3 [1, 64, 128] 4,544 4,544
PreAffinePostLayerScale-4 [1, 64, 128] 66,304 66,304
Affine-5 [1, 64, 128] 256 256
Reduce-6 [1, 64, 128] 0 0
Linear-7 [1, 128] 2,580 2,580
=================================================================================
Total params: 79,956
Trainable params: 79,956
Non-trainable params: 0
---------------------------------------------------------------------------------
MLPMixer
#MLPMixer idea from paper https://arxiv.org/pdf/2105.01601.pdf
pair = lambda x: x if isinstance(x, tuple) else (x, x)
class PreNormResidual(nn.Module):
def __init__(self, dim, fn):
super().__init__()
self.fn = fn
self.norm = nn.LayerNorm(dim)
def forward(self, x):
return self.fn(self.norm(x)) + x
def FeedForward(dim, expansion_factor = 4, dropout = 0., dense = nn.Linear, act_fun=nn.GELU()):
inner_dim = int(dim * expansion_factor)
return nn.Sequential(
dense(dim, inner_dim),
act_fun,
nn.Dropout(dropout),
dense(inner_dim, dim),
nn.Dropout(dropout)
)
class MLPMixer(nn.Module):
def __init__(self, input_dim, output_dim, act_fun=nn.GELU()):
super().__init__()
self.patch_size = 4
self.num_patches = (input_dim // self.patch_size) * (input_dim // self.patch_size)
self.dim = 256
self.expansion_factor = 4
self.expansion_factor_token = 0.5
self.dropout = 0.0
self.rearrange = Rearrange('b (h p1) (w p2) c -> b (h w) (p1 p2 c)', p1 = self.patch_size, p2 = self.patch_size)
self.ln1 = nn.Linear((self.patch_size ** 2) * 3, self.dim)
self.ln2 = nn.Sequential(
PreNormResidual(self.dim, FeedForward(self.num_patches, self.expansion_factor, self.dropout, partial(nn.Conv1d, kernel_size = 1),act_fun)),
PreNormResidual(self.dim, FeedForward(self.dim, self.expansion_factor_token, self.dropout, nn.Linear)),act_fun )
self.ln3 = nn.Sequential(nn.LayerNorm(self.dim),Reduce('b n c -> b c', 'mean'),nn.Linear(self.dim, output_dim))
def forward(self, x):
x = self.rearrange(x)
h1 = self.ln1(x)
h2 = self.ln2(h1)
pred = self.ln3(h2)
return pred
# create model
INPUT_DIM = 32
OUTPUT_DIM = number_classes
model = MLPMixer(INPUT_DIM, OUTPUT_DIM,act_fun=nn.ReLU())
#number of params pytorch model
input_sample = torch.zeros((1, 32, 32, 3))
print(summary(model, input_sample, show_input=True))
-------------------------------------------------------------------------
Layer (type) Input Shape Param # Tr. Param #
=========================================================================
Rearrange-1 [1, 32, 32, 3] 0 0
Linear-2 [1, 64, 48] 12,544 12,544
PreNormResidual-3 [1, 64, 256] 33,600 33,600
ReLU-4 [1, 256, 256] 0 0
PreNormResidual-5 [1, 64, 256] 66,432 66,432
LayerNorm-6 [1, 64, 256] 512 512
Reduce-7 [1, 64, 256] 0 0
Linear-8 [1, 256] 5,140 5,140
=========================================================================
Total params: 118,228
Trainable params: 118,228
Non-trainable params: 0
-------------------------------------------------------------------------
# #visualize model
# fig = plt.figure(figsize=(25, 25))
# y = model(input_sample)
# make_dot(y[0].mean(), params=dict(model.named_parameters()))
Training model
#metric to calculate model performance
def calculate_accuracy(y_pred, y):
top_pred = y_pred.argmax(1, keepdim=True)
correct = top_pred.eq(y.view_as(top_pred)).sum()
acc = correct.float() / y.shape[0]
return acc
#calculate timing
def epoch_time(start_time, end_time):
elapsed_time = end_time - start_time
elapsed_mins = int(elapsed_time / 60)
elapsed_secs = int(elapsed_time - (elapsed_mins * 60))
return elapsed_mins, elapsed_secs
#training step function
def train(model, iterator, optimizer, criterion, device):
epoch_loss = 0
epoch_acc = 0
model.train()
for (x, y) in tqdm(iterator, desc="Training", leave=False):
x = x.to(device)
y = y.to(device)
optimizer.zero_grad()
y_pred = model(x)
loss = criterion(y_pred, y)
acc = calculate_accuracy(y_pred, y)
loss.backward()
optimizer.step()
epoch_loss += loss.item()
epoch_acc += acc.item()
return epoch_loss / len(iterator), epoch_acc / len(iterator)
#validate step function
def evaluate(model, iterator, criterion, device):
epoch_loss = 0
epoch_acc = 0
model.eval()
with torch.no_grad():
for (x, y) in tqdm(iterator, desc="Evaluating", leave=False):
x = x.to(device)
y = y.to(device)
y_pred = model(x)
loss = criterion(y_pred, y)
acc = calculate_accuracy(y_pred, y)
epoch_loss += loss.item()
epoch_acc += acc.item()
return epoch_loss / len(iterator), epoch_acc / len(iterator)
###Hyperparameter Tuning
#parameter
p = {'model_name':['mlprgb','mlpgray','resmlp','mlpmixer'],
'activation':['relu', 'gelu'],
'optimizer': ['RAdam', 'Adam'],
'losses': ['crossentropy'],
'units':[64, 128],
'batch_size': [64, 128],
'epochs': [15]}
#experiment with mutil params
def mlp_cifar(x_train, y_train, x_val, y_val, params):
val_param = params.values()
val_param = [str(x) for x in val_param]
log_name = "_".join(val_param)
print("Experiment with params: ", log_name)
log_file = log_name + '.txt'
#create log file
if os.path.isfile(log_file):
logger = open(log_file, 'w')
else:
logger = open(log_file, 'a')
# takes in a module and applies the specified weight initialization
def weights_init_uniform_rule(m):
classname = m.__class__.__name__
# for every Linear layer in a model..
if classname.find('Linear') != -1:
# get the number of the inputs
n = m.in_features
y = 1.0 / np.sqrt(n)
m.weight.data.uniform_(-y, y)
m.bias.data.fill_(0)
class MLPCIFAR(nn.Module, talos.utils.TorchHistory):
def __init__(self, params):
super(MLPCIFAR, self).__init__()
self.model_name = params['model_name']
self.act_fun = params['activation']
self.units = params['units']
if self.model_name == 'mlprgb':
#activate function
if self.act_fun == 'relu':
self.activate = F.relu
else:
self.activate = F.gelu
self.model = MLPRGB(input_dim=32*32*3,output_dim=20, units=self.units, act_fun=self.activate)
elif self.model_name == 'mlpgray':
#activate function
if self.act_fun == 'relu':
self.activate = F.relu
else:
self.activate = F.gelu
self.model = MLPGRAY(input_dim=32*32*3,output_dim=20, units=self.units, act_fun=self.activate)
elif self.model_name == 'resmlp':
#activate function
if self.act_fun == 'relu':
self.activate = nn.ReLU()
else:
self.activate = nn.GELU()
self.model = ResMLP(input_dim=32,output_dim=20, act_fun=self.activate)
else:
#activate function
if self.act_fun == 'relu':
self.activate = nn.ReLU()
else:
self.activate = nn.GELU()
self.model = MLPMixer(input_dim=32,output_dim=20, act_fun=self.activate)
def forward(self, x):
x = self.model(x)
return x
model = MLPCIFAR(params)
model.apply(weights_init_uniform_rule)
# loss function
if params['losses'] == 'crossentropy':
criterion = nn.CrossEntropyLoss()
else:
criterion = HingeLoss(task="multiclass", num_classes=20)
#optimizer
if params['optimizer'] == 'Adam':
optimizer = optim.Adam(model.parameters())
else:
optimizer = optim.RAdam(model.parameters())
#move model to GPU for faster training
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model = model.to(device)
criterion = criterion.to(device)
#create dataset
train_data = Cifar100(X = x_train, y = y_train,transform = train_transforms)
valid_data = Cifar100(X = x_val, y = y_val,transform = train_transforms)
# data loader
train_iterator = data.DataLoader(train_data,
shuffle=True,
batch_size=params['batch_size'])
valid_iterator = data.DataLoader(valid_data,
batch_size=params['batch_size'])
# Initialize history of net
model.init_history()
#start training
best_valid_loss = float('inf')
for epoch in trange(params['epochs']):
start_time = time.monotonic()
train_loss, train_acc = train(model, train_iterator, optimizer, criterion, device)
valid_loss, valid_acc = evaluate(model, valid_iterator, criterion, device)
if valid_loss < best_valid_loss:
best_valid_loss = valid_loss
torch.save(model.state_dict(), f"cifar100_{log_name}.pt")
end_time = time.monotonic()
epoch_mins, epoch_secs = epoch_time(start_time, end_time)
# print and log training results
print(f'Epoch: {epoch+1:02} | Epoch Time: {epoch_mins}m {epoch_secs}s')
logger.write(f'Epoch: {epoch+1:02} | Epoch Time: {epoch_mins}m {epoch_secs}s')
print(f'\tTrain Loss: {train_loss:.3f} | Train Acc: {train_acc*100:.2f}%')
logger.write(f'\tTrain Loss: {train_loss:.3f} | Train Acc: {train_acc*100:.2f}%')
print(f'\t Val. Loss: {valid_loss:.3f} | Val. Acc: {valid_acc*100:.2f}%')
logger.write(f'\t Val. Loss: {valid_loss:.3f} | Val. Acc: {valid_acc*100:.2f}%')
# append history
model.append_loss(train_loss)
model.append_metric(train_acc)
model.append_val_loss(valid_loss)
model.append_val_metric(valid_acc)
#close log file
logger.close()
return model, model.parameters()
#start hyperparameter searching
search_object = talos.Scan(x=Xtrain,y=ytrain,x_val=Xval,y_val=yval,model=mlp_mnist, params=p, experiment_name="talos_output")
0%| | 0/64 [00:00<?, ?it/s][A[A
Experiment with params: mlprgb_relu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.993 | Train Acc: 6.90%
Val. Loss: 2.982 | Val. Acc: 9.14%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 27s
Train Loss: 2.884 | Train Acc: 9.81%
Val. Loss: 2.788 | Val. Acc: 11.98%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.735 | Train Acc: 14.64%
Val. Loss: 2.705 | Val. Acc: 15.71%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.657 | Train Acc: 16.93%
Val. Loss: 2.653 | Val. Acc: 17.56%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 27s
Train Loss: 2.603 | Train Acc: 18.22%
Val. Loss: 2.638 | Val. Acc: 17.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 27s
Train Loss: 2.568 | Train Acc: 20.33%
Val. Loss: 2.606 | Val. Acc: 18.59%
Training: 0%| | 0/79 [00:00<?, ?it/s]
0%| | 0/64 [04:46<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 28s
Train Loss: 2.523 | Train Acc: 21.62%
Val. Loss: 2.583 | Val. Acc: 18.90%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 27s
Train Loss: 2.506 | Train Acc: 22.31%
Val. Loss: 2.586 | Val. Acc: 19.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 27s
Train Loss: 2.480 | Train Acc: 23.18%
Val. Loss: 2.550 | Val. Acc: 20.98%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 27s
Train Loss: 2.456 | Train Acc: 23.62%
Val. Loss: 2.540 | Val. Acc: 20.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 27s
Train Loss: 2.445 | Train Acc: 23.60%
Val. Loss: 2.517 | Val. Acc: 21.85%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 28s
Train Loss: 2.415 | Train Acc: 25.10%
Val. Loss: 2.507 | Val. Acc: 22.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 26s
Train Loss: 2.382 | Train Acc: 25.28%
Val. Loss: 2.487 | Val. Acc: 22.91%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 26s
Train Loss: 2.355 | Train Acc: 26.60%
Val. Loss: 2.510 | Val. Acc: 22.80%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 27s
Train Loss: 2.343 | Train Acc: 26.64%
Val. Loss: 2.487 | Val. Acc: 22.74%
2%|▏ | 1/64 [06:57<7:18:21, 417.48s/it][A[A
Experiment with params: mlprgb_relu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 27s
Train Loss: 2.995 | Train Acc: 5.57%
Val. Loss: 2.994 | Val. Acc: 5.01%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 26s
Train Loss: 2.981 | Train Acc: 8.40%
Val. Loss: 2.951 | Val. Acc: 8.78%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 26s
Train Loss: 2.905 | Train Acc: 9.98%
Val. Loss: 2.834 | Val. Acc: 10.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.792 | Train Acc: 11.23%
Val. Loss: 2.759 | Val. Acc: 13.33%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 26s
Train Loss: 2.720 | Train Acc: 14.47%
Val. Loss: 2.719 | Val. Acc: 14.71%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 26s
Train Loss: 2.677 | Train Acc: 16.04%
Val. Loss: 2.697 | Val. Acc: 16.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 26s
Train Loss: 2.651 | Train Acc: 17.32%
Val. Loss: 2.661 | Val. Acc: 17.24%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 26s
Train Loss: 2.611 | Train Acc: 18.81%
Val. Loss: 2.661 | Val. Acc: 17.65%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 26s
Train Loss: 2.582 | Train Acc: 19.59%
Val. Loss: 2.636 | Val. Acc: 17.75%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 26s
Train Loss: 2.569 | Train Acc: 19.49%
Val. Loss: 2.612 | Val. Acc: 18.35%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 29s
Train Loss: 2.525 | Train Acc: 20.96%
Val. Loss: 2.601 | Val. Acc: 19.03%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 29s
Train Loss: 2.503 | Train Acc: 22.09%
Val. Loss: 2.584 | Val. Acc: 19.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 30s
Train Loss: 2.491 | Train Acc: 23.01%
Val. Loss: 2.559 | Val. Acc: 20.12%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 30s
Train Loss: 2.475 | Train Acc: 23.55%
Val. Loss: 2.533 | Val. Acc: 21.30%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.442 | Train Acc: 24.41%
Val. Loss: 2.518 | Val. Acc: 21.45%
3%|▎ | 2/64 [13:56<7:12:22, 418.43s/it][A[A
Experiment with params: mlprgb_relu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 30s
Train Loss: 2.986 | Train Acc: 9.24%
Val. Loss: 2.932 | Val. Acc: 11.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.787 | Train Acc: 13.23%
Val. Loss: 2.698 | Val. Acc: 16.14%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.640 | Train Acc: 18.51%
Val. Loss: 2.617 | Val. Acc: 18.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.566 | Train Acc: 20.37%
Val. Loss: 2.569 | Val. Acc: 19.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.517 | Train Acc: 21.72%
Val. Loss: 2.535 | Val. Acc: 21.09%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 28s
Train Loss: 2.466 | Train Acc: 22.25%
Val. Loss: 2.570 | Val. Acc: 21.12%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 28s
Train Loss: 2.445 | Train Acc: 24.27%
Val. Loss: 2.487 | Val. Acc: 22.80%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 28s
Train Loss: 2.400 | Train Acc: 25.83%
Val. Loss: 2.507 | Val. Acc: 22.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 28s
Train Loss: 2.357 | Train Acc: 27.53%
Val. Loss: 2.476 | Val. Acc: 23.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 28s
Train Loss: 2.346 | Train Acc: 27.37%
Val. Loss: 2.467 | Val. Acc: 23.90%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 28s
Train Loss: 2.320 | Train Acc: 27.75%
Val. Loss: 2.466 | Val. Acc: 24.02%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 28s
Train Loss: 2.286 | Train Acc: 29.27%
Val. Loss: 2.459 | Val. Acc: 23.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 27s
Train Loss: 2.259 | Train Acc: 29.55%
Val. Loss: 2.461 | Val. Acc: 24.45%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 27s
Train Loss: 2.231 | Train Acc: 30.50%
Val. Loss: 2.438 | Val. Acc: 25.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 27s
Train Loss: 2.208 | Train Acc: 30.93%
Val. Loss: 2.467 | Val. Acc: 24.91%
5%|▍ | 3/64 [21:09<7:11:57, 424.88s/it][A[A
Experiment with params: mlprgb_relu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 27s
Train Loss: 2.995 | Train Acc: 6.88%
Val. Loss: 2.991 | Val. Acc: 9.16%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 27s
Train Loss: 2.949 | Train Acc: 11.45%
Val. Loss: 2.862 | Val. Acc: 11.29%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 27s
Train Loss: 2.767 | Train Acc: 13.81%
Val. Loss: 2.720 | Val. Acc: 14.78%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.688 | Train Acc: 15.90%
Val. Loss: 2.665 | Val. Acc: 16.74%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 26s
Train Loss: 2.638 | Train Acc: 17.85%
Val. Loss: 2.633 | Val. Acc: 18.42%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 26s
Train Loss: 2.605 | Train Acc: 17.81%
Val. Loss: 2.605 | Val. Acc: 17.80%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 25s
Train Loss: 2.561 | Train Acc: 20.00%
Val. Loss: 2.575 | Val. Acc: 19.30%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 26s
Train Loss: 2.515 | Train Acc: 21.88%
Val. Loss: 2.562 | Val. Acc: 19.79%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 26s
Train Loss: 2.491 | Train Acc: 22.05%
Val. Loss: 2.548 | Val. Acc: 20.64%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 26s
Train Loss: 2.464 | Train Acc: 22.42%
Val. Loss: 2.523 | Val. Acc: 21.28%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 26s
Train Loss: 2.428 | Train Acc: 23.77%
Val. Loss: 2.534 | Val. Acc: 21.75%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 25s
Train Loss: 2.383 | Train Acc: 26.29%
Val. Loss: 2.510 | Val. Acc: 22.07%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 26s
Train Loss: 2.391 | Train Acc: 25.70%
Val. Loss: 2.481 | Val. Acc: 22.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 26s
Train Loss: 2.338 | Train Acc: 27.48%
Val. Loss: 2.497 | Val. Acc: 22.74%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 26s
Train Loss: 2.335 | Train Acc: 27.42%
Val. Loss: 2.473 | Val. Acc: 23.52%
6%|▋ | 4/64 [27:50<6:55:46, 415.77s/it][A[A
Experiment with params: mlprgb_relu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 27s
Train Loss: 2.891 | Train Acc: 9.02%
Val. Loss: 2.723 | Val. Acc: 13.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 27s
Train Loss: 2.676 | Train Acc: 15.98%
Val. Loss: 2.688 | Val. Acc: 15.88%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 27s
Train Loss: 2.604 | Train Acc: 18.47%
Val. Loss: 2.613 | Val. Acc: 18.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 26s
Train Loss: 2.546 | Train Acc: 20.71%
Val. Loss: 2.582 | Val. Acc: 19.51%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 27s
Train Loss: 2.504 | Train Acc: 21.56%
Val. Loss: 2.566 | Val. Acc: 20.41%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 27s
Train Loss: 2.464 | Train Acc: 23.06%
Val. Loss: 2.538 | Val. Acc: 21.40%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 27s
Train Loss: 2.437 | Train Acc: 23.89%
Val. Loss: nan | Val. Acc: 21.91%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 27s
