Fix wizard template module to fit new mindspore operator API

mindinsight/wizard/conf/templates/network/alexnet/dataset/cifar10/dataset.py-tpl

@@ -19,7 +19,7 @@ Produce the dataset
 import os
 import mindspore.dataset as ds
 import mindspore.dataset.transforms.c_transforms as C
-import mindspore.dataset.transforms.vision.c_transforms as CV
+import mindspore.dataset.vision.c_transforms as CV
 from mindspore.common import dtype as mstype
 from .config import cfg
 from mindspore.communication.management import init, get_rank, get_group_size
@@ -53,14 +53,14 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
         random_horizontal_op = CV.RandomHorizontalFlip()
     channel_swap_op = CV.HWC2CHW()
     typecast_op = C.TypeCast(mstype.int32)
-    cifar_ds = cifar_ds.map(input_columns="label", operations=typecast_op)
+    cifar_ds = cifar_ds.map(operations=typecast_op, input_columns="label")
     if do_train:
-        cifar_ds = cifar_ds.map(input_columns="image", operations=random_crop_op)
-        cifar_ds = cifar_ds.map(input_columns="image", operations=random_horizontal_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=resize_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=rescale_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=normalize_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=channel_swap_op)
+        cifar_ds = cifar_ds.map(operations=random_crop_op, input_columns="image")
+        cifar_ds = cifar_ds.map(operations=random_horizontal_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=resize_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=rescale_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=normalize_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=channel_swap_op, input_columns="image")
 
     cifar_ds = cifar_ds.shuffle(buffer_size=1000)
     cifar_ds = cifar_ds.batch(batch_size, drop_remainder=True)

mindinsight/wizard/conf/templates/network/alexnet/dataset/imagenet/dataset.py-tpl

@@ -18,7 +18,7 @@ create train or eval dataset.
 import os
 import mindspore.common.dtype as mstype
 import mindspore.dataset.engine as de
-import mindspore.dataset.transforms.vision.c_transforms as C
+import mindspore.dataset.vision.c_transforms as C
 import mindspore.dataset.transforms.c_transforms as C2
 from mindspore.communication.management import init, get_rank, get_group_size
 
@@ -50,10 +50,10 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
         device_num = 1
 
     if device_num == 1:
-        ds = de.ImageFolderDatasetV2(data_path, num_parallel_workers=8, shuffle=True)
+        ds = de.ImageFolderDataset(data_path, num_parallel_workers=8, shuffle=True)
     else:
-        ds = de.ImageFolderDatasetV2(data_path, num_parallel_workers=8, shuffle=True,
-                                     num_shards=device_num, shard_id=rank_id)
+        ds = de.ImageFolderDataset(data_path, num_parallel_workers=8, shuffle=True,
+                                   num_shards=device_num, shard_id=rank_id)
 
     image_size = cfg.image_height
     mean = [0.485 * 255, 0.456 * 255, 0.406 * 255]
@@ -78,8 +78,8 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
 
     type_cast_op = C2.TypeCast(mstype.int32)
 
-    ds = ds.map(input_columns="image", num_parallel_workers=8, operations=trans)
-    ds = ds.map(input_columns="label", num_parallel_workers=8, operations=type_cast_op)
+    ds = ds.map(operations=trans, input_columns="image", num_parallel_workers=8)
+    ds = ds.map(operations=type_cast_op, input_columns="label", num_parallel_workers=8)
 
     # apply batch operations
     ds = ds.batch(batch_size, drop_remainder=True)

mindinsight/wizard/conf/templates/network/alexnet/train.py-tpl

@@ -88,13 +88,13 @@ if __name__ == "__main__":
     else:
         for _, cell in net.cells_and_names():
             if isinstance(cell, nn.Conv2d):
-                cell.weight.default_input = weight_init.initializer(weight_init.XavierUniform(),
-                                                                    cell.weight.default_input.shape,
-                                                                    cell.weight.default_input.dtype).to_tensor()
+                cell.weight.set_data(weight_init.initializer(weight_init.XavierUniform(),
+                                                             cell.weight.shape,
+                                                             cell.weight.dtype))
             if isinstance(cell, nn.Dense):
-                cell.weight.default_input = weight_init.initializer(weight_init.TruncatedNormal(),
-                                                                    cell.weight.default_input.shape,
-                                                                    cell.weight.default_input.dtype).to_tensor()
+                cell.weight.set_data(weight_init.initializer(weight_init.TruncatedNormal(),
+                                                             cell.weight.shape,
+                                                             cell.weight.dtype))
     # define learning rate
     lr = Tensor(get_lr(0, cfg.lr, cfg.epoch_size, ds_train.get_dataset_size()))
 

