!8775 remove "is_grad" in SoftmaxCrossEntropyWithLogits and correct its comments

From: @wanyiming
Reviewed-by: @kingxian,@zh_qh
Signed-off-by: @kingxian

mindspore/nn/probability/README.md

@@ -100,7 +100,7 @@ The loss function `SoftmaxCrossEntropyWithLogits` and the optimizer `AdamWeightD
 if __name__ == "__main__":
     ...
     # define the loss function
-    criterion = SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True, reduction="mean")
+    criterion = SoftmaxCrossEntropyWithLogits(sparse=True, reduction="mean")
     optimizer = AdamWeightDecay(params=network.trainable_params(), learning_rate=0.0001)
     ...
 ```
@@ -320,7 +320,7 @@ from mindspore.nn import WithLossCell, TrainOneStepCell
 if __name__ == "__main__":
     network = LeNet5()
 
-    criterion = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True, reduction="mean")
+    criterion = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction="mean")
     optimizer = nn.AdamWeightDecay(params=network.trainable_params(), learning_rate=0.0001)
 
     net_with_loss = WithLossCell(network, criterion)

mindspore/nn/probability/transforms/transform_bnn.py

@@ -53,7 +53,7 @@ class TransformToBNN:
         >>>         return out
         >>>
         >>> net = Net()
-        >>> criterion = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+        >>> criterion = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
         >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
         >>> net_with_loss = WithLossCell(network, criterion)
         >>> train_network = TrainOneStepCell(net_with_loss, optim)
@@ -107,7 +107,7 @@ class TransformToBNN:
 
         Examples:
             >>> net = Net()
-            >>> criterion = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> criterion = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
             >>> net_with_loss = WithLossCell(network, criterion)
             >>> train_network = TrainOneStepCell(net_with_loss, optim)
@@ -149,7 +149,7 @@ class TransformToBNN:
 
         Examples:
             >>> net = Net()
-            >>> criterion = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> criterion = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
             >>> net_with_loss = WithLossCell(network, criterion)
             >>> train_network = TrainOneStepCell(net_with_loss, optim)

model_zoo/official/cv/googlenet/README.md

@@ -425,8 +425,7 @@ If you need to use the trained model to perform inference on multiple hardware p
   net = GoogleNet(num_classes=cfg.num_classes)
   opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 0.01,
                  cfg.momentum, weight_decay=cfg.weight_decay)
-  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean', 
-                                          is_grad=False)
+  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
   model = Model(net, loss_fn=loss, optimizer=opt, metrics={'acc'})
   
   # Load pre-trained model
@@ -452,8 +451,7 @@ If you need to use the trained model to perform inference on multiple hardware p
   net = GoogleNet(num_classes=cfg.num_classes)
   opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 0.01,
                  cfg.momentum, weight_decay=cfg.weight_decay)
-  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean', 
-                                          is_grad=False)
+  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
   model = Model(net, loss_fn=loss, optimizer=opt, metrics={'acc'})
   
   # Load pre-trained model
@@ -486,7 +484,7 @@ If you need to use the trained model to perform inference on multiple hardware p
                 steps_per_epoch=batch_num)
   opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 
                  Tensor(lr), cfg.momentum, weight_decay=cfg.weight_decay)
-  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean', is_grad=False)
+  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
   model = Model(net, loss_fn=loss, optimizer=opt, metrics={'acc'},
                 amp_level="O2", keep_batchnorm_fp32=False, loss_scale_manager=None)
   
@@ -520,7 +518,7 @@ If you need to use the trained model to perform inference on multiple hardware p
                 steps_per_epoch=batch_num)
   opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 
                  Tensor(lr), cfg.momentum, weight_decay=cfg.weight_decay)
-  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean', is_grad=False)
+  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