Train Loss: 2.387 | Train Acc: 25.59%
Val. Loss: 2.516 | Val. Acc: 22.06%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 27s
Train Loss: 2.363 | Train Acc: 26.48%
Val. Loss: 2.499 | Val. Acc: 23.51%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 27s
Train Loss: 2.349 | Train Acc: 26.72%
Val. Loss: 2.472 | Val. Acc: 23.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 27s
Train Loss: 2.327 | Train Acc: 27.77%
Val. Loss: 2.478 | Val. Acc: 23.90%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 26s
Train Loss: 2.289 | Train Acc: 28.50%
Val. Loss: 2.474 | Val. Acc: 23.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 27s
Train Loss: 2.267 | Train Acc: 29.49%
Val. Loss: 2.478 | Val. Acc: 24.04%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 27s
Train Loss: 2.250 | Train Acc: 29.85%
Val. Loss: 2.467 | Val. Acc: 24.04%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 27s
Train Loss: 2.222 | Train Acc: 30.72%
Val. Loss: 2.466 | Val. Acc: 24.74%
8%|▊ | 5/64 [34:38<6:45:58, 412.86s/it][A[A
Experiment with params: mlprgb_relu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 26s
Train Loss: 2.951 | Train Acc: 6.95%
Val. Loss: 2.836 | Val. Acc: 11.03%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 26s
Train Loss: 2.742 | Train Acc: 14.32%
Val. Loss: 2.704 | Val. Acc: 15.26%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 26s
Train Loss: 2.646 | Train Acc: 17.11%
Val. Loss: 2.637 | Val. Acc: 17.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 26s
Train Loss: 2.583 | Train Acc: 19.34%
Val. Loss: 2.606 | Val. Acc: 19.13%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 26s
Train Loss: 2.527 | Train Acc: 21.23%
Val. Loss: 2.597 | Val. Acc: 19.69%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 26s
Train Loss: 2.511 | Train Acc: 22.19%
Val. Loss: 2.562 | Val. Acc: 20.34%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 26s
Train Loss: 2.483 | Train Acc: 22.91%
Val. Loss: 2.559 | Val. Acc: 20.29%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 26s
Train Loss: 2.453 | Train Acc: 24.53%
Val. Loss: 2.538 | Val. Acc: 21.24%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 26s
Train Loss: 2.422 | Train Acc: 24.84%
Val. Loss: 2.509 | Val. Acc: 22.05%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 26s
Train Loss: 2.398 | Train Acc: 24.82%
Val. Loss: 2.505 | Val. Acc: 22.13%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 26s
Train Loss: 2.371 | Train Acc: 25.90%
Val. Loss: 2.514 | Val. Acc: 22.27%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 26s
Train Loss: 2.372 | Train Acc: 25.53%
Val. Loss: 2.508 | Val. Acc: 22.19%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 26s
Train Loss: 2.335 | Train Acc: 27.38%
Val. Loss: 2.506 | Val. Acc: 22.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 26s
Train Loss: 2.313 | Train Acc: 27.40%
Val. Loss: 2.492 | Val. Acc: 23.32%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 25s
Train Loss: 2.296 | Train Acc: 28.16%
Val. Loss: 2.501 | Val. Acc: 23.67%
9%|▉ | 6/64 [41:17<6:34:32, 408.15s/it][A[A
Experiment with params: mlprgb_relu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 26s
Train Loss: 2.836 | Train Acc: 10.38%
Val. Loss: 2.737 | Val. Acc: 14.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 27s
Train Loss: 2.668 | Train Acc: 15.86%
Val. Loss: 2.660 | Val. Acc: 16.81%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 27s
Train Loss: 2.581 | Train Acc: 18.28%
Val. Loss: 2.630 | Val. Acc: 19.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.538 | Train Acc: 20.98%
Val. Loss: 2.605 | Val. Acc: 19.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 27s
Train Loss: 2.492 | Train Acc: 22.39%
Val. Loss: 2.557 | Val. Acc: 20.75%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 27s
Train Loss: 2.449 | Train Acc: 23.14%
Val. Loss: 2.535 | Val. Acc: 21.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 26s
Train Loss: 2.390 | Train Acc: 25.08%
Val. Loss: 2.491 | Val. Acc: 23.13%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 26s
Train Loss: 2.365 | Train Acc: 26.86%
Val. Loss: 2.484 | Val. Acc: 23.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 27s
Train Loss: 2.333 | Train Acc: 27.33%
Val. Loss: 2.480 | Val. Acc: 23.87%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 27s
Train Loss: 2.287 | Train Acc: 29.17%
Val. Loss: 2.472 | Val. Acc: 24.13%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 26s
Train Loss: 2.256 | Train Acc: 29.39%
Val. Loss: 2.511 | Val. Acc: 24.31%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 27s
Train Loss: 2.222 | Train Acc: 31.05%
Val. Loss: 2.495 | Val. Acc: 24.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 27s
Train Loss: 2.206 | Train Acc: 30.74%
Val. Loss: 2.496 | Val. Acc: 24.68%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 27s
Train Loss: 2.184 | Train Acc: 31.76%
Val. Loss: 2.471 | Val. Acc: 24.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 27s
Train Loss: 2.164 | Train Acc: 32.58%
Val. Loss: 2.486 | Val. Acc: 24.85%
11%|█ | 7/64 [48:06<6:28:05, 408.51s/it][A[A
Experiment with params: mlprgb_relu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 26s
Train Loss: 2.915 | Train Acc: 9.63%
Val. Loss: 2.805 | Val. Acc: 11.84%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 26s
Train Loss: 2.727 | Train Acc: 14.10%
Val. Loss: 2.694 | Val. Acc: 16.43%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 27s
Train Loss: 2.637 | Train Acc: 16.95%
Val. Loss: 2.641 | Val. Acc: 17.61%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.564 | Train Acc: 20.08%
Val. Loss: 2.588 | Val. Acc: 19.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 26s
Train Loss: 2.495 | Train Acc: 22.34%
Val. Loss: 2.582 | Val. Acc: 20.08%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 26s
Train Loss: 2.489 | Train Acc: 22.07%
Val. Loss: 2.527 | Val. Acc: 21.46%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 26s
Train Loss: 2.446 | Train Acc: 24.02%
Val. Loss: 2.521 | Val. Acc: 22.17%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 26s
Train Loss: 2.399 | Train Acc: 24.65%
Val. Loss: 2.540 | Val. Acc: 21.43%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 27s
Train Loss: 2.373 | Train Acc: 25.74%
Val. Loss: 2.504 | Val. Acc: 23.14%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 26s
Train Loss: 2.347 | Train Acc: 27.27%
Val. Loss: 2.485 | Val. Acc: 23.37%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 27s
Train Loss: 2.302 | Train Acc: 27.05%
Val. Loss: 2.479 | Val. Acc: 23.53%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 26s
Train Loss: 2.280 | Train Acc: 29.10%
Val. Loss: 2.464 | Val. Acc: 24.16%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 26s
Train Loss: 2.271 | Train Acc: 28.98%
Val. Loss: 2.477 | Val. Acc: 24.18%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 26s
Train Loss: 2.271 | Train Acc: 28.83%
Val. Loss: 2.472 | Val. Acc: 23.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 26s
Train Loss: 2.216 | Train Acc: 30.74%
Val. Loss: 2.474 | Val. Acc: 24.29%
12%|█▎ | 8/64 [54:49<6:19:38, 406.75s/it][A[A
Experiment with params: mlprgb_gelu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 27s
Train Loss: 2.993 | Train Acc: 7.14%
Val. Loss: 2.981 | Val. Acc: 8.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 27s
Train Loss: 2.938 | Train Acc: 8.80%
Val. Loss: 2.865 | Val. Acc: 11.32%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 27s
Train Loss: 2.750 | Train Acc: 13.67%
Val. Loss: 2.691 | Val. Acc: 16.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.641 | Train Acc: 16.93%
Val. Loss: 2.644 | Val. Acc: 17.10%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 27s
Train Loss: 2.579 | Train Acc: 19.80%
Val. Loss: 2.603 | Val. Acc: 19.03%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 26s
Train Loss: 2.537 | Train Acc: 20.95%
Val. Loss: 2.588 | Val. Acc: 18.85%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 27s
Train Loss: 2.504 | Train Acc: 22.31%
Val. Loss: 2.563 | Val. Acc: 20.00%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 28s
Train Loss: 2.489 | Train Acc: 22.17%
Val. Loss: 2.569 | Val. Acc: 20.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 27s
Train Loss: 2.457 | Train Acc: 23.42%
Val. Loss: 2.530 | Val. Acc: 21.36%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 27s
Train Loss: 2.432 | Train Acc: 23.62%
Val. Loss: 2.510 | Val. Acc: 21.96%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 28s
Train Loss: 2.392 | Train Acc: 25.79%
Val. Loss: 2.542 | Val. Acc: 22.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 27s
Train Loss: 2.377 | Train Acc: 25.55%
Val. Loss: 2.498 | Val. Acc: 22.68%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 27s
Train Loss: 2.350 | Train Acc: 26.58%
Val. Loss: 2.526 | Val. Acc: 22.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 27s
Train Loss: 2.325 | Train Acc: 27.49%
Val. Loss: 2.481 | Val. Acc: 23.10%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 28s
Train Loss: 2.308 | Train Acc: 27.57%
Val. Loss: 2.487 | Val. Acc: 22.98%
14%|█▍ | 9/64 [1:01:44<6:15:10, 409.29s/it][A[A
Experiment with params: mlprgb_gelu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 27s
Train Loss: 2.995 | Train Acc: 7.46%
Val. Loss: 2.994 | Val. Acc: 8.32%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 27s
Train Loss: 2.981 | Train Acc: 7.50%
Val. Loss: 2.946 | Val. Acc: 7.48%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 27s
Train Loss: 2.893 | Train Acc: 9.24%
Val. Loss: 2.838 | Val. Acc: 9.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 27s
Train Loss: 2.771 | Train Acc: 12.05%
Val. Loss: 2.731 | Val. Acc: 13.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 26s
Train Loss: 2.678 | Train Acc: 15.82%
Val. Loss: 2.676 | Val. Acc: 16.77%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 26s
Train Loss: 2.617 | Train Acc: 18.67%
Val. Loss: 2.643 | Val. Acc: 17.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 27s
Train Loss: 2.603 | Train Acc: 17.99%
Val. Loss: 2.628 | Val. Acc: 17.80%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 27s
Train Loss: 2.576 | Train Acc: 19.36%
Val. Loss: 2.615 | Val. Acc: 18.63%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 27s
Train Loss: 2.557 | Train Acc: 20.25%
Val. Loss: 2.583 | Val. Acc: 19.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 27s
Train Loss: 2.515 | Train Acc: 20.68%
Val. Loss: 2.569 | Val. Acc: 19.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 27s
Train Loss: 2.482 | Train Acc: 22.71%
Val. Loss: 2.558 | Val. Acc: 20.26%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 28s
Train Loss: 2.466 | Train Acc: 23.11%
Val. Loss: 2.545 | Val. Acc: 21.03%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 28s
Train Loss: 2.442 | Train Acc: 23.71%
Val. Loss: 2.535 | Val. Acc: 20.92%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 27s
Train Loss: 2.435 | Train Acc: 25.10%
Val. Loss: 2.541 | Val. Acc: 21.52%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 28s
Train Loss: 2.426 | Train Acc: 24.88%
Val. Loss: 2.508 | Val. Acc: 22.27%
16%|█▌ | 10/64 [1:08:38<6:09:43, 410.80s/it][A[A
Experiment with params: mlprgb_gelu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.982 | Train Acc: 8.52%
Val. Loss: 2.942 | Val. Acc: 10.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.790 | Train Acc: 12.01%
Val. Loss: 2.686 | Val. Acc: 16.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.639 | Train Acc: 17.70%
Val. Loss: 2.626 | Val. Acc: 17.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.571 | Train Acc: 19.94%
Val. Loss: 2.591 | Val. Acc: 19.25%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 30s
Train Loss: 2.531 | Train Acc: 21.06%
Val. Loss: 2.559 | Val. Acc: 20.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.492 | Train Acc: 22.77%
Val. Loss: 2.537 | Val. Acc: 21.38%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.432 | Train Acc: 24.47%
Val. Loss: 2.548 | Val. Acc: 21.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 29s
Train Loss: 2.409 | Train Acc: 24.78%
Val. Loss: 2.505 | Val. Acc: 22.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 30s
Train Loss: 2.374 | Train Acc: 25.91%
Val. Loss: 2.498 | Val. Acc: 22.41%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 30s
Train Loss: 2.359 | Train Acc: 27.08%
Val. Loss: 2.466 | Val. Acc: 23.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 30s
Train Loss: 2.324 | Train Acc: 28.03%
Val. Loss: 2.485 | Val. Acc: 23.33%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.276 | Train Acc: 28.86%
Val. Loss: 2.441 | Val. Acc: 24.81%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 30s
Train Loss: 2.239 | Train Acc: 29.92%
Val. Loss: 2.459 | Val. Acc: 25.08%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.248 | Train Acc: 29.61%
Val. Loss: 2.448 | Val. Acc: 25.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.203 | Train Acc: 30.81%
Val. Loss: 2.440 | Val. Acc: 26.02%
17%|█▋ | 11/64 [1:16:11<6:14:14, 423.67s/it][A[A
Experiment with params: mlprgb_gelu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.995 | Train Acc: 9.00%
Val. Loss: 2.992 | Val. Acc: 10.42%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.931 | Train Acc: 10.76%
Val. Loss: 2.831 | Val. Acc: 10.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 28s
Train Loss: 2.762 | Train Acc: 13.63%
Val. Loss: 2.711 | Val. Acc: 15.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 28s
Train Loss: 2.658 | Train Acc: 17.34%
Val. Loss: 2.647 | Val. Acc: 17.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 28s
Train Loss: 2.592 | Train Acc: 19.16%
Val. Loss: 2.624 | Val. Acc: 18.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 29s
Train Loss: 2.557 | Train Acc: 20.33%
Val. Loss: 2.591 | Val. Acc: 19.29%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 28s
Train Loss: 2.525 | Train Acc: 21.19%
Val. Loss: 2.567 | Val. Acc: 19.89%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 29s
Train Loss: 2.490 | Train Acc: 22.11%
Val. Loss: 2.577 | Val. Acc: 18.86%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 28s
Train Loss: 2.467 | Train Acc: 23.79%
Val. Loss: 2.521 | Val. Acc: 21.81%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 27s
Train Loss: 2.448 | Train Acc: 24.43%
Val. Loss: 2.514 | Val. Acc: 22.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 28s
Train Loss: 2.417 | Train Acc: 25.16%
Val. Loss: 2.492 | Val. Acc: 23.22%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 28s
Train Loss: 2.371 | Train Acc: 25.94%
Val. Loss: 2.476 | Val. Acc: 23.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 28s
Train Loss: 2.345 | Train Acc: 26.58%
Val. Loss: 2.465 | Val. Acc: 23.79%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 28s
Train Loss: 2.334 | Train Acc: 26.68%
Val. Loss: 2.483 | Val. Acc: 23.43%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 29s
Train Loss: 2.303 | Train Acc: 28.18%
Val. Loss: 2.476 | Val. Acc: 23.97%
19%|█▉ | 12/64 [1:23:24<6:09:36, 426.47s/it][A[A
Experiment with params: mlprgb_gelu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 30s
Train Loss: 2.867 | Train Acc: 10.90%
Val. Loss: 2.691 | Val. Acc: 16.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 30s
Train Loss: 2.644 | Train Acc: 17.90%
Val. Loss: 2.631 | Val. Acc: 17.91%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.557 | Train Acc: 20.61%
Val. Loss: 2.593 | Val. Acc: 19.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 30s
Train Loss: 2.517 | Train Acc: 20.77%
Val. Loss: 2.543 | Val. Acc: 21.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.470 | Train Acc: 22.43%
Val. Loss: 2.532 | Val. Acc: 21.70%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 30s
Train Loss: 2.439 | Train Acc: 24.13%
Val. Loss: 2.520 | Val. Acc: 21.87%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.416 | Train Acc: 25.30%
Val. Loss: 2.505 | Val. Acc: 22.83%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 30s
Train Loss: 2.380 | Train Acc: 26.05%
Val. Loss: 2.483 | Val. Acc: 23.96%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 29s
Train Loss: 2.359 | Train Acc: 26.96%
Val. Loss: 2.487 | Val. Acc: 23.66%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 30s
Train Loss: 2.339 | Train Acc: 27.23%
Val. Loss: 2.486 | Val. Acc: 23.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 29s
Train Loss: 2.303 | Train Acc: 28.26%
Val. Loss: 2.501 | Val. Acc: 23.36%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.289 | Train Acc: 29.11%
Val. Loss: 2.472 | Val. Acc: 23.79%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 29s
Train Loss: 2.265 | Train Acc: 29.55%
Val. Loss: 2.479 | Val. Acc: 24.18%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 30s
Train Loss: 2.250 | Train Acc: 29.69%
Val. Loss: 2.490 | Val. Acc: 24.06%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.209 | Train Acc: 31.51%
Val. Loss: 2.510 | Val. Acc: 24.29%
20%|██ | 13/64 [1:30:55<6:08:45, 433.84s/it][A[A
Experiment with params: mlprgb_gelu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.909 | Train Acc: 9.36%
Val. Loss: 2.792 | Val. Acc: 12.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.729 | Train Acc: 13.85%
Val. Loss: 2.699 | Val. Acc: 15.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 28s
Train Loss: 2.624 | Train Acc: 18.73%
Val. Loss: 2.666 | Val. Acc: 17.44%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.575 | Train Acc: 19.90%
Val. Loss: 2.578 | Val. Acc: 19.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.501 | Train Acc: 22.44%
Val. Loss: 2.540 | Val. Acc: 20.76%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 29s
Train Loss: 2.461 | Train Acc: 23.48%