mindinsight/wizard/conf/templates/network/lenet/dataset/mnist/dataset.py-tpl

@@ -18,9 +18,9 @@ Produce the dataset
 import os
 
 import mindspore.dataset as ds
-import mindspore.dataset.transforms.vision.c_transforms as CV
+import mindspore.dataset.vision.c_transforms as CV
 import mindspore.dataset.transforms.c_transforms as C
-from mindspore.dataset.transforms.vision import Inter
+from mindspore.dataset.vision import Inter
 from mindspore.common import dtype as mstype
 from mindspore.communication.management import init, get_rank, get_group_size
 
@@ -67,11 +67,11 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
     type_cast_op = C.TypeCast(mstype.int32)
 
     # apply map operations on images
-    mnist_ds = mnist_ds.map(input_columns="label", operations=type_cast_op)
-    mnist_ds = mnist_ds.map(input_columns="image", operations=resize_op)
-    mnist_ds = mnist_ds.map(input_columns="image", operations=rescale_op)
-    mnist_ds = mnist_ds.map(input_columns="image", operations=rescale_nml_op)
-    mnist_ds = mnist_ds.map(input_columns="image", operations=hwc2chw_op)
+    mnist_ds = mnist_ds.map(operations=type_cast_op, input_columns="label")
+    mnist_ds = mnist_ds.map(operations=resize_op, input_columns="image")
+    mnist_ds = mnist_ds.map(operations=rescale_op, input_columns="image")
+    mnist_ds = mnist_ds.map(operations=rescale_nml_op, input_columns="image")
+    mnist_ds = mnist_ds.map(operations=hwc2chw_op, input_columns="image")
 
     # apply DatasetOps
     buffer_size = 10000

mindinsight/wizard/conf/templates/network/resnet50/dataset/cifar10/dataset.py-tpl

@@ -19,7 +19,7 @@ import os
 
 import mindspore.dataset as ds
 import mindspore.dataset.transforms.c_transforms as C
-import mindspore.dataset.transforms.vision.c_transforms as CV
+import mindspore.dataset.vision.c_transforms as CV
 from mindspore.common import dtype as mstype
 from mindspore.communication.management import init, get_rank, get_group_size
 
@@ -46,8 +46,6 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
 
     rescale = 1.0 / 255.0
     shift = 0.0
-    rescale = 1.0 / 255.0
-    shift = 0.0
 
     resize_op = CV.Resize((cfg.image_height, cfg.image_width))
     rescale_op = CV.Rescale(rescale, shift)
@@ -57,14 +55,14 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
         random_horizontal_op = CV.RandomHorizontalFlip()
     channel_swap_op = CV.HWC2CHW()
     typecast_op = C.TypeCast(mstype.int32)
-    cifar_ds = cifar_ds.map(input_columns="label", operations=typecast_op)
+    cifar_ds = cifar_ds.map(operations=typecast_op, input_columns="label")
     if do_train:
-        cifar_ds = cifar_ds.map(input_columns="image", operations=random_crop_op)
-        cifar_ds = cifar_ds.map(input_columns="image", operations=random_horizontal_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=resize_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=rescale_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=normalize_op)
-    cifar_ds = cifar_ds.map(input_columns="image", operations=channel_swap_op)
+        cifar_ds = cifar_ds.map(operations=random_crop_op, input_columns="image")
+        cifar_ds = cifar_ds.map(operations=random_horizontal_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=resize_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=rescale_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=normalize_op, input_columns="image")
+    cifar_ds = cifar_ds.map(operations=channel_swap_op, input_columns="image")
 
     cifar_ds = cifar_ds.shuffle(buffer_size=1000)
     cifar_ds = cifar_ds.batch(batch_size, drop_remainder=True)

mindinsight/wizard/conf/templates/network/resnet50/dataset/imagenet/dataset.py-tpl

@@ -18,7 +18,7 @@ create train or eval dataset.
 import os
 import mindspore.common.dtype as mstype
 import mindspore.dataset.engine as de
-import mindspore.dataset.transforms.vision.c_transforms as C
+import mindspore.dataset.vision.c_transforms as C
 import mindspore.dataset.transforms.c_transforms as C2
 from mindspore.communication.management import init, get_rank, get_group_size
 
