Resolve conflicts for #4760 (#4762)

docs/tools/translation_patch.py

+"""
+Fix a troublsome translation in sphinx.
+Related PR: https://github.com/sphinx-doc/sphinx/pull/10303
+"""
+
+import subprocess
+from pathlib import Path
+
+import sphinx
+
+sphinx_path = Path(sphinx.__path__[0]) / 'locale/zh_CN/LC_MESSAGES'
+po_content = (sphinx_path / 'sphinx.po').read_text()
+po_content = po_content.replace('%s的别名', '%s 的别名')
+(sphinx_path / 'sphinx.po').write_text(po_content)
+
+# build po -> mo
+subprocess.run(['msgfmt', '-c', str(sphinx_path / 'sphinx.po'), '-o', str(sphinx_path / 'sphinx.mo')], check=True)

examples/model_compress/end2end_compression.py

@@ -15,7 +15,7 @@ import torch.optim as optim
 from torch.optim.lr_scheduler import StepLR
 from torchvision import datasets, transforms
 
-from nni.compression.pytorch.utils.counter import count_flops_params
+from nni.compression.pytorch.utils import count_flops_params
 from nni.compression.pytorch import ModelSpeedup
 
 from nni.algorithms.compression.pytorch.pruning import L1FilterPruner
@@ -277,7 +277,7 @@ if __name__ == '__main__':
     parser.add_argument('--sparsity', type=float, default=0.5,
                         help='target overall target sparsity')
     parser.add_argument('--dependency-aware', action='store_true', default=False,
-                        help='toggle dependency aware mode')
+                        help='toggle dependency-aware mode')
 
     # finetuning
     parser.add_argument('--finetune-epochs', type=int, default=5,
@@ -290,8 +290,8 @@ if __name__ == '__main__':
     #                     help='learning rate to finetune the model')
 
     # speedup
-    # parser.add_argument('--speed-up', action='store_true', default=False,
-    #                     help='whether to speed-up the pruned model')
+    # parser.add_argument('--speedup', action='store_true', default=False,
+    #                     help='whether to speedup the pruned model')
 
     # parser.add_argument('--nni', action='store_true', default=False,
     #                     help="whether to tune the pruners using NNi tuners")

examples/model_compress/pruning/v2/activation_pruning_torch.py → examples/model_compress/pruning/activation_pruning_torch.py

@@ -4,7 +4,7 @@
 '''
 NNI example for supported ActivationAPoZRank and ActivationMeanRank pruning algorithms.
 In this example, we show the end-to-end pruning process: pre-training -> pruning -> fine-tuning.
-Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speed up is required.
+Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speedup is required.
 
 '''
 import argparse
@@ -16,11 +16,11 @@ from torch.optim.lr_scheduler import MultiStepLR
 
 import nni
 from nni.compression.pytorch import ModelSpeedup
-from nni.compression.pytorch.utils.counter import count_flops_params
-from nni.algorithms.compression.v2.pytorch.pruning.basic_pruner import ActivationAPoZRankPruner, ActivationMeanRankPruner
+from nni.compression.pytorch.utils import count_flops_params
+from nni.compression.pytorch.pruning import ActivationAPoZRankPruner, ActivationMeanRankPruner
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
 from cifar10.vgg import VGG
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

examples/model_compress/pruning/v2/admm_pruning_torch.py → examples/model_compress/pruning/admm_pruning_torch.py

@@ -4,7 +4,7 @@
 '''
 NNI example for supported ADMM pruning algorithms.
 In this example, we show the end-to-end pruning process: pre-training -> pruning -> fine-tuning.
-Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speed up is required.
+Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speedup is required.
 
 '''
 import argparse
@@ -15,11 +15,12 @@ from torchvision import datasets, transforms
 from torch.optim.lr_scheduler import MultiStepLR
 
 import nni
-from nni.compression.pytorch.utils.counter import count_flops_params
-from nni.algorithms.compression.v2.pytorch.pruning.basic_pruner import ADMMPruner
+from nni.compression.pytorch.speedup import ModelSpeedup
+from nni.compression.pytorch.utils import count_flops_params
+from nni.compression.pytorch.pruning import ADMMPruner
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
 from cifar10.vgg import VGG
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
@@ -108,18 +109,17 @@ if __name__ == '__main__':
     config_list = [{
         'sparsity': 0.8,
         'op_types': ['Conv2d'],
-    }, {
-        'sparsity': 0.92,
-        'op_types': ['Conv2d'],
     }]
 
     # make sure you have used nni.trace to wrap the optimizer class before initialize
     traced_optimizer = nni.trace(torch.optim.SGD)(model.parameters(), lr=0.01, momentum=0.9, weight_decay=5e-4)
-    pruner = ADMMPruner(model, config_list, trainer, traced_optimizer, criterion, iterations=2, training_epochs=2)
+    pruner = ADMMPruner(model, config_list, trainer, traced_optimizer, criterion, iterations=10, training_epochs=1, granularity='coarse-grained')
     _, masks = pruner.compress()
     pruner.show_pruned_weights()
 