Val. Loss: 2.538 | Val. Acc: 21.25%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 29s
Train Loss: 2.434 | Train Acc: 24.02%
Val. Loss: 2.523 | Val. Acc: 21.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 29s
Train Loss: 2.415 | Train Acc: 23.87%
Val. Loss: 2.488 | Val. Acc: 22.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 28s
Train Loss: 2.354 | Train Acc: 26.54%
Val. Loss: 2.500 | Val. Acc: 22.76%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 28s
Train Loss: 2.351 | Train Acc: 26.31%
Val. Loss: 2.485 | Val. Acc: 23.53%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 29s
Train Loss: 2.302 | Train Acc: 28.38%
Val. Loss: 2.480 | Val. Acc: 23.55%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 29s
Train Loss: 2.312 | Train Acc: 27.54%
Val. Loss: 2.458 | Val. Acc: 24.27%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 29s
Train Loss: 2.271 | Train Acc: 29.34%
Val. Loss: 2.462 | Val. Acc: 24.33%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 29s
Train Loss: 2.282 | Train Acc: 28.09%
Val. Loss: 2.457 | Val. Acc: 24.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 29s
Train Loss: 2.266 | Train Acc: 29.10%
Val. Loss: 2.453 | Val. Acc: 24.48%
22%|██▏ | 14/64 [1:38:13<6:02:41, 435.24s/it][A[A
Experiment with params: mlprgb_gelu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.824 | Train Acc: 11.99%
Val. Loss: 2.708 | Val. Acc: 16.35%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.639 | Train Acc: 17.78%
Val. Loss: 2.613 | Val. Acc: 18.46%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.561 | Train Acc: 21.08%
Val. Loss: 2.565 | Val. Acc: 19.73%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.502 | Train Acc: 21.58%
Val. Loss: 2.548 | Val. Acc: 21.00%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.453 | Train Acc: 23.42%
Val. Loss: 2.537 | Val. Acc: 21.69%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 29s
Train Loss: 2.421 | Train Acc: 25.04%
Val. Loss: 2.490 | Val. Acc: 23.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.374 | Train Acc: 25.55%
Val. Loss: 2.485 | Val. Acc: 23.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 29s
Train Loss: 2.325 | Train Acc: 26.94%
Val. Loss: 2.495 | Val. Acc: 22.97%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 29s
Train Loss: 2.282 | Train Acc: 28.40%
Val. Loss: 2.473 | Val. Acc: 24.35%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 29s
Train Loss: 2.227 | Train Acc: 29.59%
Val. Loss: 2.483 | Val. Acc: 24.57%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 30s
Train Loss: 2.220 | Train Acc: 30.06%
Val. Loss: 2.454 | Val. Acc: 25.00%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.192 | Train Acc: 31.13%
Val. Loss: 2.497 | Val. Acc: 24.10%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 29s
Train Loss: 2.148 | Train Acc: 31.49%
Val. Loss: 2.479 | Val. Acc: 25.22%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 30s
Train Loss: 2.131 | Train Acc: 32.42%
Val. Loss: 2.469 | Val. Acc: 25.99%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.121 | Train Acc: 33.29%
Val. Loss: 2.487 | Val. Acc: 25.51%
23%|██▎ | 15/64 [1:45:43<5:58:57, 439.54s/it][A[A
Experiment with params: mlprgb_gelu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.869 | Train Acc: 9.73%
Val. Loss: 2.797 | Val. Acc: 13.77%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.689 | Train Acc: 16.41%
Val. Loss: 2.657 | Val. Acc: 17.24%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 28s
Train Loss: 2.601 | Train Acc: 18.52%
Val. Loss: 2.602 | Val. Acc: 19.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.541 | Train Acc: 21.15%
Val. Loss: 2.583 | Val. Acc: 19.85%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.505 | Train Acc: 22.66%
Val. Loss: 2.540 | Val. Acc: 21.24%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 29s
Train Loss: 2.451 | Train Acc: 23.85%
Val. Loss: 2.538 | Val. Acc: 21.82%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 29s
Train Loss: 2.421 | Train Acc: 24.30%
Val. Loss: 2.529 | Val. Acc: 22.15%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 29s
Train Loss: 2.385 | Train Acc: 25.88%
Val. Loss: 2.482 | Val. Acc: 23.23%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 28s
Train Loss: 2.336 | Train Acc: 27.30%
Val. Loss: 2.476 | Val. Acc: 24.17%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 28s
Train Loss: 2.295 | Train Acc: 28.89%
Val. Loss: 2.477 | Val. Acc: 23.99%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 28s
Train Loss: 2.311 | Train Acc: 28.40%
Val. Loss: 2.477 | Val. Acc: 24.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 29s
Train Loss: 2.282 | Train Acc: 29.08%
Val. Loss: 2.480 | Val. Acc: 25.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 29s
Train Loss: 2.223 | Train Acc: 31.11%
Val. Loss: 2.496 | Val. Acc: 24.55%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 29s
Train Loss: 2.226 | Train Acc: 30.55%
Val. Loss: 2.458 | Val. Acc: 25.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.156 | Train Acc: 32.73%
Val. Loss: 2.459 | Val. Acc: 25.57%
25%|██▌ | 16/64 [1:53:01<5:51:21, 439.20s/it][A[A
Experiment with params: mlpgray_relu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.994 | Train Acc: 4.94%
Val. Loss: 2.989 | Val. Acc: 5.45%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 30s
Train Loss: 2.984 | Train Acc: 5.46%
Val. Loss: 2.974 | Val. Acc: 6.64%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.965 | Train Acc: 7.06%
Val. Loss: 2.951 | Val. Acc: 8.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.934 | Train Acc: 8.19%
Val. Loss: 2.877 | Val. Acc: 11.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.834 | Train Acc: 10.52%
Val. Loss: 2.760 | Val. Acc: 13.25%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 30s
Train Loss: 2.771 | Train Acc: 11.73%
Val. Loss: 2.727 | Val. Acc: 14.27%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.736 | Train Acc: 13.41%
Val. Loss: 2.694 | Val. Acc: 14.72%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 30s
Train Loss: 2.717 | Train Acc: 13.63%
Val. Loss: 2.670 | Val. Acc: 16.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 30s
Train Loss: 2.676 | Train Acc: 15.03%
Val. Loss: 2.657 | Val. Acc: 17.10%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 31s
Train Loss: 2.673 | Train Acc: 15.66%
Val. Loss: 2.642 | Val. Acc: 17.58%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 30s
Train Loss: 2.656 | Train Acc: 16.91%
Val. Loss: 2.626 | Val. Acc: 18.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.624 | Train Acc: 17.58%
Val. Loss: 2.600 | Val. Acc: 18.59%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.623 | Train Acc: 17.48%
Val. Loss: 2.602 | Val. Acc: 19.12%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.602 | Train Acc: 18.51%
Val. Loss: 2.575 | Val. Acc: 19.78%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.576 | Train Acc: 19.30%
Val. Loss: 2.580 | Val. Acc: 19.70%
27%|██▋ | 17/64 [2:00:47<5:50:19, 447.22s/it][A[A
Experiment with params: mlpgray_relu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.996 | Train Acc: 5.14%
Val. Loss: 2.995 | Val. Acc: 6.06%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.992 | Train Acc: 5.86%
Val. Loss: 2.983 | Val. Acc: 5.55%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.954 | Train Acc: 5.08%
Val. Loss: 2.919 | Val. Acc: 6.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.929 | Train Acc: 6.88%
Val. Loss: 2.893 | Val. Acc: 10.30%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 30s
Train Loss: 2.900 | Train Acc: 7.97%
Val. Loss: 2.873 | Val. Acc: 11.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 30s
Train Loss: 2.875 | Train Acc: 9.84%
Val. Loss: 2.842 | Val. Acc: 12.36%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.841 | Train Acc: 10.53%
Val. Loss: 2.788 | Val. Acc: 13.65%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 30s
Train Loss: 2.803 | Train Acc: 12.09%
Val. Loss: 2.732 | Val. Acc: 14.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 30s
Train Loss: 2.765 | Train Acc: 12.34%
Val. Loss: 2.715 | Val. Acc: 14.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 30s
Train Loss: 2.753 | Train Acc: 12.97%
Val. Loss: 2.704 | Val. Acc: 15.36%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 30s
Train Loss: 2.731 | Train Acc: 14.30%
Val. Loss: 2.678 | Val. Acc: 15.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.709 | Train Acc: 14.30%
Val. Loss: 2.665 | Val. Acc: 15.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 30s
Train Loss: 2.707 | Train Acc: 15.33%
Val. Loss: 2.655 | Val. Acc: 16.79%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 29s
Train Loss: 2.678 | Train Acc: 15.74%
Val. Loss: 2.646 | Val. Acc: 17.06%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.684 | Train Acc: 16.04%
Val. Loss: 2.636 | Val. Acc: 17.32%
28%|██▊ | 18/64 [2:08:20<5:44:06, 448.84s/it][A[A
Experiment with params: mlpgray_relu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 2.994 | Train Acc: 6.09%
Val. Loss: 2.983 | Val. Acc: 5.91%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.936 | Train Acc: 7.00%
Val. Loss: 2.883 | Val. Acc: 9.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 32s
Train Loss: 2.859 | Train Acc: 9.61%
Val. Loss: 2.787 | Val. Acc: 11.58%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.779 | Train Acc: 12.52%
Val. Loss: 2.726 | Val. Acc: 15.00%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.731 | Train Acc: 14.32%
Val. Loss: 2.672 | Val. Acc: 16.44%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.684 | Train Acc: 16.06%
Val. Loss: 2.635 | Val. Acc: 17.65%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.633 | Train Acc: 18.08%
Val. Loss: 2.617 | Val. Acc: 17.99%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 32s
Train Loss: 2.617 | Train Acc: 17.60%
Val. Loss: 2.579 | Val. Acc: 19.37%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 31s
Train Loss: 2.585 | Train Acc: 19.26%
Val. Loss: 2.582 | Val. Acc: 20.19%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 31s
Train Loss: 2.558 | Train Acc: 20.19%
Val. Loss: 2.559 | Val. Acc: 20.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.545 | Train Acc: 20.67%
Val. Loss: 2.544 | Val. Acc: 21.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 31s
Train Loss: 2.536 | Train Acc: 21.30%
Val. Loss: 2.533 | Val. Acc: 21.32%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.503 | Train Acc: 21.60%
Val. Loss: 2.529 | Val. Acc: 21.98%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.497 | Train Acc: 21.78%
Val. Loss: 2.523 | Val. Acc: 22.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.474 | Train Acc: 22.57%
Val. Loss: 2.510 | Val. Acc: 22.72%
30%|██▉ | 19/64 [2:16:21<5:43:50, 458.46s/it][A[A
Experiment with params: mlpgray_relu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 30s
Train Loss: 2.995 | Train Acc: 5.41%
Val. Loss: 2.993 | Val. Acc: 6.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.970 | Train Acc: 6.11%
Val. Loss: 2.929 | Val. Acc: 5.12%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.922 | Train Acc: 6.84%
Val. Loss: 2.891 | Val. Acc: 9.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.884 | Train Acc: 9.06%
Val. Loss: 2.820 | Val. Acc: 9.69%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.809 | Train Acc: 10.06%
Val. Loss: 2.786 | Val. Acc: 10.01%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 29s
Train Loss: 2.781 | Train Acc: 11.07%
Val. Loss: 2.749 | Val. Acc: 11.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.758 | Train Acc: 11.56%
Val. Loss: 2.719 | Val. Acc: 13.22%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 30s
Train Loss: 2.714 | Train Acc: 13.12%
Val. Loss: 2.714 | Val. Acc: 14.07%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 29s
Train Loss: 2.695 | Train Acc: 13.91%
Val. Loss: 2.680 | Val. Acc: 16.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 30s
Train Loss: 2.679 | Train Acc: 14.63%
Val. Loss: 2.665 | Val. Acc: 17.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 29s
Train Loss: 2.652 | Train Acc: 16.02%
Val. Loss: 2.646 | Val. Acc: 17.53%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.646 | Train Acc: 16.99%
Val. Loss: 2.619 | Val. Acc: 18.65%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 30s
Train Loss: 2.599 | Train Acc: 18.30%
Val. Loss: 2.607 | Val. Acc: 19.09%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 30s
Train Loss: 2.596 | Train Acc: 18.40%
Val. Loss: 2.587 | Val. Acc: 19.28%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.555 | Train Acc: 18.98%
Val. Loss: 2.569 | Val. Acc: 20.39%
31%|███▏ | 20/64 [2:23:51<5:34:22, 455.98s/it][A[A
Experiment with params: mlpgray_relu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 2.961 | Train Acc: 7.24%
Val. Loss: 2.831 | Val. Acc: 10.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.823 | Train Acc: 10.15%
Val. Loss: 2.764 | Val. Acc: 12.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.771 | Train Acc: 11.87%
Val. Loss: 2.711 | Val. Acc: 15.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 32s
Train Loss: 2.728 | Train Acc: 14.36%
Val. Loss: 2.665 | Val. Acc: 16.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.696 | Train Acc: 15.15%
Val. Loss: 2.649 | Val. Acc: 18.03%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.656 | Train Acc: 16.48%
Val. Loss: 2.632 | Val. Acc: 18.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.645 | Train Acc: 16.08%
Val. Loss: 2.625 | Val. Acc: 19.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.636 | Train Acc: 17.60%
Val. Loss: 2.591 | Val. Acc: 19.56%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.626 | Train Acc: 17.44%
Val. Loss: 2.585 | Val. Acc: 19.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 31s
Train Loss: 2.586 | Train Acc: 18.77%
Val. Loss: 2.562 | Val. Acc: 20.23%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 33s
Train Loss: 2.584 | Train Acc: 19.46%
Val. Loss: 2.557 | Val. Acc: 20.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.564 | Train Acc: 19.78%
Val. Loss: 2.551 | Val. Acc: 20.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.552 | Train Acc: 20.15%
Val. Loss: 2.559 | Val. Acc: 20.81%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 33s
Train Loss: 2.539 | Train Acc: 20.63%
Val. Loss: 2.535 | Val. Acc: 20.63%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 33s
Train Loss: 2.532 | Train Acc: 20.57%
Val. Loss: 2.531 | Val. Acc: 21.27%
33%|███▎ | 21/64 [2:31:59<5:33:43, 465.66s/it][A[A
Experiment with params: mlpgray_relu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.983 | Train Acc: 5.14%
Val. Loss: 2.943 | Val. Acc: 5.87%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.913 | Train Acc: 7.23%
Val. Loss: 2.855 | Val. Acc: 10.42%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.852 | Train Acc: 10.10%
Val. Loss: 2.794 | Val. Acc: 11.63%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 30s
Train Loss: 2.810 | Train Acc: 10.31%
Val. Loss: 2.748 | Val. Acc: 13.86%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.760 | Train Acc: 12.89%
Val. Loss: 2.715 | Val. Acc: 14.31%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.741 | Train Acc: 12.54%
Val. Loss: 2.707 | Val. Acc: 14.54%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.728 | Train Acc: 13.81%
Val. Loss: 2.695 | Val. Acc: 15.39%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.733 | Train Acc: 12.70%
Val. Loss: 2.692 | Val. Acc: 15.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.685 | Train Acc: 14.59%
Val. Loss: 2.663 | Val. Acc: 16.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.679 | Train Acc: 15.20%
Val. Loss: 2.642 | Val. Acc: 17.51%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 33s
Train Loss: 2.658 | Train Acc: 16.76%
Val. Loss: 2.637 | Val. Acc: 17.51%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.659 | Train Acc: 16.05%
Val. Loss: 2.617 | Val. Acc: 18.21%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.646 | Train Acc: 16.05%
Val. Loss: 2.622 | Val. Acc: 18.56%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.616 | Train Acc: 18.03%
Val. Loss: 2.610 | Val. Acc: 18.59%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.601 | Train Acc: 17.44%
Val. Loss: 2.592 | Val. Acc: 19.19%
34%|███▍ | 22/64 [2:39:56<5:28:17, 468.98s/it][A[A
Experiment with params: mlpgray_relu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 2.930 | Train Acc: 7.54%
Val. Loss: 2.796 | Val. Acc: 10.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.781 | Train Acc: 12.42%
Val. Loss: 2.734 | Val. Acc: 14.78%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 32s
Train Loss: 2.701 | Train Acc: 14.83%
Val. Loss: 2.666 | Val. Acc: 16.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 32s
Train Loss: 2.667 | Train Acc: 17.15%
Val. Loss: 2.639 | Val. Acc: 17.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.630 | Train Acc: 18.71%
Val. Loss: 2.595 | Val. Acc: 18.74%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 32s
Train Loss: 2.615 | Train Acc: 17.37%
Val. Loss: 2.585 | Val. Acc: 19.92%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 32s
Train Loss: 2.569 | Train Acc: 19.80%
Val. Loss: 2.567 | Val. Acc: 19.96%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 32s
Train Loss: 2.563 | Train Acc: 19.66%
Val. Loss: 2.538 | Val. Acc: 20.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.535 | Train Acc: 20.31%
Val. Loss: 2.550 | Val. Acc: 20.76%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.507 | Train Acc: 22.17%
Val. Loss: 2.522 | Val. Acc: 21.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.496 | Train Acc: 22.59%