@@ -50,10 +50,10 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
         device_num = 1
 
     if device_num == 1:
-        ds = de.ImageFolderDatasetV2(data_path, num_parallel_workers=8, shuffle=True)
+        ds = de.ImageFolderDataset(data_path, num_parallel_workers=8, shuffle=True)
     else:
-        ds = de.ImageFolderDatasetV2(data_path, num_parallel_workers=8, shuffle=True,
-                                     num_shards=device_num, shard_id=rank_id)
+        ds = de.ImageFolderDataset(data_path, num_parallel_workers=8, shuffle=True,
+                                   num_shards=device_num, shard_id=rank_id)
 
     image_size = cfg.image_height
     mean = [0.485 * 255, 0.456 * 255, 0.406 * 255]
@@ -78,8 +78,8 @@ def create_dataset(data_path, batch_size=32, repeat_size=1, do_train=True, targe
 
     type_cast_op = C2.TypeCast(mstype.int32)
 
-    ds = ds.map(input_columns="image", num_parallel_workers=8, operations=trans)
-    ds = ds.map(input_columns="label", num_parallel_workers=8, operations=type_cast_op)
+    ds = ds.map(operations=trans, input_columns="image", num_parallel_workers=8)
+    ds = ds.map(operations=type_cast_op, input_columns="label", num_parallel_workers=8)
 
     # apply batch operations
     ds = ds.batch(batch_size, drop_remainder=True)

mindinsight/wizard/conf/templates/network/resnet50/train.py-tpl

@@ -91,13 +91,13 @@ if __name__ == '__main__':
     else:
         for _, cell in net.cells_and_names():
             if isinstance(cell, nn.Conv2d):
-                cell.weight.default_input = weight_init.initializer(weight_init.XavierUniform(),
-                                                                    cell.weight.default_input.shape,
-                                                                    cell.weight.default_input.dtype).to_tensor()
+                cell.weight.set_data(weight_init.initializer(weight_init.XavierUniform(),
+                                                             cell.weight.shape,
+                                                             cell.weight.dtype))
             if isinstance(cell, nn.Dense):
-                cell.weight.default_input = weight_init.initializer(weight_init.TruncatedNormal(),
-                                                                    cell.weight.default_input.shape,
-                                                                    cell.weight.default_input.dtype).to_tensor()
+                cell.weight.set_data(weight_init.initializer(weight_init.TruncatedNormal(),
+                                                             cell.weight.shape,
+                                                             cell.weight.dtype))
 
     # init lr
 

tests/st/func/wizard/test_alexnet.py

@@ -26,7 +26,7 @@ NETWORK_NAME = 'alexnet'
 
 
 class TestAlexNet:
-    """Test AlexNet Module"""
+    """Test AlexNet Module."""
 
     @pytest.mark.level0
     @pytest.mark.env_single
@@ -68,35 +68,35 @@ class TestAlexNet:
         'config': {'loss': 'SoftmaxCrossEntropyWithLogits',
                    'optimizer': 'Momentum',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyWithLogits',
                    'optimizer': 'Adam',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyWithLogits',
                    'optimizer': 'SGD',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyExpand',
                    'optimizer': 'Momentum',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyExpand',
                    'optimizer': 'Adam',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyExpand',
                    'optimizer': 'SGD',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }])
     def test_combinations(self, params):
-        """Do testing"""
+        """Do testing."""
 
         network_maker_name = NETWORK_NAME
         config = params['config']
@@ -112,7 +112,7 @@ class TestAlexNet:
         self.check_train_eval_readme(config['dataset'], config['loss'], config['optimizer'])
 
     def check_src(self, dataset_name, config):
-        """Check src file"""
+        """Check src file."""
         dataset_is_right = False
         config_dataset_is_right = False
         config_optimizer_is_right = False
@@ -140,7 +140,7 @@ class TestAlexNet:
 
     @staticmethod
     def _check_config_dataset(config, content):
-        """Check dataset in config"""
+        """Check dataset in config."""
         config_dataset_is_right = False
         if config['dataset'] == 'Cifar10':
             if "'num_classes': 10" in content:
@@ -152,7 +152,7 @@ class TestAlexNet:
 
     @staticmethod
     def _check_config_optimizer(config, content):
-        """Check optimizer in config"""
+        """Check optimizer in config."""
         config_optimizer_is_right = False
         if config['optimizer'] == 'Momentum':
             if "'lr': 0.002" in content:
@@ -166,7 +166,7 @@ class TestAlexNet:
         return config_optimizer_is_right
 
     def check_train_eval_readme(self, dataset_name, loss_name, optimizer_name):
-        """Check train and eval"""
+        """Check train and eval."""
 
         train_is_right = False
         eval_is_right = False
@@ -191,7 +191,7 @@ class TestAlexNet:
         assert readme_is_right
 
     def check_scripts(self):
-        """Check scripts"""
+        """Check scripts."""
 
         exist_run_distribute_train = False
         exist_run_distribute_train_gpu = False

tests/st/func/wizard/test_lenet.py

@@ -25,7 +25,7 @@ NETWORK_NAME = 'lenet'
 
 
 class TestLeNet:
-    """Test LeNet Module"""
+    """Test LeNet Module."""
 