-    # Fine-grained method does not need to speedup
+    pruner._unwrap_model()
+    ModelSpeedup(model, torch.randn([128, 3, 32, 32]).to(device), masks).speedup_model()
+
     print('\n' + '=' * 50 + ' EVALUATE THE MODEL AFTER PRUNING ' + '=' * 50)
     evaluator(model)
 

examples/model_compress/pruning/v2/amc_pruning_torch.py → examples/model_compress/pruning/amc_pruning_torch.py

@@ -5,11 +5,11 @@ import torch
 from torchvision import datasets, transforms
 from torch.optim.lr_scheduler import MultiStepLR
 
-from nni.algorithms.compression.v2.pytorch.pruning import AMCPruner
-from nni.compression.pytorch.utils.counter import count_flops_params
+from nni.compression.pytorch.pruning import AMCPruner
+from nni.compression.pytorch.utils import count_flops_params
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
 from cifar10.vgg import VGG
 
 

examples/model_compress/pruning/v2/auto_compress_pruner.py → examples/model_compress/pruning/auto_compress_pruner.py

@@ -5,10 +5,10 @@ import torch
 from torchvision import datasets, transforms
 
 import nni
-from nni.algorithms.compression.v2.pytorch.pruning import AutoCompressPruner
+from nni.compression.pytorch.pruning import AutoCompressPruner
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
 from cifar10.vgg import VGG
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

examples/model_compress/pruning/v2/fpgm_pruning_torch.py → examples/model_compress/pruning/fpgm_pruning_torch.py

@@ -4,7 +4,7 @@
 '''
 NNI example for supported fpgm pruning algorithms.
 In this example, we show the end-to-end pruning process: pre-training -> pruning -> fine-tuning.
-Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speed up is required.
+Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speedup is required.
 
 '''
 import argparse
@@ -15,11 +15,11 @@ from torchvision import datasets, transforms
 from torch.optim.lr_scheduler import MultiStepLR
 
 from nni.compression.pytorch import ModelSpeedup
-from nni.compression.pytorch.utils.counter import count_flops_params
-from nni.algorithms.compression.v2.pytorch.pruning.basic_pruner import FPGMPruner
+from nni.compression.pytorch.utils import count_flops_params
+from nni.compression.pytorch.pruning import FPGMPruner
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
 from cifar10.vgg import VGG
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

examples/model_compress/pruning/v2/iterative_pruning_torch.py → examples/model_compress/pruning/iterative_pruning_torch.py

@@ -13,14 +13,14 @@ from tqdm import tqdm
 import torch
 from torchvision import datasets, transforms
 
-from nni.algorithms.compression.v2.pytorch.pruning import (
+from nni.compression.pytorch.pruning import (
     LinearPruner,
     AGPPruner,
     LotteryTicketPruner
 )
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
 from cifar10.vgg import VGG
 
 
@@ -94,8 +94,8 @@ if __name__ == '__main__':
                         choices=['level', 'l1', 'l2', 'fpgm', 'slim', 'apoz',
                                  'mean_activation', 'taylorfo', 'admm'],
                         help='algorithm to evaluate weights to prune')
-    parser.add_argument('--speed-up', type=bool, default=False,
-                        help='Whether to speed-up the pruned model')
+    parser.add_argument('--speedup', type=bool, default=False,
+                        help='Whether to speedup the pruned model')
     parser.add_argument('--reset-weight', type=bool, default=True,
                         help='Whether to reset weight during each iteration')
 
@@ -120,8 +120,8 @@ if __name__ == '__main__':
                'evaluator': None,
                'finetuner': finetuner}
 
-    if args.speed_up:
-        kw_args['speed_up'] = args.speed_up
+    if args.speedup:
+        kw_args['speedup'] = args.speedup
         kw_args['dummy_input'] = torch.rand(10, 3, 32, 32).to(device)
 
     if args.pruner == 'linear':

examples/model_compress/pruning/amc/amc_train.py → examples/model_compress/pruning/legacy/amc/amc_train.py

@@ -23,7 +23,7 @@ from data import get_dataset
 from utils import AverageMeter, accuracy, progress_bar
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[3] / 'models'))
 from mobilenet import MobileNet
 from mobilenet_v2 import MobileNetV2
 

examples/model_compress/pruning/auto_pruners_torch.py → examples/model_compress/pruning/legacy/auto_pruners_torch.py