Val. Loss: 2.522 | Val. Acc: 21.88%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.474 | Train Acc: 22.78%
Val. Loss: 2.511 | Val. Acc: 21.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.469 | Train Acc: 23.34%
Val. Loss: 2.518 | Val. Acc: 21.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.434 | Train Acc: 23.73%
Val. Loss: 2.500 | Val. Acc: 23.12%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.428 | Train Acc: 23.28%
Val. Loss: 2.495 | Val. Acc: 22.66%
36%|███▌ | 23/64 [2:48:04<5:24:29, 474.87s/it][A[A
Experiment with params: mlpgray_relu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.974 | Train Acc: 6.60%
Val. Loss: 2.932 | Val. Acc: 7.85%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.910 | Train Acc: 8.11%
Val. Loss: 2.863 | Val. Acc: 12.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 30s
Train Loss: 2.827 | Train Acc: 12.68%
Val. Loss: 2.759 | Val. Acc: 13.72%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 30s
Train Loss: 2.744 | Train Acc: 13.30%
Val. Loss: 2.701 | Val. Acc: 14.96%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 30s
Train Loss: 2.701 | Train Acc: 15.23%
Val. Loss: 2.670 | Val. Acc: 15.66%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 30s
Train Loss: 2.665 | Train Acc: 16.25%
Val. Loss: 2.647 | Val. Acc: 17.12%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.646 | Train Acc: 17.15%
Val. Loss: 2.614 | Val. Acc: 18.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.623 | Train Acc: 18.12%
Val. Loss: 2.599 | Val. Acc: 19.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 31s
Train Loss: 2.581 | Train Acc: 20.06%
Val. Loss: 2.580 | Val. Acc: 20.56%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 31s
Train Loss: 2.562 | Train Acc: 20.41%
Val. Loss: 2.569 | Val. Acc: 19.85%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 31s
Train Loss: 2.552 | Train Acc: 20.68%
Val. Loss: 2.553 | Val. Acc: 20.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 31s
Train Loss: 2.527 | Train Acc: 21.23%
Val. Loss: 2.538 | Val. Acc: 21.33%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.544 | Train Acc: 20.98%
Val. Loss: 2.549 | Val. Acc: 21.19%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.509 | Train Acc: 21.33%
Val. Loss: 2.523 | Val. Acc: 21.68%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.489 | Train Acc: 22.50%
Val. Loss: 2.522 | Val. Acc: 21.90%
38%|███▊ | 24/64 [2:55:51<5:14:59, 472.49s/it][A[A
Experiment with params: mlpgray_gelu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 2.995 | Train Acc: 6.49%
Val. Loss: 2.988 | Val. Acc: 9.54%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.941 | Train Acc: 8.50%
Val. Loss: 2.879 | Val. Acc: 11.15%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 32s
Train Loss: 2.869 | Train Acc: 10.76%
Val. Loss: 2.802 | Val. Acc: 13.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 32s
Train Loss: 2.793 | Train Acc: 12.34%
Val. Loss: 2.718 | Val. Acc: 16.12%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 33s
Train Loss: 2.747 | Train Acc: 14.36%
Val. Loss: 2.685 | Val. Acc: 16.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.717 | Train Acc: 14.81%
Val. Loss: 2.659 | Val. Acc: 17.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.692 | Train Acc: 16.48%
Val. Loss: 2.628 | Val. Acc: 18.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.661 | Train Acc: 17.17%
Val. Loss: 2.607 | Val. Acc: 18.82%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.651 | Train Acc: 17.72%
Val. Loss: 2.587 | Val. Acc: 19.90%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.630 | Train Acc: 18.47%
Val. Loss: 2.576 | Val. Acc: 20.28%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.623 | Train Acc: 18.85%
Val. Loss: 2.541 | Val. Acc: 21.23%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 31s
Train Loss: 2.567 | Train Acc: 19.64%
Val. Loss: 2.549 | Val. Acc: 20.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.584 | Train Acc: 20.17%
Val. Loss: 2.544 | Val. Acc: 20.75%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 30s
Train Loss: 2.544 | Train Acc: 20.59%
Val. Loss: 2.523 | Val. Acc: 21.88%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.534 | Train Acc: 21.93%
Val. Loss: 2.527 | Val. Acc: 21.80%
39%|███▉ | 25/64 [3:03:57<5:09:39, 476.39s/it][A[A
Experiment with params: mlpgray_gelu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 30s
Train Loss: 2.996 | Train Acc: 5.68%
Val. Loss: 2.995 | Val. Acc: 8.97%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 30s
Train Loss: 2.992 | Train Acc: 7.23%
Val. Loss: 2.984 | Val. Acc: 8.37%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 30s
Train Loss: 2.958 | Train Acc: 8.20%
Val. Loss: 2.914 | Val. Acc: 10.63%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 30s
Train Loss: 2.904 | Train Acc: 9.28%
Val. Loss: 2.845 | Val. Acc: 11.60%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 30s
Train Loss: 2.841 | Train Acc: 11.45%
Val. Loss: 2.769 | Val. Acc: 13.19%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 30s
Train Loss: 2.805 | Train Acc: 12.85%
Val. Loss: 2.736 | Val. Acc: 14.69%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.757 | Train Acc: 13.59%
Val. Loss: 2.716 | Val. Acc: 15.32%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 30s
Train Loss: 2.730 | Train Acc: 13.69%
Val. Loss: 2.686 | Val. Acc: 15.71%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 30s
Train Loss: 2.710 | Train Acc: 15.39%
Val. Loss: 2.671 | Val. Acc: 15.66%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 30s
Train Loss: 2.702 | Train Acc: 14.75%
Val. Loss: 2.655 | Val. Acc: 17.18%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 30s
Train Loss: 2.667 | Train Acc: 16.27%
Val. Loss: 2.638 | Val. Acc: 17.83%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 30s
Train Loss: 2.658 | Train Acc: 16.88%
Val. Loss: 2.628 | Val. Acc: 18.11%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 30s
Train Loss: 2.657 | Train Acc: 16.15%
Val. Loss: 2.624 | Val. Acc: 18.74%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 30s
Train Loss: 2.638 | Train Acc: 17.15%
Val. Loss: 2.596 | Val. Acc: 19.65%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.625 | Train Acc: 18.07%
Val. Loss: 2.596 | Val. Acc: 19.23%
41%|████ | 26/64 [3:11:33<4:57:46, 470.17s/it][A[A
Experiment with params: mlpgray_gelu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 2.992 | Train Acc: 7.65%
Val. Loss: 2.968 | Val. Acc: 9.08%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.879 | Train Acc: 10.48%
Val. Loss: 2.782 | Val. Acc: 12.36%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 32s
Train Loss: 2.763 | Train Acc: 12.94%
Val. Loss: 2.698 | Val. Acc: 16.05%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 33s
Train Loss: 2.704 | Train Acc: 15.62%
Val. Loss: 2.659 | Val. Acc: 18.22%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.673 | Train Acc: 16.83%
Val. Loss: 2.637 | Val. Acc: 18.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 32s
Train Loss: 2.641 | Train Acc: 18.31%
Val. Loss: 2.596 | Val. Acc: 20.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 32s
Train Loss: 2.606 | Train Acc: 19.28%
Val. Loss: 2.589 | Val. Acc: 19.75%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 32s
Train Loss: 2.564 | Train Acc: 20.47%
Val. Loss: 2.569 | Val. Acc: 20.09%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.555 | Train Acc: 21.16%
Val. Loss: 2.532 | Val. Acc: 21.11%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.533 | Train Acc: 21.72%
Val. Loss: 2.529 | Val. Acc: 21.87%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 33s
Train Loss: 2.504 | Train Acc: 22.25%
Val. Loss: 2.510 | Val. Acc: 22.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.488 | Train Acc: 22.35%
Val. Loss: 2.495 | Val. Acc: 22.99%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.459 | Train Acc: 23.24%
Val. Loss: 2.484 | Val. Acc: 23.20%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.457 | Train Acc: 23.79%
Val. Loss: 2.482 | Val. Acc: 23.27%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.444 | Train Acc: 23.93%
Val. Loss: 2.469 | Val. Acc: 23.47%
42%|████▏ | 27/64 [3:19:46<4:54:15, 477.16s/it][A[A
Experiment with params: mlpgray_gelu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 29s
Train Loss: 2.995 | Train Acc: 6.97%
Val. Loss: 2.994 | Val. Acc: 12.17%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 29s
Train Loss: 2.974 | Train Acc: 9.14%
Val. Loss: 2.907 | Val. Acc: 9.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 29s
Train Loss: 2.843 | Train Acc: 9.73%
Val. Loss: 2.772 | Val. Acc: 12.25%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 29s
Train Loss: 2.779 | Train Acc: 11.46%
Val. Loss: 2.737 | Val. Acc: 14.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 29s
Train Loss: 2.739 | Train Acc: 13.98%
Val. Loss: 2.707 | Val. Acc: 15.06%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 29s
Train Loss: 2.705 | Train Acc: 15.06%
Val. Loss: 2.670 | Val. Acc: 16.55%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 30s
Train Loss: 2.685 | Train Acc: 16.27%
Val. Loss: 2.640 | Val. Acc: 18.17%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 29s
Train Loss: 2.640 | Train Acc: 17.73%
Val. Loss: 2.615 | Val. Acc: 18.51%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 29s
Train Loss: 2.623 | Train Acc: 18.89%
Val. Loss: 2.595 | Val. Acc: 19.20%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 29s
Train Loss: 2.602 | Train Acc: 19.22%
Val. Loss: 2.579 | Val. Acc: 19.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 29s
Train Loss: 2.596 | Train Acc: 19.45%
Val. Loss: 2.568 | Val. Acc: 20.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 29s
Train Loss: 2.575 | Train Acc: 19.84%
Val. Loss: 2.556 | Val. Acc: 20.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 30s
Train Loss: 2.521 | Train Acc: 20.72%
Val. Loss: 2.544 | Val. Acc: 21.13%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.539 | Train Acc: 22.27%
Val. Loss: 2.528 | Val. Acc: 21.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 30s
Train Loss: 2.510 | Train Acc: 21.52%
Val. Loss: 2.523 | Val. Acc: 22.00%
44%|████▍ | 28/64 [3:27:18<4:41:43, 469.55s/it][A[A
Experiment with params: mlpgray_gelu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 2.913 | Train Acc: 8.82%
Val. Loss: 2.804 | Val. Acc: 12.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.808 | Train Acc: 12.06%
Val. Loss: 2.708 | Val. Acc: 15.12%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.746 | Train Acc: 13.92%
Val. Loss: 2.695 | Val. Acc: 16.69%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 32s
Train Loss: 2.715 | Train Acc: 15.03%
Val. Loss: 2.664 | Val. Acc: 18.02%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.694 | Train Acc: 16.53%
Val. Loss: 2.650 | Val. Acc: 18.33%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 32s
Train Loss: 2.654 | Train Acc: 18.14%
Val. Loss: 2.629 | Val. Acc: 18.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.630 | Train Acc: 18.53%
Val. Loss: 2.613 | Val. Acc: 19.27%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.613 | Train Acc: 18.45%
Val. Loss: 2.598 | Val. Acc: 19.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.598 | Train Acc: 19.44%
Val. Loss: 2.589 | Val. Acc: 19.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.591 | Train Acc: 19.54%
Val. Loss: 2.573 | Val. Acc: 20.58%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.589 | Train Acc: 19.44%
Val. Loss: 2.572 | Val. Acc: 20.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.561 | Train Acc: 20.95%
Val. Loss: 2.561 | Val. Acc: 20.63%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 33s
Train Loss: 2.560 | Train Acc: 20.77%
Val. Loss: 2.552 | Val. Acc: 20.60%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.530 | Train Acc: 21.82%
Val. Loss: 2.539 | Val. Acc: 21.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 33s
Train Loss: 2.509 | Train Acc: 21.68%
Val. Loss: 2.530 | Val. Acc: 21.74%
45%|████▌ | 29/64 [3:35:25<4:37:02, 474.94s/it][A[A
Experiment with params: mlpgray_gelu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.953 | Train Acc: 8.75%
Val. Loss: 2.848 | Val. Acc: 11.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.827 | Train Acc: 11.07%
Val. Loss: 2.749 | Val. Acc: 14.02%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.779 | Train Acc: 13.22%
Val. Loss: 2.719 | Val. Acc: 15.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.743 | Train Acc: 13.52%
Val. Loss: 2.683 | Val. Acc: 16.42%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.718 | Train Acc: 15.04%
Val. Loss: 2.670 | Val. Acc: 16.72%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 30s
Train Loss: 2.695 | Train Acc: 15.37%
Val. Loss: 2.650 | Val. Acc: 17.60%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.669 | Train Acc: 16.02%
Val. Loss: 2.628 | Val. Acc: 17.98%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 30s
Train Loss: 2.652 | Train Acc: 17.17%
Val. Loss: 2.613 | Val. Acc: 18.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 30s
Train Loss: 2.625 | Train Acc: 16.89%
Val. Loss: 2.603 | Val. Acc: 18.64%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 30s
Train Loss: 2.614 | Train Acc: 19.51%
Val. Loss: 2.596 | Val. Acc: 19.06%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 30s
Train Loss: 2.604 | Train Acc: 18.81%
Val. Loss: 2.605 | Val. Acc: 19.22%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 31s
Train Loss: 2.585 | Train Acc: 19.06%
Val. Loss: 2.596 | Val. Acc: 19.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.582 | Train Acc: 19.30%
Val. Loss: 2.573 | Val. Acc: 19.75%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.565 | Train Acc: 19.57%
Val. Loss: 2.561 | Val. Acc: 20.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.544 | Train Acc: 21.02%
Val. Loss: 2.554 | Val. Acc: 21.04%
47%|████▋ | 30/64 [3:43:14<4:28:01, 472.97s/it][A[A
Experiment with params: mlpgray_gelu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 33s
Train Loss: 2.894 | Train Acc: 9.79%
Val. Loss: 2.770 | Val. Acc: 14.55%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.740 | Train Acc: 14.46%
Val. Loss: 2.675 | Val. Acc: 16.39%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 33s
Train Loss: 2.663 | Train Acc: 16.59%
Val. Loss: 2.624 | Val. Acc: 18.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 33s
Train Loss: 2.630 | Train Acc: 17.92%
Val. Loss: 2.605 | Val. Acc: 19.35%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 33s
Train Loss: 2.597 | Train Acc: 19.54%
Val. Loss: 2.582 | Val. Acc: 19.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.569 | Train Acc: 19.62%
Val. Loss: 2.569 | Val. Acc: 19.92%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.537 | Train Acc: 20.83%
Val. Loss: 2.534 | Val. Acc: 21.60%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 33s
Train Loss: 2.512 | Train Acc: 22.27%
Val. Loss: 2.521 | Val. Acc: 21.85%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.493 | Train Acc: 22.21%
Val. Loss: 2.506 | Val. Acc: 22.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.479 | Train Acc: 23.18%
Val. Loss: 2.504 | Val. Acc: 22.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.459 | Train Acc: 23.28%
Val. Loss: 2.496 | Val. Acc: 22.62%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.439 | Train Acc: 23.87%
Val. Loss: 2.498 | Val. Acc: 22.70%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.433 | Train Acc: 24.25%
Val. Loss: 2.491 | Val. Acc: 22.84%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.399 | Train Acc: 25.34%
Val. Loss: 2.463 | Val. Acc: 23.90%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 33s
Train Loss: 2.396 | Train Acc: 24.76%
Val. Loss: 2.469 | Val. Acc: 23.38%
48%|████▊ | 31/64 [3:51:31<4:24:07, 480.23s/it][A[A
Experiment with params: mlpgray_gelu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.948 | Train Acc: 9.39%
Val. Loss: 2.843 | Val. Acc: 11.88%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.791 | Train Acc: 12.60%
Val. Loss: 2.718 | Val. Acc: 15.05%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.712 | Train Acc: 15.23%
Val. Loss: 2.680 | Val. Acc: 16.00%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.657 | Train Acc: 17.23%
Val. Loss: 2.634 | Val. Acc: 18.34%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.627 | Train Acc: 17.75%
Val. Loss: 2.611 | Val. Acc: 19.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.609 | Train Acc: 18.14%
Val. Loss: 2.592 | Val. Acc: 19.86%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 32s
Train Loss: 2.592 | Train Acc: 19.16%
Val. Loss: 2.576 | Val. Acc: 20.08%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 32s
Train Loss: 2.553 | Train Acc: 20.49%
Val. Loss: 2.552 | Val. Acc: 20.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.528 | Train Acc: 21.95%
Val. Loss: 2.538 | Val. Acc: 21.33%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.507 | Train Acc: 22.54%
Val. Loss: 2.532 | Val. Acc: 21.35%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.487 | Train Acc: 22.38%
Val. Loss: 2.522 | Val. Acc: 21.96%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.492 | Train Acc: 21.46%
Val. Loss: 2.506 | Val. Acc: 22.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.467 | Train Acc: 23.93%
Val. Loss: 2.491 | Val. Acc: 22.85%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.432 | Train Acc: 24.86%
Val. Loss: 2.484 | Val. Acc: 22.74%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.434 | Train Acc: 23.55%
Val. Loss: 2.471 | Val. Acc: 23.61%
50%|█████ | 32/64 [3:59:32<4:16:16, 480.51s/it][A[A
Experiment with params: resmlp_relu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 38s