     @pytest.mark.level0
     @pytest.mark.env_single
@@ -65,7 +65,7 @@ class TestLeNet:
         'dataset_loader_name': 'MnistDataset'
     }])
     def test_combinations(self, params):
-        """Do testing"""
+        """Do testing."""
 
         network_maker_name = NETWORK_NAME
         config = params['config']
@@ -81,7 +81,7 @@ class TestLeNet:
         self.check_train_eval_readme(config['loss'], config['optimizer'])
 
     def check_src(self, dataset_name, config):
-        """Check src file"""
+        """Check src file."""
         dataset_is_right = False
         config_optimizer_is_right = False
         network_is_right = False
@@ -109,7 +109,7 @@ class TestLeNet:
         assert network_is_right
 
     def check_train_eval_readme(self, loss_name, optimizer_name):
-        """Check train and eval"""
+        """Check train and eval."""
 
         train_is_right = False
         eval_is_right = False
@@ -134,7 +134,7 @@ class TestLeNet:
         assert readme_is_right
 
     def check_scripts(self):
-        """Check scripts"""
+        """Check scripts."""
 
         exist_run_distribute_train = False
         exist_run_distribute_train_gpu = False

tests/st/func/wizard/test_resnet50.py

@@ -26,7 +26,7 @@ NETWORK_NAME = 'resnet50'
 
 
 class TestResNet50:
-    """Test ResNet50 Module"""
+    """Test ResNet50 Module."""
 
     @pytest.mark.level0
     @pytest.mark.env_single
@@ -68,35 +68,35 @@ class TestResNet50:
         'config': {'loss': 'SoftmaxCrossEntropyWithLogits',
                    'optimizer': 'Momentum',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyWithLogits',
                    'optimizer': 'Adam',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyWithLogits',
                    'optimizer': 'SGD',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyExpand',
                    'optimizer': 'Momentum',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyExpand',
                    'optimizer': 'Adam',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }, {
         'config': {'loss': 'SoftmaxCrossEntropyExpand',
                    'optimizer': 'SGD',
                    'dataset': 'ImageNet'},
-        'dataset_loader_name': 'ImageFolderDatasetV2'
+        'dataset_loader_name': 'ImageFolderDataset'
     }])
     def test_combinations(self, params):
-        """Do testing"""
+        """Do testing."""
 
         network_maker_name = NETWORK_NAME
         config = params['config']
@@ -112,7 +112,7 @@ class TestResNet50:
         self.check_train_eval_readme(config['dataset'], config['loss'], config['optimizer'])
 
     def check_src(self, dataset_name, config):
-        """Check src file"""
+        """Check src file."""
         dataset_is_right = False
         config_dataset_is_right = False
         config_optimizer_is_right = False
@@ -144,7 +144,7 @@ class TestResNet50:
 
     @staticmethod
     def _check_config_dataset(config, content):
-        """Check dataset in config"""
+        """Check dataset in config."""
         config_dataset_is_right = False
         if config['dataset'] == 'Cifar10':
             if "'num_classes': 10" in content \
@@ -160,7 +160,7 @@ class TestResNet50:
 
     @staticmethod
     def _check_config_optimizer(config, content):
-        """Check optimizer in config"""
+        """Check optimizer in config."""
         config_optimizer_is_right = False
         if config['optimizer'] == 'Momentum':
             if "'lr': 0.01" in content and \
@@ -175,7 +175,7 @@ class TestResNet50:
         return config_optimizer_is_right
 
     def check_train_eval_readme(self, dataset_name, loss_name, optimizer_name):
-        """Check train and eval"""
+        """Check train and eval."""
 
         train_is_right = False
         eval_is_right = False
@@ -208,7 +208,7 @@ class TestResNet50:
         assert readme_is_right
 
     def check_scripts(self):
-        """Check scripts"""
+        """Check scripts."""
 
         exist_run_distribute_train = False
         exist_run_distribute_train_gpu = False