@@ -17,10 +17,10 @@ from torchvision import datasets, transforms
 from nni.algorithms.compression.pytorch.pruning import L1FilterPruner, L2FilterPruner, FPGMPruner
 from nni.algorithms.compression.pytorch.pruning import SimulatedAnnealingPruner, ADMMPruner, NetAdaptPruner, AutoCompressPruner
 from nni.compression.pytorch import ModelSpeedup
-from nni.compression.pytorch.utils.counter import count_flops_params
+from nni.compression.pytorch.utils import count_flops_params
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
 from mnist.lenet import LeNet
 from cifar10.vgg import VGG
 from cifar10.resnet import ResNet18, ResNet50
@@ -292,8 +292,8 @@ def main(args):
             os.path.join(args.experiment_data_dir, 'model_masked.pth'), os.path.join(args.experiment_data_dir, 'mask.pth'))
         print('Masked model saved to %s' % args.experiment_data_dir)
 
-    # model speed up
-    if args.speed_up:
+    # model speedup
+    if args.speedup:
         if args.pruner != 'AutoCompressPruner':
             if args.model == 'LeNet':
                 model = LeNet().to(device)
@@ -310,11 +310,11 @@ def main(args):
             m_speedup = ModelSpeedup(model, dummy_input, masks_file, device)
             m_speedup.speedup_model()
             evaluation_result = evaluator(model)
-            print('Evaluation result (speed up model): %s' % evaluation_result)
+            print('Evaluation result (speedup model): %s' % evaluation_result)
             result['performance']['speedup'] = evaluation_result
 
-            torch.save(model.state_dict(), os.path.join(args.experiment_data_dir, 'model_speed_up.pth'))
-            print('Speed up model saved to %s' % args.experiment_data_dir)
+            torch.save(model.state_dict(), os.path.join(args.experiment_data_dir, 'model_speedup.pth'))
+            print('Speedup model saved to %s' % args.experiment_data_dir)
         flops, params, _ = count_flops_params(model, get_input_size(args.dataset))
         result['flops']['speedup'] = flops
         result['params']['speedup'] = params
@@ -402,9 +402,9 @@ if __name__ == '__main__':
     parser.add_argument('--sparsity-per-iteration', type=float, default=0.05,
                         help='sparsity_per_iteration of NetAdaptPruner')
 
-    # speed-up
-    parser.add_argument('--speed-up', type=str2bool, default=False,
-                        help='Whether to speed-up the pruned model')
+    # speedup
+    parser.add_argument('--speedup', type=str2bool, default=False,
+                        help='Whether to speedup the pruned model')
 
     # others
     parser.add_argument('--log-interval', type=int, default=200,

examples/model_compress/pruning/basic_pruners_torch.py → examples/model_compress/pruning/legacy/basic_pruners_torch.py

@@ -4,7 +4,7 @@
 '''
 NNI example for supported basic pruning algorithms.
 In this example, we show the end-to-end pruning process: pre-training -> pruning -> fine-tuning.
-Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speed up is required.
+Note that pruners use masks to simulate the real pruning. In order to obtain a real compressed model, model speedup is required.
 You can also try auto_pruners_torch.py to see the usage of some automatic pruning algorithms.
 
 '''
@@ -18,12 +18,12 @@ from torch.optim.lr_scheduler import StepLR, MultiStepLR
 from torchvision import datasets, transforms
 
 from pathlib import Path
-sys.path.append(str(Path(__file__).absolute().parents[1] / 'models'))
+sys.path.append(str(Path(__file__).absolute().parents[2] / 'models'))
 from mnist.lenet import LeNet
 from cifar10.vgg import VGG
 from cifar10.resnet import ResNet18
 
-from nni.compression.pytorch.utils.counter import count_flops_params
+from nni.compression.pytorch.utils import count_flops_params
 
 import nni
 from nni.compression.pytorch import ModelSpeedup
@@ -292,7 +292,7 @@ def main(args):
     if args.test_only:
         test(args, model, device, criterion, test_loader)
 
-    if args.speed_up:
+    if args.speedup:
         # Unwrap all modules to normal state
         pruner._unwrap_model()
         m_speedup = ModelSpeedup(model, dummy_input, mask_path, device)
@@ -356,7 +356,7 @@ if __name__ == '__main__':
     parser.add_argument('--sparsity', type=float, default=0.5,
                         help='target overall target sparsity')
     parser.add_argument('--dependency-aware', action='store_true', default=False,
-                        help='toggle dependency aware mode')
+                        help='toggle dependency-aware mode')
     parser.add_argument('--global-sort', action='store_true', default=False,
                         help='toggle global sort mode')
     parser.add_argument('--pruner', type=str, default='l1filter',
@@ -364,9 +364,9 @@ if __name__ == '__main__':
                                  'fpgm', 'mean_activation', 'apoz', 'taylorfo'],
                         help='pruner to use')
 
-    # speed-up
-    parser.add_argument('--speed-up', action='store_true', default=False,
-                        help='Whether to speed-up the pruned model')
+    # speedup
+    parser.add_argument('--speedup', action='store_true', default=False,
+                        help='Whether to speedup the pruned model')
 
     # fine-tuning
     parser.add_argument('--fine-tune-epochs', type=int, default=160,