Train Loss: 2.973 | Train Acc: 7.97%
Val. Loss: 2.926 | Val. Acc: 10.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 34s
Train Loss: 2.883 | Train Acc: 10.32%
Val. Loss: 2.840 | Val. Acc: 10.98%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 34s
Train Loss: 2.814 | Train Acc: 10.92%
Val. Loss: 2.799 | Val. Acc: 12.56%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 34s
Train Loss: 2.789 | Train Acc: 11.83%
Val. Loss: 2.777 | Val. Acc: 12.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 34s
Train Loss: 2.767 | Train Acc: 13.33%
Val. Loss: 2.756 | Val. Acc: 14.85%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.740 | Train Acc: 15.21%
Val. Loss: 2.726 | Val. Acc: 16.04%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.712 | Train Acc: 15.43%
Val. Loss: 2.683 | Val. Acc: 18.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 34s
Train Loss: 2.660 | Train Acc: 18.02%
Val. Loss: 2.642 | Val. Acc: 18.71%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 34s
Train Loss: 2.623 | Train Acc: 18.55%
Val. Loss: 2.608 | Val. Acc: 19.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 34s
Train Loss: 2.601 | Train Acc: 19.17%
Val. Loss: 2.594 | Val. Acc: 20.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 34s
Train Loss: 2.584 | Train Acc: 19.54%
Val. Loss: 2.575 | Val. Acc: 20.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 34s
Train Loss: 2.567 | Train Acc: 20.43%
Val. Loss: 2.560 | Val. Acc: 21.42%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 34s
Train Loss: 2.547 | Train Acc: 21.34%
Val. Loss: 2.551 | Val. Acc: 21.55%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 35s
Train Loss: 2.540 | Train Acc: 21.62%
Val. Loss: 2.531 | Val. Acc: 22.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 34s
Train Loss: 2.520 | Train Acc: 22.29%
Val. Loss: 2.518 | Val. Acc: 22.41%
52%|█████▏ | 33/64 [4:08:12<4:14:22, 492.33s/it][A[A
Experiment with params: resmlp_relu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 32s
Train Loss: 3.008 | Train Acc: 5.66%
Val. Loss: 2.982 | Val. Acc: 8.55%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.950 | Train Acc: 9.63%
Val. Loss: 2.927 | Val. Acc: 10.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 32s
Train Loss: 2.904 | Train Acc: 9.90%
Val. Loss: 2.873 | Val. Acc: 10.44%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.843 | Train Acc: 11.52%
Val. Loss: 2.825 | Val. Acc: 11.21%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.811 | Train Acc: 11.07%
Val. Loss: 2.804 | Val. Acc: 12.17%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 32s
Train Loss: 2.803 | Train Acc: 12.05%
Val. Loss: 2.790 | Val. Acc: 13.05%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 32s
Train Loss: 2.780 | Train Acc: 13.09%
Val. Loss: 2.777 | Val. Acc: 12.89%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 32s
Train Loss: 2.765 | Train Acc: 13.40%
Val. Loss: 2.764 | Val. Acc: 14.39%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 32s
Train Loss: 2.758 | Train Acc: 13.96%
Val. Loss: 2.753 | Val. Acc: 14.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.744 | Train Acc: 14.00%
Val. Loss: 2.738 | Val. Acc: 15.36%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.723 | Train Acc: 15.45%
Val. Loss: 2.724 | Val. Acc: 15.98%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.707 | Train Acc: 15.96%
Val. Loss: 2.700 | Val. Acc: 16.51%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.692 | Train Acc: 16.46%
Val. Loss: 2.677 | Val. Acc: 17.02%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.681 | Train Acc: 16.72%
Val. Loss: 2.656 | Val. Acc: 18.54%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 33s
Train Loss: 2.644 | Train Acc: 17.29%
Val. Loss: 2.639 | Val. Acc: 18.88%
53%|█████▎ | 34/64 [4:16:18<4:05:14, 490.47s/it][A[A
Experiment with params: resmlp_relu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 34s
Train Loss: 2.983 | Train Acc: 5.80%
Val. Loss: 2.937 | Val. Acc: 8.08%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.896 | Train Acc: 9.77%
Val. Loss: 2.853 | Val. Acc: 11.18%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 34s
Train Loss: 2.812 | Train Acc: 11.67%
Val. Loss: 2.803 | Val. Acc: 11.78%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 34s
Train Loss: 2.788 | Train Acc: 12.03%
Val. Loss: 2.779 | Val. Acc: 12.85%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 34s
Train Loss: 2.769 | Train Acc: 13.37%
Val. Loss: 2.757 | Val. Acc: 13.97%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 34s
Train Loss: 2.740 | Train Acc: 14.66%
Val. Loss: 2.732 | Val. Acc: 15.37%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 34s
Train Loss: 2.709 | Train Acc: 16.22%
Val. Loss: 2.696 | Val. Acc: 16.14%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 33s
Train Loss: 2.670 | Train Acc: 16.34%
Val. Loss: 2.658 | Val. Acc: 18.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.639 | Train Acc: 18.55%
Val. Loss: 2.635 | Val. Acc: 18.70%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.618 | Train Acc: 18.85%
Val. Loss: 2.598 | Val. Acc: 19.46%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.587 | Train Acc: 19.86%
Val. Loss: 2.573 | Val. Acc: 21.03%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.577 | Train Acc: 20.77%
Val. Loss: 2.555 | Val. Acc: 21.68%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 33s
Train Loss: 2.545 | Train Acc: 21.70%
Val. Loss: 2.546 | Val. Acc: 21.41%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 33s
Train Loss: 2.539 | Train Acc: 21.52%
Val. Loss: 2.538 | Val. Acc: 22.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 33s
Train Loss: 2.516 | Train Acc: 22.55%
Val. Loss: 2.521 | Val. Acc: 22.35%
55%|█████▍ | 35/64 [4:24:46<3:59:36, 495.73s/it][A[A
Experiment with params: resmlp_relu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 3.008 | Train Acc: 5.29%
Val. Loss: 2.984 | Val. Acc: 6.15%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.959 | Train Acc: 8.32%
Val. Loss: 2.939 | Val. Acc: 9.29%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.914 | Train Acc: 9.75%
Val. Loss: 2.895 | Val. Acc: 10.43%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.870 | Train Acc: 10.62%
Val. Loss: 2.845 | Val. Acc: 11.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.823 | Train Acc: 12.15%
Val. Loss: 2.810 | Val. Acc: 11.75%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.801 | Train Acc: 12.15%
Val. Loss: 2.791 | Val. Acc: 13.01%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.780 | Train Acc: 13.57%
Val. Loss: 2.780 | Val. Acc: 13.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.775 | Train Acc: 12.89%
Val. Loss: 2.767 | Val. Acc: 13.99%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 31s
Train Loss: 2.758 | Train Acc: 13.85%
Val. Loss: 2.752 | Val. Acc: 14.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 31s
Train Loss: 2.750 | Train Acc: 14.18%
Val. Loss: 2.733 | Val. Acc: 15.82%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 31s
Train Loss: 2.719 | Train Acc: 15.61%
Val. Loss: 2.717 | Val. Acc: 15.86%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 31s
Train Loss: 2.687 | Train Acc: 16.31%
Val. Loss: 2.685 | Val. Acc: 18.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 31s
Train Loss: 2.657 | Train Acc: 16.70%
Val. Loss: 2.656 | Val. Acc: 18.40%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.651 | Train Acc: 17.85%
Val. Loss: 2.635 | Val. Acc: 19.12%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.626 | Train Acc: 17.97%
Val. Loss: 2.619 | Val. Acc: 19.08%
56%|█████▋ | 36/64 [4:32:40<3:48:17, 489.19s/it][A[A
Experiment with params: resmlp_relu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 33s
Train Loss: 2.858 | Train Acc: 9.77%
Val. Loss: 2.804 | Val. Acc: 10.41%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.778 | Train Acc: 13.07%
Val. Loss: 2.743 | Val. Acc: 15.20%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 33s
Train Loss: 2.709 | Train Acc: 15.61%
Val. Loss: 2.673 | Val. Acc: 16.55%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 33s
Train Loss: 2.654 | Train Acc: 17.05%
Val. Loss: 2.631 | Val. Acc: 18.58%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 33s
Train Loss: 2.616 | Train Acc: 18.87%
Val. Loss: 2.595 | Val. Acc: 20.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.580 | Train Acc: 20.39%
Val. Loss: 2.565 | Val. Acc: 20.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.553 | Train Acc: 20.89%
Val. Loss: 2.543 | Val. Acc: 21.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 33s
Train Loss: 2.540 | Train Acc: 21.91%
Val. Loss: 2.528 | Val. Acc: 21.96%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.513 | Train Acc: 21.72%
Val. Loss: 2.508 | Val. Acc: 22.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.491 | Train Acc: 23.40%
Val. Loss: 2.494 | Val. Acc: 22.71%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.482 | Train Acc: 23.34%
Val. Loss: 2.495 | Val. Acc: 22.99%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.462 | Train Acc: 24.01%
Val. Loss: 2.462 | Val. Acc: 24.07%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 34s
Train Loss: 2.439 | Train Acc: 25.10%
Val. Loss: 2.450 | Val. Acc: 24.40%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 33s
Train Loss: 2.434 | Train Acc: 24.64%
Val. Loss: 2.444 | Val. Acc: 24.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.417 | Train Acc: 25.44%
Val. Loss: 2.431 | Val. Acc: 24.94%
58%|█████▊ | 37/64 [4:41:01<3:41:45, 492.80s/it][A[A
Experiment with params: resmlp_relu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.896 | Train Acc: 9.39%
Val. Loss: 2.813 | Val. Acc: 11.16%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.800 | Train Acc: 12.71%
Val. Loss: 2.779 | Val. Acc: 13.83%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.771 | Train Acc: 14.00%
Val. Loss: 2.744 | Val. Acc: 14.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.725 | Train Acc: 14.94%
Val. Loss: 2.704 | Val. Acc: 16.15%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.686 | Train Acc: 15.96%
Val. Loss: 2.672 | Val. Acc: 16.70%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.651 | Train Acc: 17.36%
Val. Loss: 2.629 | Val. Acc: 18.54%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.616 | Train Acc: 18.81%
Val. Loss: 2.597 | Val. Acc: 20.45%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 31s
Train Loss: 2.593 | Train Acc: 19.10%
Val. Loss: 2.577 | Val. Acc: 20.59%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 31s
Train Loss: 2.570 | Train Acc: 20.37%
Val. Loss: 2.556 | Val. Acc: 21.54%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 31s
Train Loss: 2.555 | Train Acc: 21.19%
Val. Loss: 2.552 | Val. Acc: 21.31%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 31s
Train Loss: 2.532 | Train Acc: 21.86%
Val. Loss: 2.524 | Val. Acc: 22.51%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 31s
Train Loss: 2.511 | Train Acc: 22.03%
Val. Loss: 2.521 | Val. Acc: 22.44%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.499 | Train Acc: 22.85%
Val. Loss: 2.502 | Val. Acc: 23.12%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 31s
Train Loss: 2.497 | Train Acc: 22.50%
Val. Loss: 2.500 | Val. Acc: 22.81%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 31s
Train Loss: 2.487 | Train Acc: 23.59%
Val. Loss: 2.489 | Val. Acc: 23.32%
59%|█████▉ | 38/64 [4:48:56<3:31:12, 487.41s/it][A[A
Experiment with params: resmlp_relu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 33s
Train Loss: 2.867 | Train Acc: 9.22%
Val. Loss: 2.797 | Val. Acc: 11.79%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 34s
Train Loss: 2.774 | Train Acc: 13.29%
Val. Loss: 2.734 | Val. Acc: 15.13%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 34s
Train Loss: 2.698 | Train Acc: 16.02%
Val. Loss: 2.663 | Val. Acc: 17.23%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 34s
Train Loss: 2.635 | Train Acc: 18.65%
Val. Loss: 2.615 | Val. Acc: 19.03%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 33s
Train Loss: 2.594 | Train Acc: 19.11%
Val. Loss: 2.587 | Val. Acc: 20.19%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.564 | Train Acc: 20.69%
Val. Loss: 2.543 | Val. Acc: 21.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.548 | Train Acc: 21.88%
Val. Loss: 2.527 | Val. Acc: 22.02%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 33s
Train Loss: 2.519 | Train Acc: 22.21%
Val. Loss: 2.534 | Val. Acc: 21.71%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.509 | Train Acc: 22.33%
Val. Loss: 2.519 | Val. Acc: 22.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 32s
Train Loss: 2.482 | Train Acc: 23.02%
Val. Loss: 2.484 | Val. Acc: 23.42%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 33s
Train Loss: 2.459 | Train Acc: 24.23%
Val. Loss: 2.470 | Val. Acc: 23.83%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.458 | Train Acc: 24.05%
Val. Loss: 2.458 | Val. Acc: 24.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 34s
Train Loss: 2.456 | Train Acc: 24.47%
Val. Loss: 2.453 | Val. Acc: 24.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 34s
Train Loss: 2.413 | Train Acc: 25.04%
Val. Loss: 2.439 | Val. Acc: 25.13%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 34s
Train Loss: 2.421 | Train Acc: 25.67%
Val. Loss: 2.424 | Val. Acc: 25.65%
61%|██████ | 39/64 [4:57:24<3:25:39, 493.58s/it][A[A
Experiment with params: resmlp_relu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 31s
Train Loss: 2.892 | Train Acc: 9.88%
Val. Loss: 2.819 | Val. Acc: 11.74%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 31s
Train Loss: 2.797 | Train Acc: 12.30%
Val. Loss: 2.786 | Val. Acc: 12.27%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 31s
Train Loss: 2.760 | Train Acc: 13.52%
Val. Loss: 2.749 | Val. Acc: 13.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 31s
Train Loss: 2.728 | Train Acc: 14.98%
Val. Loss: 2.705 | Val. Acc: 16.80%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 31s
Train Loss: 2.676 | Train Acc: 17.36%
Val. Loss: 2.661 | Val. Acc: 18.75%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 31s
Train Loss: 2.640 | Train Acc: 17.54%
Val. Loss: 2.620 | Val. Acc: 19.33%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 31s
Train Loss: 2.598 | Train Acc: 19.59%
Val. Loss: 2.594 | Val. Acc: 19.88%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 32s
Train Loss: 2.574 | Train Acc: 20.49%
Val. Loss: 2.583 | Val. Acc: 20.80%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 34s
Train Loss: 2.556 | Train Acc: 21.25%
Val. Loss: 2.546 | Val. Acc: 21.77%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 35s
Train Loss: 2.544 | Train Acc: 21.95%
Val. Loss: 2.530 | Val. Acc: 22.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.512 | Train Acc: 22.54%
Val. Loss: 2.518 | Val. Acc: 22.72%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.507 | Train Acc: 22.42%
Val. Loss: 2.505 | Val. Acc: 22.77%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.496 | Train Acc: 24.10%
Val. Loss: 2.495 | Val. Acc: 23.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 32s
Train Loss: 2.473 | Train Acc: 22.85%
Val. Loss: 2.487 | Val. Acc: 23.40%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.460 | Train Acc: 24.06%
Val. Loss: 2.486 | Val. Acc: 23.34%
62%|██████▎ | 40/64 [5:05:32<3:16:43, 491.80s/it][A[A
Experiment with params: resmlp_gelu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 36s
Train Loss: 2.991 | Train Acc: 5.58%
Val. Loss: 2.945 | Val. Acc: 8.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 36s
Train Loss: 2.897 | Train Acc: 10.21%
Val. Loss: 2.846 | Val. Acc: 10.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 36s
Train Loss: 2.812 | Train Acc: 11.71%
Val. Loss: 2.803 | Val. Acc: 11.87%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 36s
Train Loss: 2.790 | Train Acc: 12.66%
Val. Loss: 2.774 | Val. Acc: 14.46%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 35s
Train Loss: 2.756 | Train Acc: 14.34%
Val. Loss: 2.748 | Val. Acc: 15.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 37s
Train Loss: 2.732 | Train Acc: 14.89%
Val. Loss: 2.715 | Val. Acc: 16.26%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.688 | Train Acc: 16.81%
Val. Loss: 2.674 | Val. Acc: 17.83%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 38s
Train Loss: 2.660 | Train Acc: 17.21%
Val. Loss: 2.656 | Val. Acc: 18.00%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 36s
Train Loss: 2.627 | Train Acc: 18.39%
Val. Loss: 2.614 | Val. Acc: 19.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 35s
Train Loss: 2.599 | Train Acc: 19.42%
Val. Loss: 2.586 | Val. Acc: 20.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 36s
Train Loss: 2.579 | Train Acc: 20.29%
Val. Loss: 2.577 | Val. Acc: 20.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.556 | Train Acc: 20.79%
Val. Loss: 2.550 | Val. Acc: 21.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 38s
Train Loss: 2.541 | Train Acc: 21.24%
Val. Loss: 2.534 | Val. Acc: 21.83%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 36s
Train Loss: 2.523 | Train Acc: 22.11%
Val. Loss: 2.516 | Val. Acc: 22.42%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 36s
Train Loss: 2.507 | Train Acc: 22.19%
Val. Loss: 2.505 | Val. Acc: 22.61%
64%|██████▍ | 41/64 [5:14:44<3:15:28, 509.94s/it][A[A
Experiment with params: resmlp_gelu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 33s
Train Loss: 2.999 | Train Acc: 5.61%
Val. Loss: 2.975 | Val. Acc: 6.75%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.945 | Train Acc: 8.24%
Val. Loss: 2.922 | Val. Acc: 10.13%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 33s
