Remove useless files in model compression examples (#3242)

examples/model_compress/model_prune_torch.py → examples/model_compress/pruning/model_prune_torch.py

@@ -212,7 +212,7 @@ def main(args):
     train_loader, test_loader = get_data_loaders(dataset_name, args.batch_size)
     dummy_input, _ = next(iter(train_loader))
     dummy_input = dummy_input.to(device)
-    model = create_model(model_name).cuda()
+    model = create_model(model_name).to(device)
     if args.resume_from is not None and os.path.exists(args.resume_from):
         print('loading checkpoint {} ...'.format(args.resume_from))
         model.load_state_dict(torch.load(args.resume_from))

examples/model_compress/model_speedup.py → examples/model_compress/pruning/model_speedup.py

@@ -16,25 +16,21 @@ compare_results = True
 config = {
     'apoz': {
         'model_name': 'vgg16',
-        'device': 'cuda',
         'input_shape': [64, 3, 32, 32],
         'masks_file': './checkpoints/mask_vgg16_cifar10_apoz.pth'
     },
     'l1filter': {
         'model_name': 'vgg16',
-        'device': 'cuda',
         'input_shape': [64, 3, 32, 32],
         'masks_file': './checkpoints/mask_vgg16_cifar10_l1filter.pth'
     },
     'fpgm': {
         'model_name': 'naive',
-        'device': 'cpu',
         'input_shape': [64, 1, 28, 28],
         'masks_file': './checkpoints/mask_naive_mnist_fpgm.pth'
     },
     'slim': {
         'model_name': 'vgg19',
-        'device': 'cuda',
         'input_shape': [64, 3, 32, 32],
         'masks_file': './checkpoints/mask_vgg19_cifar10_slim.pth' #'mask_vgg19_cifar10.pth'
     }
@@ -42,7 +38,10 @@ config = {
 
 def model_inference(config):
     masks_file = config['masks_file']
-    device = torch.device(config['device'])
+    device = torch.device(
+        'cuda') if torch.cuda.is_available() else torch.device('cpu')
+        
+    # device = torch.device(config['device'])
     if config['model_name'] == 'vgg16':
         model = VGG(depth=16)
     elif config['model_name'] == 'vgg19':
@@ -57,14 +56,13 @@ def model_inference(config):
     use_mask_out = use_speedup_out = None
     # must run use_mask before use_speedup because use_speedup modify the model
     if use_mask:
-        apply_compression_results(model, masks_file, 'cpu' if config['device'] == 'cpu' else None)
+        apply_compression_results(model, masks_file, device)
         start = time.time()
         for _ in range(32):
             use_mask_out = model(dummy_input)
         print('elapsed time when use mask: ', time.time() - start)
     if use_speedup:
-        m_speedup = ModelSpeedup(model, dummy_input, masks_file,
-                                 'cpu' if config['device'] == 'cpu' else None)
+        m_speedup = ModelSpeedup(model, dummy_input, masks_file, device)
         m_speedup.speedup_model()
         start = time.time()
         for _ in range(32):

examples/model_compress/L1_torch_cifar10.py → examples/model_compress/pruning/reproduced/L1_torch_pruner.py

@@ -119,3 +119,4 @@ def main():
 
 if __name__ == '__main__':
     main()
+    
\ No newline at end of file

examples/model_compress/lottery_torch_mnist_fc.py → examples/model_compress/pruning/reproduced/lottery_torch_mnist_fc.py

@@ -28,13 +28,17 @@ class fc1(nn.Module):
 def train(model, train_loader, optimizer, criterion):
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model.train()
-    for imgs, targets in train_loader:
+    for batch_idx, (imgs, targets) in enumerate(train_loader):
         optimizer.zero_grad()
         imgs, targets = imgs.to(device), targets.to(device)
         output = model(imgs)
         train_loss = criterion(output, targets)
         train_loss.backward()
         optimizer.step()
+        if batch_idx % 100 == 0:
+            print('{:2.0f}%  Loss {}'.format(
+                100 * batch_idx / len(train_loader), train_loss.item()))
+
     return train_loss.item()
 
 def test(model, test_loader, criterion):

examples/model_compress/slim_torch_cifar10.py → examples/model_compress/pruning/reproduced/slim_torch_pruner.py

