# License: BSD
# Author: Sasank Chilamkurthy
from __future__ import print_function, division
import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim import lr_scheduler
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
import copy
from tqdm import tqdm
plt.ion() # interactive mode
mean = [0.5671, 0.5770, 0.5728]
std = [0.1897, 0.1718, 0.1553]
train_transforms = transforms.Compose([
transforms.RandomHorizontalFlip(p=0.5),
transforms.RandomCrop(1),
transforms.RandomRotation(degrees=5),
transforms.Resize((1500, 1500)),
transforms.ToTensor(),
transforms.Normalize(mean, std)])
test_transforms = transforms.Compose([
transforms.Resize((1500, 1500)),
transforms.ToTensor(),
transforms.Normalize(mean, std)])
num_workers = 0
batch_size = 3
root_train = '/home/dev-25/Occupancy_Type_Image_Classification_NF_CI/image_classification_occupancy_type/Development/archive/flower_data/temp_data'
root_test = '/home/dev-25/Occupancy_Type_Image_Classification_NF_CI/image_classification_occupancy_type/Development/archive/flower_data/temp_data'
train_data = datasets.ImageFolder(root_train, transform=train_transforms)
test_data = datasets.ImageFolder(root_test, transform=test_transforms)
print("Train size:{}".format(len(train_data)))
print("Valid size:{}".format(len(test_data)))
train_loader = torch.utils.data.DataLoader(
train_data, batch_size=batch_size, shuffle=True, num_workers=num_workers)
test_loader = torch.utils.data.DataLoader(
test_data, batch_size=batch_size, num_workers=num_workers)
class_names = train_data.classes
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
def train_model(model, criterion, optimizer, scheduler, num_epochs=25):
since = time.time()
best_model_wts = copy.deepcopy(model.state_dict())
best_acc = 0.0
for epoch in range(num_epochs):
print('Epoch {}/{}'.format(epoch, num_epochs - 1))
print('-' * 10)
# Each epoch has a training and validation phase
for phase in ['train', 'val']:
if phase == 'train':
model.train() # Set model to training mode
else:
model.eval() # Set model to evaluate mode
running_loss = 0.0
running_corrects = 0
# Iterate over data.
if phase=='train':
dataloader = test_loader
elif phase=='val':
dataloader = test_loader
for inputs, labels in tqdm(dataloader):
inputs = inputs.cuda()
labels = labels.cuda()
# zero the parameter gradients
optimizer.zero_grad()
# forward
# track history if only in train
with torch.set_grad_enabled(phase == 'train'):
outputs = model(inputs)
_, preds = torch.max(outputs, 1)
loss = criterion(outputs, labels)
# backward + optimize only if in training phase
if phase == 'train':
loss.backward()
optimizer.step()
# statistics
running_loss += loss.item() * inputs.size(0)
running_corrects += torch.sum(preds == labels.data)
if phase == 'train':
scheduler.step()
if phase=='train':
epoch_loss = running_loss / len(train_data)
epoch_acc = running_corrects.double() / len(train_data)
if phase=='val':
epoch_loss = running_loss / len(test_data)
epoch_acc = running_corrects.double() / len(test_data)
print('{} Loss: {:.4f} Acc: {:.4f}'.format(
phase, epoch_loss, epoch_acc))
# deep copy the model
if phase == 'val' and epoch_acc > best_acc:
best_acc = epoch_acc
best_model_wts = copy.deepcopy(model.state_dict())
time_elapsed = time.time() - since
print('Training complete in {:.0f}m {:.0f}s'.format(
time_elapsed // 60, time_elapsed % 60))
print('Best val Acc: {:4f}'.format(best_acc))
# load best model weights
model.load_state_dict(best_model_wts)
return model
model_ft = models.resnet18(pretrained=True)
num_ftrs = model_ft.fc.in_features
# Here the size of each output sample is set to 2.
# Alternatively, it can be generalized to nn.Linear(num_ftrs, len(class_names)).
model_ft.fc = nn.Linear(num_ftrs, 3)
model_ft = model_ft.cuda()
criterion = nn.CrossEntropyLoss()
# Observe that all parameters are being optimized
optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.003)
# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=250, gamma=0.1)
model_ft = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler,
num_epochs=400)
I am using resnet18 model with SGD Optimizer with LR 0.03.
Epoch 253/399
----------
100%|██████████████████████████████████████████████████████████████████| 6/6 [00:02<00:00, 2.01it/s]
train Loss: 0.0130 Acc: 1.0000
100%|██████████████████████████████████████████████████████████████████| 6/6 [00:01<00:00, 4.11it/s]
val Loss: 0.6453 Acc: 0.6250
Epoch 254/399
----------
100%|██████████████████████████████████████████████████████████████████| 6/6 [00:02<00:00, 2.01it/s]
train Loss: 0.0130 Acc: 1.0000
100%|██████████████████████████████████████████████████████████████████| 6/6 [00:01<00:00, 4.10it/s]
val Loss: 0.6454 Acc: 0.6250
Epoch 255/399
----------
100%|██████████████████████████████████████████████████████████████████| 6/6 [00:03<00:00, 1.99it/s]
train Loss: 0.0130 Acc: 1.0000
100%|██████████████████████████████████████████████████████████████████| 6/6 [00:01<00:00, 4.14it/s]
val Loss: 0.6454 Acc: 0.6250