Train Loss: 2.893 | Train Acc: 10.90%
Val. Loss: 2.863 | Val. Acc: 10.76%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 33s
Train Loss: 2.833 | Train Acc: 10.08%
Val. Loss: 2.819 | Val. Acc: 11.09%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 33s
Train Loss: 2.810 | Train Acc: 11.04%
Val. Loss: 2.802 | Val. Acc: 11.61%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.789 | Train Acc: 12.07%
Val. Loss: 2.787 | Val. Acc: 12.76%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.780 | Train Acc: 12.93%
Val. Loss: 2.775 | Val. Acc: 12.78%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 34s
Train Loss: 2.762 | Train Acc: 13.91%
Val. Loss: 2.765 | Val. Acc: 13.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.753 | Train Acc: 14.02%
Val. Loss: 2.746 | Val. Acc: 14.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.736 | Train Acc: 14.94%
Val. Loss: 2.731 | Val. Acc: 15.25%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 33s
Train Loss: 2.708 | Train Acc: 15.88%
Val. Loss: 2.715 | Val. Acc: 16.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.704 | Train Acc: 16.58%
Val. Loss: 2.691 | Val. Acc: 16.98%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 32s
Train Loss: 2.684 | Train Acc: 16.02%
Val. Loss: 2.667 | Val. Acc: 17.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 33s
Train Loss: 2.662 | Train Acc: 17.50%
Val. Loss: 2.652 | Val. Acc: 17.88%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 32s
Train Loss: 2.631 | Train Acc: 18.96%
Val. Loss: 2.633 | Val. Acc: 19.19%
66%|██████▌ | 42/64 [5:23:06<3:06:07, 507.64s/it][A[A
Experiment with params: resmlp_gelu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 35s
Train Loss: 2.967 | Train Acc: 8.76%
Val. Loss: 2.930 | Val. Acc: 10.64%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 35s
Train Loss: 2.890 | Train Acc: 10.52%
Val. Loss: 2.840 | Val. Acc: 11.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 35s
Train Loss: 2.807 | Train Acc: 11.47%
Val. Loss: 2.798 | Val. Acc: 11.87%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 35s
Train Loss: 2.790 | Train Acc: 12.18%
Val. Loss: 2.772 | Val. Acc: 14.04%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 35s
Train Loss: 2.757 | Train Acc: 13.94%
Val. Loss: 2.750 | Val. Acc: 15.21%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 36s
Train Loss: 2.734 | Train Acc: 15.31%
Val. Loss: 2.717 | Val. Acc: 15.71%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 35s
Train Loss: 2.694 | Train Acc: 16.04%
Val. Loss: 2.671 | Val. Acc: 17.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 35s
Train Loss: 2.653 | Train Acc: 17.48%
Val. Loss: 2.629 | Val. Acc: 19.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 35s
Train Loss: 2.617 | Train Acc: 19.58%
Val. Loss: 2.605 | Val. Acc: 19.24%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 36s
Train Loss: 2.597 | Train Acc: 18.75%
Val. Loss: 2.598 | Val. Acc: 19.23%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 36s
Train Loss: 2.582 | Train Acc: 20.25%
Val. Loss: 2.574 | Val. Acc: 20.62%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.562 | Train Acc: 20.19%
Val. Loss: 2.551 | Val. Acc: 21.18%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 36s
Train Loss: 2.546 | Train Acc: 21.00%
Val. Loss: 2.545 | Val. Acc: 21.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 35s
Train Loss: 2.527 | Train Acc: 21.89%
Val. Loss: 2.528 | Val. Acc: 22.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 36s
Train Loss: 2.516 | Train Acc: 22.33%
Val. Loss: 2.511 | Val. Acc: 23.12%
67%|██████▋ | 43/64 [5:32:03<3:00:42, 516.29s/it][A[A
Experiment with params: resmlp_gelu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 33s
Train Loss: 3.008 | Train Acc: 4.90%
Val. Loss: 2.984 | Val. Acc: 6.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 32s
Train Loss: 2.951 | Train Acc: 8.75%
Val. Loss: 2.930 | Val. Acc: 10.10%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 33s
Train Loss: 2.903 | Train Acc: 10.00%
Val. Loss: 2.878 | Val. Acc: 10.26%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 32s
Train Loss: 2.837 | Train Acc: 11.74%
Val. Loss: 2.822 | Val. Acc: 11.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 33s
Train Loss: 2.809 | Train Acc: 11.82%
Val. Loss: 2.801 | Val. Acc: 12.45%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 33s
Train Loss: 2.791 | Train Acc: 12.66%
Val. Loss: 2.787 | Val. Acc: 13.20%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 33s
Train Loss: 2.775 | Train Acc: 13.83%
Val. Loss: 2.774 | Val. Acc: 14.13%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 33s
Train Loss: 2.766 | Train Acc: 14.04%
Val. Loss: 2.760 | Val. Acc: 14.43%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.734 | Train Acc: 15.27%
Val. Loss: 2.745 | Val. Acc: 15.21%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.738 | Train Acc: 15.00%
Val. Loss: 2.732 | Val. Acc: 15.39%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 32s
Train Loss: 2.727 | Train Acc: 14.96%
Val. Loss: 2.711 | Val. Acc: 16.61%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 32s
Train Loss: 2.696 | Train Acc: 16.05%
Val. Loss: 2.689 | Val. Acc: 17.37%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 33s
Train Loss: 2.675 | Train Acc: 16.60%
Val. Loss: 2.662 | Val. Acc: 18.60%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 33s
Train Loss: 2.645 | Train Acc: 17.93%
Val. Loss: 2.640 | Val. Acc: 18.51%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 33s
Train Loss: 2.630 | Train Acc: 18.91%
Val. Loss: 2.621 | Val. Acc: 19.38%
69%|██████▉ | 44/64 [5:40:23<2:50:26, 511.34s/it][A[A
Experiment with params: resmlp_gelu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 36s
Train Loss: 2.870 | Train Acc: 9.67%
Val. Loss: 2.801 | Val. Acc: 12.12%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 35s
Train Loss: 2.771 | Train Acc: 13.25%
Val. Loss: 2.739 | Val. Acc: 14.51%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 35s
Train Loss: 2.692 | Train Acc: 16.12%
Val. Loss: 2.658 | Val. Acc: 17.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 35s
Train Loss: 2.632 | Train Acc: 17.62%
Val. Loss: 2.594 | Val. Acc: 20.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 35s
Train Loss: 2.577 | Train Acc: 20.49%
Val. Loss: 2.579 | Val. Acc: 19.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 35s
Train Loss: 2.563 | Train Acc: 20.83%
Val. Loss: 2.545 | Val. Acc: 21.14%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 35s
Train Loss: 2.527 | Train Acc: 22.05%
Val. Loss: 2.524 | Val. Acc: 22.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 35s
Train Loss: 2.525 | Train Acc: 22.25%
Val. Loss: 2.513 | Val. Acc: 21.96%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 35s
Train Loss: 2.498 | Train Acc: 21.86%
Val. Loss: 2.501 | Val. Acc: 22.31%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 35s
Train Loss: 2.479 | Train Acc: 23.24%
Val. Loss: 2.490 | Val. Acc: 23.45%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 35s
Train Loss: 2.470 | Train Acc: 23.40%
Val. Loss: 2.485 | Val. Acc: 23.15%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.458 | Train Acc: 23.66%
Val. Loss: 2.478 | Val. Acc: 23.48%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 36s
Train Loss: 2.440 | Train Acc: 24.72%
Val. Loss: 2.472 | Val. Acc: 23.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 35s
Train Loss: 2.435 | Train Acc: 24.78%
Val. Loss: 2.444 | Val. Acc: 24.98%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 35s
Train Loss: 2.427 | Train Acc: 25.45%
Val. Loss: 2.448 | Val. Acc: 24.55%
70%|███████ | 45/64 [5:49:17<2:44:07, 518.27s/it][A[A
Experiment with params: resmlp_gelu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 33s
Train Loss: 2.893 | Train Acc: 9.63%
Val. Loss: 2.816 | Val. Acc: 11.20%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.803 | Train Acc: 11.33%
Val. Loss: 2.782 | Val. Acc: 13.70%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 33s
Train Loss: 2.766 | Train Acc: 13.52%
Val. Loss: 2.736 | Val. Acc: 15.31%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 32s
Train Loss: 2.711 | Train Acc: 14.77%
Val. Loss: 2.684 | Val. Acc: 16.85%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 32s
Train Loss: 2.669 | Train Acc: 17.29%
Val. Loss: 2.646 | Val. Acc: 17.99%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 32s
Train Loss: 2.626 | Train Acc: 18.67%
Val. Loss: 2.611 | Val. Acc: 19.52%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 32s
Train Loss: 2.589 | Train Acc: 20.14%
Val. Loss: 2.579 | Val. Acc: 20.72%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 33s
Train Loss: 2.580 | Train Acc: 20.23%
Val. Loss: 2.552 | Val. Acc: 21.12%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 33s
Train Loss: 2.546 | Train Acc: 20.78%
Val. Loss: 2.546 | Val. Acc: 21.25%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 33s
Train Loss: 2.521 | Train Acc: 21.80%
Val. Loss: 2.529 | Val. Acc: 21.86%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 34s
Train Loss: 2.511 | Train Acc: 21.25%
Val. Loss: 2.510 | Val. Acc: 22.36%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.501 | Train Acc: 21.64%
Val. Loss: 2.516 | Val. Acc: 22.03%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 36s
Train Loss: 2.489 | Train Acc: 22.87%
Val. Loss: 2.502 | Val. Acc: 22.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 44s
Train Loss: 2.469 | Train Acc: 23.83%
Val. Loss: 2.483 | Val. Acc: 23.63%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 40s
Train Loss: 2.458 | Train Acc: 24.36%
Val. Loss: 2.474 | Val. Acc: 23.93%
72%|███████▏ | 46/64 [5:58:01<2:35:59, 519.98s/it][A[A
Experiment with params: resmlp_gelu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 41s
Train Loss: 2.869 | Train Acc: 9.99%
Val. Loss: 2.812 | Val. Acc: 11.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 41s
Train Loss: 2.769 | Train Acc: 12.66%
Val. Loss: 2.735 | Val. Acc: 15.64%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 41s
Train Loss: 2.695 | Train Acc: 16.00%
Val. Loss: 2.671 | Val. Acc: 17.37%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 39s
Train Loss: 2.634 | Train Acc: 17.92%
Val. Loss: 2.603 | Val. Acc: 19.62%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 37s
Train Loss: 2.582 | Train Acc: 19.22%
Val. Loss: 2.562 | Val. Acc: 21.58%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 38s
Train Loss: 2.557 | Train Acc: 21.58%
Val. Loss: 2.541 | Val. Acc: 21.92%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.525 | Train Acc: 22.15%
Val. Loss: 2.524 | Val. Acc: 22.69%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 37s
Train Loss: 2.505 | Train Acc: 22.23%
Val. Loss: 2.497 | Val. Acc: 22.89%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.473 | Train Acc: 23.40%
Val. Loss: 2.487 | Val. Acc: 23.30%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 37s
Train Loss: 2.467 | Train Acc: 24.07%
Val. Loss: 2.468 | Val. Acc: 23.61%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 37s
Train Loss: 2.450 | Train Acc: 24.66%
Val. Loss: 2.461 | Val. Acc: 23.92%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 34s
Train Loss: 2.420 | Train Acc: 25.36%
Val. Loss: 2.536 | Val. Acc: 22.35%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 37s
Train Loss: 2.419 | Train Acc: 25.67%
Val. Loss: 2.443 | Val. Acc: 24.15%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 36s
Train Loss: 2.409 | Train Acc: 24.90%
Val. Loss: 2.454 | Val. Acc: 24.51%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 35s
Train Loss: 2.387 | Train Acc: 26.98%
Val. Loss: 2.418 | Val. Acc: 25.56%
73%|███████▎ | 47/64 [6:07:35<2:31:55, 536.20s/it][A[A
Experiment with params: resmlp_gelu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 34s
Train Loss: 2.888 | Train Acc: 9.26%
Val. Loss: 2.814 | Val. Acc: 11.97%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 34s
Train Loss: 2.800 | Train Acc: 11.93%
Val. Loss: 2.780 | Val. Acc: 13.68%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 34s
Train Loss: 2.755 | Train Acc: 14.16%
Val. Loss: 2.740 | Val. Acc: 14.39%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 34s
Train Loss: 2.723 | Train Acc: 15.43%
Val. Loss: 2.683 | Val. Acc: 17.32%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 34s
Train Loss: 2.666 | Train Acc: 17.52%
Val. Loss: 2.634 | Val. Acc: 19.32%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 35s
Train Loss: 2.612 | Train Acc: 19.77%
Val. Loss: 2.603 | Val. Acc: 19.87%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 34s
Train Loss: 2.601 | Train Acc: 19.79%
Val. Loss: 2.579 | Val. Acc: 20.21%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 34s
Train Loss: 2.567 | Train Acc: 19.10%
Val. Loss: 2.551 | Val. Acc: 21.64%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 34s
Train Loss: 2.540 | Train Acc: 20.70%
Val. Loss: 2.533 | Val. Acc: 22.18%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 34s
Train Loss: 2.515 | Train Acc: 22.50%
Val. Loss: 2.520 | Val. Acc: 22.19%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 35s
Train Loss: 2.508 | Train Acc: 21.97%
Val. Loss: 2.505 | Val. Acc: 22.88%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 34s
Train Loss: 2.497 | Train Acc: 22.60%
Val. Loss: 2.494 | Val. Acc: 23.19%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 35s
Train Loss: 2.485 | Train Acc: 22.66%
Val. Loss: 2.480 | Val. Acc: 23.96%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 35s
Train Loss: 2.464 | Train Acc: 23.40%
Val. Loss: 2.484 | Val. Acc: 23.11%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 36s
Train Loss: 2.451 | Train Acc: 23.50%
Val. Loss: 2.485 | Val. Acc: 23.20%
75%|███████▌ | 48/64 [6:16:19<2:22:03, 532.70s/it][A[A
Experiment with params: mlpmixer_relu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 42s
Train Loss: 2.936 | Train Acc: 9.06%
Val. Loss: 2.827 | Val. Acc: 12.52%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 42s
Train Loss: 2.771 | Train Acc: 14.10%
Val. Loss: 2.721 | Val. Acc: 16.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 43s
Train Loss: 2.685 | Train Acc: 17.78%
Val. Loss: 2.654 | Val. Acc: 18.17%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 43s
Train Loss: 2.617 | Train Acc: 20.02%
Val. Loss: 2.592 | Val. Acc: 20.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 42s
Train Loss: 2.552 | Train Acc: 21.84%
Val. Loss: 2.540 | Val. Acc: 21.92%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 40s
Train Loss: 2.490 | Train Acc: 23.28%
Val. Loss: 2.502 | Val. Acc: 22.70%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 40s
Train Loss: 2.431 | Train Acc: 24.29%
Val. Loss: 2.461 | Val. Acc: 24.91%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 40s
Train Loss: 2.396 | Train Acc: 25.99%
Val. Loss: 2.413 | Val. Acc: 25.83%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.359 | Train Acc: 27.16%
Val. Loss: 2.385 | Val. Acc: 26.86%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 38s
Train Loss: 2.322 | Train Acc: 28.82%
Val. Loss: 2.354 | Val. Acc: 27.76%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 39s
Train Loss: 2.277 | Train Acc: 29.75%
Val. Loss: 2.352 | Val. Acc: 28.15%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 39s
Train Loss: 2.251 | Train Acc: 30.76%
Val. Loss: 2.333 | Val. Acc: 27.79%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 38s
Train Loss: 2.229 | Train Acc: 31.72%
Val. Loss: 2.297 | Val. Acc: 29.01%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 40s
Train Loss: 2.188 | Train Acc: 32.02%
Val. Loss: 2.264 | Val. Acc: 30.72%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 39s
Train Loss: 2.159 | Train Acc: 33.19%
Val. Loss: 2.262 | Val. Acc: 30.46%
77%|███████▋ | 49/64 [6:26:31<2:19:03, 556.24s/it][A[A
Experiment with params: mlpmixer_relu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 38s
Train Loss: 2.989 | Train Acc: 5.25%
Val. Loss: 2.917 | Val. Acc: 8.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 38s
Train Loss: 2.855 | Train Acc: 12.66%
Val. Loss: 2.802 | Val. Acc: 14.39%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 37s
Train Loss: 2.769 | Train Acc: 14.28%
Val. Loss: 2.743 | Val. Acc: 16.14%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 38s
Train Loss: 2.714 | Train Acc: 16.17%
Val. Loss: 2.696 | Val. Acc: 16.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 38s
Train Loss: 2.677 | Train Acc: 17.73%
Val. Loss: 2.650 | Val. Acc: 18.97%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 37s
Train Loss: 2.644 | Train Acc: 19.18%
Val. Loss: 2.615 | Val. Acc: 20.16%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 39s
Train Loss: 2.589 | Train Acc: 21.37%
Val. Loss: 2.572 | Val. Acc: 21.45%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 38s
Train Loss: 2.544 | Train Acc: 22.05%
Val. Loss: 2.538 | Val. Acc: 22.71%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 35s
Train Loss: 2.507 | Train Acc: 23.79%
Val. Loss: 2.513 | Val. Acc: 23.14%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 35s
Train Loss: 2.493 | Train Acc: 23.14%
Val. Loss: 2.491 | Val. Acc: 23.79%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 35s
Train Loss: 2.446 | Train Acc: 25.29%
Val. Loss: 2.456 | Val. Acc: 24.76%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.426 | Train Acc: 25.62%
Val. Loss: 2.441 | Val. Acc: 25.20%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 35s
Train Loss: 2.384 | Train Acc: 26.97%
Val. Loss: 2.428 | Val. Acc: 25.72%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 36s
Train Loss: 2.387 | Train Acc: 26.74%
Val. Loss: 2.394 | Val. Acc: 26.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 35s