@@ -143,3 +143,4 @@ def main():
 
 if __name__ == '__main__':
     main()
+    
\ No newline at end of file

examples/model_compress/pruning_kd.py

-import math
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from torchvision import datasets, transforms
-from nni.algorithms.compression.pytorch.pruning import L1FilterPruner
-from knowledge_distill.knowledge_distill import KnowledgeDistill
-from models.cifar10.vgg import VGG
-
-
-def train(model, device, train_loader, optimizer, kd=None):
-    alpha = 1
-    beta = 0.8
-    model.train()
-    for batch_idx, (data, target) in enumerate(train_loader):
-        data, target = data.to(device), target.to(device)
-        optimizer.zero_grad()
-        output = model(data)
-        student_loss = F.cross_entropy(output, target)
-        if kd is not None:
-            kd_loss = kd.loss(data=data, student_out=output)
-            loss = alpha * student_loss + beta * kd_loss
-        else:
-            loss = student_loss
-        loss.backward()
-        optimizer.step()
-        if batch_idx % 100 == 0:
-            print('{:2.0f}%  Loss {}'.format(100 * batch_idx / len(train_loader), loss.item()))
-
-
-def test(model, device, test_loader):
-    model.eval()
-    test_loss = 0
-    correct = 0
-    with torch.no_grad():
-        for data, target in test_loader:
-            data, target = data.to(device), target.to(device)
-            output = model(data)
-            test_loss += F.nll_loss(output, target, reduction='sum').item()
-            pred = output.argmax(dim=1, keepdim=True)
-            correct += pred.eq(target.view_as(pred)).sum().item()
-    test_loss /= len(test_loader.dataset)
-    acc = 100 * correct / len(test_loader.dataset)
-
-    print('Loss: {}  Accuracy: {}%)\n'.format(
-        test_loss, acc))
-    return acc
-
-
-def main():
-    torch.manual_seed(0)
-    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-    train_loader = torch.utils.data.DataLoader(
-        datasets.CIFAR10('./data.cifar10', train=True, download=True,
-                         transform=transforms.Compose([
-                             transforms.Pad(4),
-                             transforms.RandomCrop(32),
-                             transforms.RandomHorizontalFlip(),
-                             transforms.ToTensor(),
-                             transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))
-                         ])),
-        batch_size=64, shuffle=True)
-    test_loader = torch.utils.data.DataLoader(
-        datasets.CIFAR10('./data.cifar10', train=False, transform=transforms.Compose([
-            transforms.ToTensor(),
-            transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))
-        ])),
-        batch_size=200, shuffle=False)
-
-    model = VGG(depth=16)
-    model.to(device)
-
-    # Train the base VGG-16 model
-    print('=' * 10 + 'Train the unpruned base model' + '=' * 10)
-    optimizer = torch.optim.SGD(model.parameters(), lr=0.1, momentum=0.9, weight_decay=1e-4)
-    lr_scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, 160, 0)
-    for epoch in range(160):
-        print('# Epoch {} #'.format(epoch))
-        train(model, device, train_loader, optimizer)
-        test(model, device, test_loader)
-        lr_scheduler.step(epoch)
-    torch.save(model.state_dict(), 'vgg16_cifar10.pth')
-
-    # Test base model accuracy
-    print('=' * 10 + 'Test on the original model' + '=' * 10)
-    model.load_state_dict(torch.load('vgg16_cifar10.pth'))
-    test(model, device, test_loader)
-    # top1 = 93.51%
-
-    # Pruning Configuration, all convolution layers are pruned out 80% filters according to the L1 norm
-    configure_list = [{
-        'sparsity': 0.8,
-        'op_types': ['Conv2d'],
-    }]
-
-    # Prune model and test accuracy without fine tuning.
-    print('=' * 10 + 'Test on the pruned model before fine tune' + '=' * 10)
-    pruner = L1FilterPruner(model, configure_list)
-    model = pruner.compress()
-    test(model, device, test_loader)
-    # top1 = 10.00%
-
-    # Fine tune the pruned model for 40 epochs and test accuracy
-    print('=' * 10 + 'Fine tuning' + '=' * 10)
-    optimizer_finetune = torch.optim.SGD(model.parameters(), lr=0.001, momentum=0.9, weight_decay=1e-4)
-    best_top1 = 0
-    kd_teacher_model = VGG(depth=16)
-    kd_teacher_model.to(device)
-    kd_teacher_model.load_state_dict(torch.load('vgg16_cifar10.pth'))
-    kd = KnowledgeDistill(kd_teacher_model, kd_T=5)
-    for epoch in range(40):
-        pruner.update_epoch(epoch)
-        print('# Epoch {} #'.format(epoch))
-        train(model, device, train_loader, optimizer_finetune, kd)
-        top1 = test(model, device, test_loader)
-        if top1 > best_top1:
-            best_top1 = top1
-            # Export the best model, 'model_path' stores state_dict of the pruned model,
-            # mask_path stores mask_dict of the pruned model
-            pruner.export_model(model_path='pruned_vgg16_cifar10.pth', mask_path='mask_vgg16_cifar10.pth')
-
-    # Test the exported model
-    print('=' * 10 + 'Test on the pruned model after fine tune' + '=' * 10)
-    new_model = VGG(depth=16)
-    new_model.to(device)
-    new_model.load_state_dict(torch.load('pruned_vgg16_cifar10.pth'))
-    test(new_model, device, test_loader)
-    # top1 = 85.43% with kd, top1 = 85.04% without kd,
-
-
-if __name__ == '__main__':
-    main()