Train Loss: 2.354 | Train Acc: 28.09%
Val. Loss: 2.395 | Val. Acc: 27.33%
78%|███████▊ | 50/64 [6:35:46<2:09:45, 556.09s/it][A[A
Experiment with params: mlpmixer_relu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 37s
Train Loss: 2.977 | Train Acc: 6.59%
Val. Loss: 2.847 | Val. Acc: 11.83%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 38s
Train Loss: 2.780 | Train Acc: 14.36%
Val. Loss: 2.725 | Val. Acc: 16.61%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 38s
Train Loss: 2.678 | Train Acc: 18.33%
Val. Loss: 2.636 | Val. Acc: 19.46%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 38s
Train Loss: 2.600 | Train Acc: 20.15%
Val. Loss: 2.564 | Val. Acc: 21.27%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 38s
Train Loss: 2.523 | Train Acc: 23.02%
Val. Loss: 2.524 | Val. Acc: 22.82%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 37s
Train Loss: 2.464 | Train Acc: 23.79%
Val. Loss: 2.466 | Val. Acc: 23.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 37s
Train Loss: 2.418 | Train Acc: 25.38%
Val. Loss: 2.436 | Val. Acc: 24.81%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 38s
Train Loss: 2.377 | Train Acc: 27.16%
Val. Loss: 2.421 | Val. Acc: 25.92%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.341 | Train Acc: 27.87%
Val. Loss: 2.366 | Val. Acc: 27.15%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 37s
Train Loss: 2.301 | Train Acc: 28.98%
Val. Loss: 2.346 | Val. Acc: 27.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 38s
Train Loss: 2.268 | Train Acc: 30.50%
Val. Loss: 2.337 | Val. Acc: 28.14%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 37s
Train Loss: 2.233 | Train Acc: 31.37%
Val. Loss: 2.302 | Val. Acc: 28.88%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 37s
Train Loss: 2.207 | Train Acc: 31.61%
Val. Loss: 2.275 | Val. Acc: 29.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 37s
Train Loss: 2.181 | Train Acc: 33.15%
Val. Loss: 2.282 | Val. Acc: 29.59%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 37s
Train Loss: 2.148 | Train Acc: 33.72%
Val. Loss: 2.246 | Val. Acc: 30.70%
80%|███████▉ | 51/64 [6:45:17<2:01:25, 560.39s/it][A[A
Experiment with params: mlpmixer_relu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 35s
Train Loss: 2.990 | Train Acc: 7.13%
Val. Loss: 2.920 | Val. Acc: 10.08%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 35s
Train Loss: 2.866 | Train Acc: 11.97%
Val. Loss: 2.807 | Val. Acc: 13.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 36s
Train Loss: 2.780 | Train Acc: 13.75%
Val. Loss: 2.737 | Val. Acc: 16.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 35s
Train Loss: 2.707 | Train Acc: 17.05%
Val. Loss: 2.680 | Val. Acc: 18.36%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 35s
Train Loss: 2.655 | Train Acc: 18.79%
Val. Loss: 2.628 | Val. Acc: 19.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 36s
Train Loss: 2.602 | Train Acc: 19.96%
Val. Loss: 2.578 | Val. Acc: 21.42%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 35s
Train Loss: 2.534 | Train Acc: 21.99%
Val. Loss: 2.535 | Val. Acc: 22.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 35s
Train Loss: 2.485 | Train Acc: 22.34%
Val. Loss: 2.504 | Val. Acc: 23.26%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 35s
Train Loss: 2.471 | Train Acc: 22.87%
Val. Loss: 2.469 | Val. Acc: 24.76%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 36s
Train Loss: 2.421 | Train Acc: 25.49%
Val. Loss: 2.447 | Val. Acc: 25.28%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 35s
Train Loss: 2.409 | Train Acc: 25.76%
Val. Loss: 2.430 | Val. Acc: 25.62%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 34s
Train Loss: 2.383 | Train Acc: 25.96%
Val. Loss: 2.410 | Val. Acc: 25.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 34s
Train Loss: 2.353 | Train Acc: 27.95%
Val. Loss: 2.394 | Val. Acc: 26.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 34s
Train Loss: 2.343 | Train Acc: 27.68%
Val. Loss: 2.373 | Val. Acc: 27.26%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 34s
Train Loss: 2.321 | Train Acc: 28.30%
Val. Loss: 2.375 | Val. Acc: 27.50%
81%|████████▏ | 52/64 [6:54:10<1:50:28, 552.35s/it][A[A
Experiment with params: mlpmixer_relu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 37s
Train Loss: 2.780 | Train Acc: 13.39%
Val. Loss: 2.674 | Val. Acc: 17.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 37s
Train Loss: 2.602 | Train Acc: 19.22%
Val. Loss: 2.542 | Val. Acc: 21.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 37s
Train Loss: 2.492 | Train Acc: 23.16%
Val. Loss: 2.481 | Val. Acc: 23.76%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 36s
Train Loss: 2.450 | Train Acc: 24.55%
Val. Loss: 2.440 | Val. Acc: 24.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 37s
Train Loss: 2.378 | Train Acc: 26.76%
Val. Loss: 2.408 | Val. Acc: 25.67%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 36s
Train Loss: 2.323 | Train Acc: 28.24%
Val. Loss: 2.379 | Val. Acc: 26.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 36s
Train Loss: 2.287 | Train Acc: 28.94%
Val. Loss: 2.321 | Val. Acc: 28.61%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 36s
Train Loss: 2.241 | Train Acc: 30.60%
Val. Loss: 2.288 | Val. Acc: 29.56%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 37s
Train Loss: 2.205 | Train Acc: 32.02%
Val. Loss: 2.262 | Val. Acc: 29.66%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 37s
Train Loss: 2.153 | Train Acc: 33.60%
Val. Loss: 2.243 | Val. Acc: 30.85%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 36s
Train Loss: 2.127 | Train Acc: 34.30%
Val. Loss: 2.235 | Val. Acc: 31.02%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 37s
Train Loss: 2.107 | Train Acc: 33.50%
Val. Loss: 2.212 | Val. Acc: 31.72%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 36s
Train Loss: 2.054 | Train Acc: 36.79%
Val. Loss: 2.192 | Val. Acc: 32.46%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 37s
Train Loss: 2.018 | Train Acc: 36.91%
Val. Loss: 2.165 | Val. Acc: 33.23%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 36s
Train Loss: 2.015 | Train Acc: 37.18%
Val. Loss: 2.182 | Val. Acc: 32.70%
83%|████████▎ | 53/64 [7:03:26<1:41:26, 553.32s/it][A[A
Experiment with params: mlpmixer_relu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 36s
Train Loss: 2.809 | Train Acc: 13.11%
Val. Loss: 2.702 | Val. Acc: 16.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 34s
Train Loss: 2.660 | Train Acc: 17.70%
Val. Loss: 2.605 | Val. Acc: 18.97%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 34s
Train Loss: 2.556 | Train Acc: 21.66%
Val. Loss: 2.559 | Val. Acc: 21.18%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 35s
Train Loss: 2.475 | Train Acc: 23.65%
Val. Loss: 2.478 | Val. Acc: 23.02%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 34s
Train Loss: 2.413 | Train Acc: 26.00%
Val. Loss: 2.460 | Val. Acc: 24.77%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 35s
Train Loss: 2.400 | Train Acc: 26.62%
Val. Loss: 2.392 | Val. Acc: 26.66%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 34s
Train Loss: 2.351 | Train Acc: 28.14%
Val. Loss: 2.387 | Val. Acc: 26.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 35s
Train Loss: 2.339 | Train Acc: 27.52%
Val. Loss: 2.346 | Val. Acc: 28.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 35s
Train Loss: 2.268 | Train Acc: 29.77%
Val. Loss: 2.360 | Val. Acc: 27.40%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 35s
Train Loss: 2.237 | Train Acc: 30.55%
Val. Loss: 2.287 | Val. Acc: 29.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 34s
Train Loss: 2.202 | Train Acc: 31.68%
Val. Loss: 2.294 | Val. Acc: 28.95%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.177 | Train Acc: 32.23%
Val. Loss: 2.266 | Val. Acc: 29.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 35s
Train Loss: 2.135 | Train Acc: 34.84%
Val. Loss: 2.233 | Val. Acc: 30.91%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 35s
Train Loss: 2.119 | Train Acc: 34.57%
Val. Loss: 2.225 | Val. Acc: 31.72%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 35s
Train Loss: 2.100 | Train Acc: 35.20%
Val. Loss: 2.199 | Val. Acc: 32.17%
84%|████████▍ | 54/64 [7:12:14<1:30:57, 545.72s/it][A[A
Experiment with params: mlpmixer_relu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 37s
Train Loss: 2.780 | Train Acc: 13.73%
Val. Loss: 2.638 | Val. Acc: 19.04%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 36s
Train Loss: 2.598 | Train Acc: 19.98%
Val. Loss: 2.534 | Val. Acc: 21.59%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 36s
Train Loss: 2.509 | Train Acc: 23.06%
Val. Loss: 2.467 | Val. Acc: 23.80%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 37s
Train Loss: 2.423 | Train Acc: 25.02%
Val. Loss: 2.432 | Val. Acc: 25.35%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 36s
Train Loss: 2.369 | Train Acc: 27.49%
Val. Loss: 2.367 | Val. Acc: 27.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 36s
Train Loss: 2.320 | Train Acc: 28.20%
Val. Loss: 2.346 | Val. Acc: 28.15%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 37s
Train Loss: 2.284 | Train Acc: 29.55%
Val. Loss: 2.316 | Val. Acc: 28.68%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 37s
Train Loss: 2.234 | Train Acc: 31.19%
Val. Loss: 2.292 | Val. Acc: 29.58%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 36s
Train Loss: 2.185 | Train Acc: 32.99%
Val. Loss: 2.266 | Val. Acc: 30.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 36s
Train Loss: 2.148 | Train Acc: 34.20%
Val. Loss: 2.229 | Val. Acc: 31.55%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 36s
Train Loss: 2.115 | Train Acc: 34.00%
Val. Loss: 2.219 | Val. Acc: 31.90%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 35s
Train Loss: 2.091 | Train Acc: 35.30%
Val. Loss: 2.207 | Val. Acc: 32.00%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 37s
Train Loss: 2.057 | Train Acc: 35.56%
Val. Loss: 2.183 | Val. Acc: 32.75%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 37s
Train Loss: 2.022 | Train Acc: 37.03%
Val. Loss: 2.179 | Val. Acc: 32.61%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 36s
Train Loss: 1.997 | Train Acc: 38.37%
Val. Loss: 2.135 | Val. Acc: 33.85%
86%|████████▌ | 55/64 [7:21:27<1:22:11, 547.92s/it][A[A
Experiment with params: mlpmixer_relu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 34s
Train Loss: 2.801 | Train Acc: 13.01%
Val. Loss: 2.712 | Val. Acc: 16.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 33s
Train Loss: 2.667 | Train Acc: 17.66%
Val. Loss: 2.620 | Val. Acc: 18.92%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 34s
Train Loss: 2.552 | Train Acc: 21.09%
Val. Loss: 2.529 | Val. Acc: 22.11%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 34s
Train Loss: 2.473 | Train Acc: 23.96%
Val. Loss: 2.486 | Val. Acc: 22.93%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 34s
Train Loss: 2.445 | Train Acc: 25.08%
Val. Loss: 2.435 | Val. Acc: 25.11%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 34s
Train Loss: 2.396 | Train Acc: 27.17%
Val. Loss: 2.408 | Val. Acc: 26.46%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 34s
Train Loss: 2.335 | Train Acc: 28.14%
Val. Loss: 2.377 | Val. Acc: 26.81%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 34s
Train Loss: 2.295 | Train Acc: 29.38%
Val. Loss: 2.379 | Val. Acc: 27.26%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 34s
Train Loss: 2.282 | Train Acc: 30.57%
Val. Loss: 2.341 | Val. Acc: 28.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 34s
Train Loss: 2.245 | Train Acc: 31.76%
Val. Loss: 2.289 | Val. Acc: 29.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 34s
Train Loss: 2.194 | Train Acc: 31.95%
Val. Loss: 2.303 | Val. Acc: 29.33%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 33s
Train Loss: 2.191 | Train Acc: 33.07%
Val. Loss: 2.268 | Val. Acc: 30.52%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 33s
Train Loss: 2.121 | Train Acc: 34.71%
Val. Loss: 2.265 | Val. Acc: 30.28%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 35s
Train Loss: 2.153 | Train Acc: 33.71%
Val. Loss: 2.204 | Val. Acc: 32.00%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 38s
Train Loss: 2.079 | Train Acc: 36.00%
Val. Loss: 2.197 | Val. Acc: 32.41%
88%|████████▊ | 56/64 [7:30:05<1:11:50, 538.79s/it][A[A
Experiment with params: mlpmixer_gelu_RAdam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 40s
Train Loss: 2.949 | Train Acc: 8.88%
Val. Loss: 2.836 | Val. Acc: 12.23%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 39s
Train Loss: 2.776 | Train Acc: 14.22%
Val. Loss: 2.730 | Val. Acc: 16.28%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 38s
Train Loss: 2.693 | Train Acc: 16.61%
Val. Loss: 2.655 | Val. Acc: 18.43%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 38s
Train Loss: 2.622 | Train Acc: 19.78%
Val. Loss: 2.595 | Val. Acc: 21.02%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 38s
Train Loss: 2.558 | Train Acc: 22.09%
Val. Loss: 2.538 | Val. Acc: 21.75%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 39s
Train Loss: 2.486 | Train Acc: 23.95%
Val. Loss: 2.486 | Val. Acc: 23.04%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.454 | Train Acc: 24.58%
Val. Loss: 2.448 | Val. Acc: 25.65%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 38s
Train Loss: 2.402 | Train Acc: 25.87%
Val. Loss: 2.412 | Val. Acc: 25.60%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.360 | Train Acc: 27.22%
Val. Loss: 2.407 | Val. Acc: 26.01%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 37s
Train Loss: 2.340 | Train Acc: 27.97%
Val. Loss: 2.367 | Val. Acc: 27.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 37s
Train Loss: 2.299 | Train Acc: 28.86%
Val. Loss: 2.346 | Val. Acc: 27.96%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 38s
Train Loss: 2.262 | Train Acc: 30.52%
Val. Loss: 2.302 | Val. Acc: 29.34%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 38s
Train Loss: 2.228 | Train Acc: 32.56%
Val. Loss: 2.305 | Val. Acc: 28.75%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 39s
Train Loss: 2.195 | Train Acc: 32.77%
Val. Loss: 2.300 | Val. Acc: 29.47%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 38s
Train Loss: 2.164 | Train Acc: 33.94%
Val. Loss: 2.255 | Val. Acc: 31.18%
89%|████████▉ | 57/64 [7:39:48<1:04:25, 552.17s/it][A[A
Experiment with params: mlpmixer_gelu_RAdam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 38s
Train Loss: 2.982 | Train Acc: 6.25%
Val. Loss: 2.901 | Val. Acc: 11.16%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 37s
Train Loss: 2.841 | Train Acc: 12.25%
Val. Loss: 2.795 | Val. Acc: 14.69%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 38s
Train Loss: 2.766 | Train Acc: 14.51%
Val. Loss: 2.738 | Val. Acc: 16.16%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 38s
Train Loss: 2.724 | Train Acc: 15.98%
Val. Loss: 2.684 | Val. Acc: 17.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 37s
Train Loss: 2.662 | Train Acc: 18.22%
Val. Loss: 2.631 | Val. Acc: 19.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 38s
Train Loss: 2.600 | Train Acc: 19.86%
Val. Loss: 2.573 | Val. Acc: 20.88%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.537 | Train Acc: 22.07%
Val. Loss: 2.541 | Val. Acc: 21.87%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 38s
Train Loss: 2.500 | Train Acc: 22.01%
Val. Loss: 2.512 | Val. Acc: 22.30%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 37s
Train Loss: 2.454 | Train Acc: 24.04%
Val. Loss: 2.465 | Val. Acc: 24.77%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 37s
Train Loss: 2.426 | Train Acc: 25.72%
Val. Loss: 2.440 | Val. Acc: 25.07%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 36s
Train Loss: 2.398 | Train Acc: 26.07%
Val. Loss: 2.413 | Val. Acc: 26.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 39s
Train Loss: 2.352 | Train Acc: 27.60%
Val. Loss: 2.402 | Val. Acc: 26.11%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 38s
Train Loss: 2.332 | Train Acc: 29.00%
Val. Loss: 2.396 | Val. Acc: 26.27%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 37s
Train Loss: 2.320 | Train Acc: 28.12%
Val. Loss: 2.365 | Val. Acc: 27.43%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 38s
Train Loss: 2.299 | Train Acc: 30.12%
Val. Loss: 2.346 | Val. Acc: 27.93%
91%|█████████ | 58/64 [7:49:20<55:49, 558.23s/it] [A[A
Experiment with params: mlpmixer_gelu_RAdam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 39s
Train Loss: 2.972 | Train Acc: 7.93%
Val. Loss: 2.840 | Val. Acc: 12.45%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 39s
Train Loss: 2.769 | Train Acc: 14.58%
Val. Loss: 2.710 | Val. Acc: 15.77%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 39s
Train Loss: 2.665 | Train Acc: 17.94%
Val. Loss: 2.617 | Val. Acc: 18.41%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 39s
Train Loss: 2.571 | Train Acc: 20.65%
Val. Loss: 2.559 | Val. Acc: 21.54%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 39s
Train Loss: 2.502 | Train Acc: 22.90%
Val. Loss: 2.491 | Val. Acc: 23.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 38s
Train Loss: 2.447 | Train Acc: 25.00%
Val. Loss: 2.456 | Val. Acc: 23.97%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.400 | Train Acc: 25.53%
Val. Loss: 2.436 | Val. Acc: 24.89%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 38s
Train Loss: 2.368 | Train Acc: 26.76%
Val. Loss: 2.409 | Val. Acc: 25.95%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.338 | Train Acc: 27.99%
Val. Loss: 2.376 | Val. Acc: 27.08%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 38s
Train Loss: 2.282 | Train Acc: 29.41%
Val. Loss: 2.367 | Val. Acc: 27.22%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 39s
Train Loss: 2.246 | Train Acc: 30.52%
Val. Loss: 2.320 | Val. Acc: 28.94%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 38s
Train Loss: 2.205 | Train Acc: 31.35%
Val. Loss: 2.308 | Val. Acc: 28.65%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 38s
Train Loss: 2.175 | Train Acc: 32.91%
Val. Loss: 2.264 | Val. Acc: 30.18%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 38s
Train Loss: 2.152 | Train Acc: 33.64%
Val. Loss: 2.241 | Val. Acc: 31.22%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 39s
Train Loss: 2.114 | Train Acc: 35.72%
Val. Loss: 2.233 | Val. Acc: 31.14%
92%|█████████▏| 59/64 [7:59:06<47:12, 566.40s/it][A[A
Experiment with params: mlpmixer_gelu_RAdam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 38s
Train Loss: 3.006 | Train Acc: 4.75%
Val. Loss: 2.923 | Val. Acc: 9.90%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 37s
Train Loss: 2.864 | Train Acc: 11.66%
Val. Loss: 2.804 | Val. Acc: 13.37%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 38s
Train Loss: 2.778 | Train Acc: 14.18%
Val. Loss: 2.737 | Val. Acc: 16.30%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 37s
Train Loss: 2.700 | Train Acc: 16.43%
Val. Loss: 2.687 | Val. Acc: 17.60%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 38s
Train Loss: 2.660 | Train Acc: 17.79%
Val. Loss: 2.646 | Val. Acc: 18.47%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 37s
Train Loss: 2.620 | Train Acc: 19.73%
Val. Loss: 2.597 | Val. Acc: 20.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.567 | Train Acc: 21.58%
Val. Loss: 2.558 | Val. Acc: 21.83%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 37s
Train Loss: 2.526 | Train Acc: 22.05%
Val. Loss: 2.527 | Val. Acc: 22.59%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 37s
Train Loss: 2.502 | Train Acc: 22.44%
Val. Loss: 2.489 | Val. Acc: 23.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 37s
Train Loss: 2.458 | Train Acc: 23.54%
Val. Loss: 2.466 | Val. Acc: 24.37%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 38s
Train Loss: 2.418 | Train Acc: 25.43%
Val. Loss: 2.445 | Val. Acc: 24.81%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 36s
Train Loss: 2.414 | Train Acc: 25.41%
Val. Loss: 2.423 | Val. Acc: 26.04%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 37s
Train Loss: 2.361 | Train Acc: 26.99%
Val. Loss: 2.402 | Val. Acc: 26.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 38s
Train Loss: 2.333 | Train Acc: 28.67%
Val. Loss: 2.383 | Val. Acc: 26.94%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 36s
Train Loss: 2.299 | Train Acc: 29.41%
Val. Loss: 2.367 | Val. Acc: 27.85%
94%|█████████▍| 60/64 [8:08:33<37:46, 566.68s/it][A[A
Experiment with params: mlpmixer_gelu_Adam_crossentropy_64_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 38s
Train Loss: 2.799 | Train Acc: 13.67%
Val. Loss: 2.686 | Val. Acc: 16.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 38s
Train Loss: 2.605 | Train Acc: 20.02%
Val. Loss: 2.559 | Val. Acc: 20.74%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 37s
Train Loss: 2.513 | Train Acc: 22.35%
Val. Loss: 2.493 | Val. Acc: 23.25%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 37s
Train Loss: 2.429 | Train Acc: 24.39%
Val. Loss: 2.461 | Val. Acc: 24.02%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 38s
Train Loss: 2.362 | Train Acc: 27.59%
Val. Loss: 2.494 | Val. Acc: 23.50%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 38s
Train Loss: 2.334 | Train Acc: 27.97%
Val. Loss: 2.379 | Val. Acc: 26.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.271 | Train Acc: 30.18%
Val. Loss: 2.317 | Val. Acc: 28.39%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 39s
Train Loss: 2.232 | Train Acc: 32.02%
Val. Loss: 2.285 | Val. Acc: 29.20%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.163 | Train Acc: 33.48%
Val. Loss: 2.301 | Val. Acc: 29.79%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 38s
Train Loss: 2.155 | Train Acc: 33.70%
Val. Loss: 2.254 | Val. Acc: 30.08%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 38s
Train Loss: 2.101 | Train Acc: 35.32%
Val. Loss: 2.224 | Val. Acc: 31.09%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 38s
Train Loss: 2.062 | Train Acc: 35.54%
Val. Loss: 2.217 | Val. Acc: 31.54%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 37s
Train Loss: 2.032 | Train Acc: 36.81%
Val. Loss: 2.184 | Val. Acc: 32.98%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 40s
Train Loss: 2.012 | Train Acc: 37.94%
Val. Loss: 2.189 | Val. Acc: 32.45%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 43s
Train Loss: 1.996 | Train Acc: 37.92%
Val. Loss: 2.180 | Val. Acc: 32.86%
95%|█████████▌| 61/64 [8:18:15<28:33, 571.26s/it][A[A
Experiment with params: mlpmixer_gelu_Adam_crossentropy_64_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 42s
Train Loss: 2.807 | Train Acc: 13.36%
Val. Loss: 2.697 | Val. Acc: 17.60%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 46s
Train Loss: 2.646 | Train Acc: 19.24%
Val. Loss: 2.595 | Val. Acc: 19.41%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 44s
Train Loss: 2.547 | Train Acc: 21.78%
Val. Loss: 2.524 | Val. Acc: 23.14%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 43s
Train Loss: 2.469 | Train Acc: 24.38%
Val. Loss: 2.504 | Val. Acc: 22.98%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 43s
Train Loss: 2.427 | Train Acc: 26.05%
Val. Loss: 2.460 | Val. Acc: 24.67%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 44s
Train Loss: 2.425 | Train Acc: 24.94%
Val. Loss: 2.405 | Val. Acc: 25.78%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 42s
Train Loss: 2.326 | Train Acc: 27.58%
Val. Loss: 2.373 | Val. Acc: 27.01%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 42s
Train Loss: 2.290 | Train Acc: 29.08%
Val. Loss: 2.355 | Val. Acc: 27.06%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 39s
Train Loss: 2.259 | Train Acc: 29.88%
Val. Loss: 2.315 | Val. Acc: 28.28%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 40s
Train Loss: 2.219 | Train Acc: 32.01%
Val. Loss: 2.311 | Val. Acc: 28.61%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 39s
Train Loss: 2.195 | Train Acc: 32.09%
Val. Loss: 2.305 | Val. Acc: 28.53%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 39s
Train Loss: 2.189 | Train Acc: 32.09%
Val. Loss: 2.279 | Val. Acc: 29.50%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 39s
Train Loss: 2.161 | Train Acc: 33.61%
Val. Loss: 2.250 | Val. Acc: 30.40%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 38s
Train Loss: 2.110 | Train Acc: 34.30%
Val. Loss: 2.309 | Val. Acc: 29.68%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 40s
Train Loss: 2.095 | Train Acc: 34.14%
Val. Loss: 2.259 | Val. Acc: 30.35%
97%|█████████▋| 62/64 [8:28:44<19:37, 588.52s/it][A[A
Experiment with params: mlpmixer_gelu_Adam_crossentropy_128_64_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 42s
Train Loss: 2.779 | Train Acc: 14.20%
Val. Loss: 2.666 | Val. Acc: 17.82%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 40s
Train Loss: 2.591 | Train Acc: 19.76%
Val. Loss: 2.540 | Val. Acc: 21.14%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 40s
Train Loss: 2.481 | Train Acc: 23.73%
Val. Loss: 2.461 | Val. Acc: 24.65%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 40s
Train Loss: 2.427 | Train Acc: 25.87%
Val. Loss: 2.443 | Val. Acc: 25.18%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 40s
Train Loss: 2.373 | Train Acc: 26.92%
Val. Loss: 2.381 | Val. Acc: 26.93%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 40s
Train Loss: 2.325 | Train Acc: 28.26%
Val. Loss: 2.360 | Val. Acc: 27.36%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 41s
Train Loss: 2.285 | Train Acc: 29.45%
Val. Loss: 2.329 | Val. Acc: 28.29%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 41s
Train Loss: 2.255 | Train Acc: 30.12%
Val. Loss: 2.315 | Val. Acc: 28.72%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 41s
Train Loss: 2.202 | Train Acc: 31.65%
Val. Loss: 2.305 | Val. Acc: 29.11%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 39s
Train Loss: 2.188 | Train Acc: 32.99%
Val. Loss: 2.246 | Val. Acc: 31.03%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 40s
Train Loss: 2.152 | Train Acc: 32.83%
Val. Loss: 2.237 | Val. Acc: 30.84%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 39s
Train Loss: 2.116 | Train Acc: 33.62%
Val. Loss: 2.244 | Val. Acc: 30.49%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 40s
Train Loss: 2.069 | Train Acc: 35.07%
Val. Loss: 2.220 | Val. Acc: 31.69%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 41s
Train Loss: 2.061 | Train Acc: 36.06%
Val. Loss: 2.202 | Val. Acc: 32.36%
Training: 0%| | 0/79 [00:00<?, ?it/s]
Evaluating: 0%| | 0/704 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 40s
Train Loss: 2.018 | Train Acc: 37.42%
Val. Loss: 2.186 | Val. Acc: 32.63%
98%|█████████▊| 63/64 [8:38:57<09:55, 595.97s/it][A[A
Experiment with params: mlpmixer_gelu_Adam_crossentropy_128_128_15
0%| | 0/15 [00:00<?, ?it/s]
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 01 | Epoch Time: 0m 39s
Train Loss: 2.817 | Train Acc: 13.36%
Val. Loss: 2.714 | Val. Acc: 15.73%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 02 | Epoch Time: 0m 38s
Train Loss: 2.650 | Train Acc: 17.89%
Val. Loss: 2.611 | Val. Acc: 19.05%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 03 | Epoch Time: 0m 37s
Train Loss: 2.555 | Train Acc: 20.98%
Val. Loss: 2.506 | Val. Acc: 22.97%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 04 | Epoch Time: 0m 39s
Train Loss: 2.484 | Train Acc: 23.44%
Val. Loss: 2.466 | Val. Acc: 24.02%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 05 | Epoch Time: 0m 39s
Train Loss: 2.431 | Train Acc: 24.94%
Val. Loss: 2.437 | Val. Acc: 25.82%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 06 | Epoch Time: 0m 38s
Train Loss: 2.385 | Train Acc: 26.43%
Val. Loss: 2.414 | Val. Acc: 26.49%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 07 | Epoch Time: 0m 38s
Train Loss: 2.325 | Train Acc: 29.02%
Val. Loss: 2.365 | Val. Acc: 27.57%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 08 | Epoch Time: 0m 39s
Train Loss: 2.290 | Train Acc: 30.08%
Val. Loss: 2.369 | Val. Acc: 26.34%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 09 | Epoch Time: 0m 38s
Train Loss: 2.267 | Train Acc: 29.57%
Val. Loss: 2.328 | Val. Acc: 28.79%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 10 | Epoch Time: 0m 39s
Train Loss: 2.249 | Train Acc: 30.14%
Val. Loss: 2.310 | Val. Acc: 29.02%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 11 | Epoch Time: 0m 38s
Train Loss: 2.201 | Train Acc: 32.75%
Val. Loss: 2.308 | Val. Acc: 29.02%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 12 | Epoch Time: 0m 39s
Train Loss: 2.180 | Train Acc: 33.16%
Val. Loss: 2.253 | Val. Acc: 30.55%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 13 | Epoch Time: 0m 39s
Train Loss: 2.155 | Train Acc: 33.44%
Val. Loss: 2.241 | Val. Acc: 31.32%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 14 | Epoch Time: 0m 39s
Train Loss: 2.102 | Train Acc: 35.21%
Val. Loss: 2.264 | Val. Acc: 30.38%
Training: 0%| | 0/40 [00:00<?, ?it/s]
Evaluating: 0%| | 0/352 [00:00<?, ?it/s]
Epoch: 15 | Epoch Time: 0m 39s
Train Loss: 2.110 | Train Acc: 34.94%
Val. Loss: 2.227 | Val. Acc: 31.71%
100%|██████████| 64/64 [8:48:46<00:00, 495.72s/it]
Hyperparameter Tuning Results
output_path = "talos_output/031823231703.csv"
df_result = pd.read_csv(output_path)
df_result
| round_epochs | loss | metric | val_loss | val_metric | model_name | activation | optimizer | losses | units | batch_size | epochs | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 15 | 2.343323 | 0.266416 | 2.487088 | 0.227406 | mlprgb | relu | RAdam | crossentropy | 64 | 64 | 15 |
| 1 | 15 | 2.442377 | 0.244141 | 2.517502 | 0.214528 | mlprgb | relu | RAdam | crossentropy | 64 | 128 | 15 |
| 2 | 15 | 2.208389 | 0.309335 | 2.466630 | 0.249068 | mlprgb | relu | RAdam | crossentropy | 128 | 64 | 15 |
| 3 | 15 | 2.334524 | 0.274219 | 2.473362 | 0.235204 | mlprgb | relu | RAdam | crossentropy | 128 | 128 | 15 |
| 4 | 15 | 2.222436 | 0.307160 | 2.465534 | 0.247425 | mlprgb | relu | Adam | crossentropy | 64 | 64 | 15 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 59 | 15 | 2.299378 | 0.294141 | 2.366793 | 0.278505 | mlpmixer | gelu | RAdam | crossentropy | 128 | 128 | 15 |
| 60 | 15 | 1.995841 | 0.379153 | 2.180399 | 0.328569 | mlpmixer | gelu | Adam | crossentropy | 64 | 64 | 15 |
| 61 | 15 | 2.095319 | 0.341406 | 2.258898 | 0.303521 | mlpmixer | gelu | Adam | crossentropy | 64 | 128 | 15 |
| 62 | 15 | 2.017702 | 0.374209 | 2.186394 | 0.326261 | mlpmixer | gelu | Adam | crossentropy | 128 | 64 | 15 |
| 63 | 15 | 2.109830 | 0.349414 | 2.226770 | 0.317067 | mlpmixer | gelu | Adam | crossentropy | 128 | 128 | 15 |
64 rows × 12 columns
df_result['val_metric'].hist()
<Axes: >
#best model
df_best = df_result[df_result['val_metric'] == np.max(df_result['val_metric'])]
df_best
| round_epochs | loss | metric | val_loss | val_metric | model_name | activation | optimizer | losses | units | batch_size | epochs | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 54 | 15 | 1.997359 | 0.383703 | 2.134726 | 0.338534 | mlpmixer | relu | Adam | crossentropy | 128 | 64 | 15 |
best_model_name = f"cifar100_{df_best['model_name'].values[0]}_{df_best['activation'].values[0]}_{df_best['optimizer'].values[0]}_{df_best['losses'].values[0]}_{df_best['units'].values[0]}_{df_best['batch_size'].values[0]}_{df_best['epochs'].values[0]}.pt"
best_model_name
'cifar100_mlpmixer_relu_Adam_crossentropy_128_64_15.pt'
class MLPCIFAR(nn.Module, talos.utils.TorchHistory):
def __init__(self, params):
super(MLPCIFAR, self).__init__()
self.model_name = params['model_name']
self.act_fun = params['activation']
self.units = params['units']
if self.model_name == 'mlprgb':
#activate function
if self.act_fun == 'relu':
self.activate = F.relu
else:
self.activate = F.gelu
self.model = MLPRGB(input_dim=32*32*3,output_dim=20, units=self.units, act_fun=self.activate)
elif self.model_name == 'mlpgray':
#activate function
if self.act_fun == 'relu':
self.activate = F.relu
else:
self.activate = F.gelu
self.model = MLPGRAY(input_dim=32*32*3,output_dim=20, units=self.units, act_fun=self.activate)
elif self.model_name == 'resmlp':
#activate function
if self.act_fun == 'relu':
self.activate = nn.ReLU()
else:
self.activate = nn.GELU()
self.model = ResMLP(input_dim=32,output_dim=20, act_fun=self.activate)
else:
#activate function
if self.act_fun == 'relu':
self.activate = nn.ReLU()
else:
self.activate = nn.GELU()
self.model = MLPMixer(input_dim=32,output_dim=20, act_fun=self.activate)
def forward(self, x):
x = self.model(x)
return x
Inference The Best Model on Test set
# create model
best_params = {
'model_name':df_best['model_name'].values[0],
'activation':df_best['activation'].values[0],
'units':df_best['units'].values[0]
}
model = MLPCIFAR(best_params)
model.load_state_dict(torch.load(best_model_name))
#optimizer
optimizer = optim.Adam(model.parameters())
#loss function
criterion = nn.CrossEntropyLoss()
#move model to GPU for faster training
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model = model.to(device)
criterion = criterion.to(device)
test_loss, test_acc = evaluate(model, test_iterator, criterion, device)
print(f'Test Loss: {test_loss:.3f} | Test Acc: {test_acc*100:.2f}%')
Evaluating: 0%| | 0/157 [00:00<?, ?it/s]
Test Loss: 2.166 | Test Acc: 33.22%
Diagnostics The Best Model
#prediction function
def get_predictions(model, iterator, device):
model.eval()
images = []
labels = []
probs = []
with torch.no_grad():
for (x, y) in iterator:
x = x.to(device)
y_pred = model(x)
y_prob = F.softmax(y_pred, dim=-1)
images.append(x.cpu())
labels.append(y.cpu())
probs.append(y_prob.cpu())
images = torch.cat(images, dim=0)
labels = torch.cat(labels, dim=0)
probs = torch.cat(probs, dim=0)
return images, labels, probs
#get predict for test set
images, labels, probs = get_predictions(model, test_iterator, device)
pred_labels = torch.argmax(probs, 1)
#plot confusion matrix
def plot_confusion_matrix(labels, pred_labels):
fig = plt.figure(figsize=(15, 15))
ax = fig.add_subplot(1, 1, 1)
cm = metrics.confusion_matrix(labels, pred_labels)
cm = metrics.ConfusionMatrixDisplay(cm, display_labels=range(20))
cm.plot(values_format='d', cmap='Blues', ax=ax)
#show plot
plot_confusion_matrix(labels, pred_labels)
#get uncorrect label
corrects = torch.eq(labels, pred_labels)
incorrect_examples = []
for image, label, prob, correct in zip(images, labels, probs, corrects):
if not correct:
incorrect_examples.append((image, label, prob))
incorrect_examples.sort(reverse=True,
key=lambda x: torch.max(x[2], dim=0).values)
#plot uncorrect label
def plot_most_incorrect(incorrect, n_images):
rows = int(np.sqrt(n_images))
cols = int(np.sqrt(n_images))
fig = plt.figure(figsize=(20, 10))
for i in range(rows*cols):
ax = fig.add_subplot(rows, cols, i+1)
image, true_label, probs = incorrect[i]
true_prob = probs[true_label]
incorrect_prob, incorrect_label = torch.max(probs, dim=0)
ax.imshow(image.view(32, 32,3).cpu().numpy(), cmap='bone')
ax.set_title(f'true label: {true_label} ({true_prob:.3f})\n'
f'pred label: {incorrect_label} ({incorrect_prob:.3f})')
ax.axis('off')
fig.subplots_adjust(hspace=0.5)
#show uncorrect image
N_IMAGES = 9
plot_most_incorrect(incorrect_examples, N_IMAGES)
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).