   model = Model(net, loss_fn=loss, optimizer=opt, metrics={'acc'},
                 amp_level="O2", keep_batchnorm_fp32=False, loss_scale_manager=None)
   

model_zoo/official/nlp/bert_thor/src/model_thor.py

@@ -135,7 +135,7 @@ class Model:
         >>>         return out
         >>>
         >>> net = Net()
-        >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+        >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
         >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
         >>> model = Model(net, loss_fn=loss, optimizer=optim, metrics=None)
         >>> dataset = get_dataset()
@@ -307,7 +307,7 @@ class Model:
             >>> train_dataset = get_train_dataset()
             >>> valid_dataset = get_valid_dataset()
             >>> net = Net()
-            >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
             >>> model = Model(net, loss_fn=loss, optimizer=optim, metrics={'acc'})
             >>> model.init(train_dataset, valid_dataset)
@@ -597,7 +597,7 @@ class Model:
         Examples:
             >>> dataset = get_dataset()
             >>> net = Net()
-            >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> loss_scale_manager = FixedLossScaleManager()
             >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
             >>> model = Model(net, loss_fn=loss, optimizer=optim, metrics=None, loss_scale_manager=loss_scale_manager)
@@ -714,7 +714,7 @@ class Model:
         Examples:
             >>> dataset = get_dataset()
             >>> net = Net()
-            >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> model = Model(net, loss_fn=loss, optimizer=None, metrics={'acc'})
             >>> model.eval(dataset)
         """

model_zoo/official/recommend/ncf/README.md

@@ -243,8 +243,7 @@ https://www.mindspore.cn/tutorial/zh-CN/master/use/multi_platform_inference.html
   net = GoogleNet(num_classes=cfg.num_classes)
   opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 0.01,
                  cfg.momentum, weight_decay=cfg.weight_decay)
-  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean', 
-                                          is_grad=False)
+  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
   model = Model(net, loss_fn=loss, optimizer=opt, metrics={'acc'})
   
   # Load pre-trained model
@@ -275,7 +274,7 @@ https://www.mindspore.cn/tutorial/zh-CN/master/use/multi_platform_inference.html
                 steps_per_epoch=batch_num)
   opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 
                  Tensor(lr), cfg.momentum, weight_decay=cfg.weight_decay)
-  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean', is_grad=False)
+  loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
   model = Model(net, loss_fn=loss, optimizer=opt, metrics={'acc'},
                 amp_level="O2", keep_batchnorm_fp32=False, loss_scale_manager=None)
   

model_zoo/official/recommend/ncf/src/ncf.py

@@ -199,7 +199,7 @@ class NetWithLossClass(nn.Cell):
     """
     def __init__(self, network):
         super(NetWithLossClass, self).__init__(auto_prefix=False)
-        #self.loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+        #self.loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
         self.loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
         self.network = network
         self.reducesum = P.ReduceSum(keep_dims=False)

model_zoo/research/cv/ghostnet/eval.py

@@ -50,8 +50,7 @@ if __name__ == '__main__':
     else:
         raise ValueError("Unsupport platform.")
 
-    loss = nn.SoftmaxCrossEntropyWithLogits(
-        is_grad=False, sparse=True, reduction='mean')
+    loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
 
     if args_opt.model == 'ghostnet':
         net = ghostnet_1x(num_classes=config_platform.num_classes)

model_zoo/research/cv/ghostnet_quant/eval.py

@@ -49,8 +49,7 @@ if __name__ == '__main__':
     else:
         raise ValueError("Unsupport platform.")
 
-    loss = nn.SoftmaxCrossEntropyWithLogits(
-        is_grad=False, sparse=True, reduction='mean')
+    loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
 
     net = ghostnet_1x(num_classes=config_platform.num_classes)
 

model_zoo/research/cv/resnet50_adv_pruning/eval.py

@@ -53,8 +53,7 @@ if __name__ == '__main__':
     else:
         raise ValueError("Unsupport platform.")
 
-    loss = nn.SoftmaxCrossEntropyWithLogits(
-        is_grad=False, sparse=True, reduction='mean')
+    loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
 
     if args_opt.platform == "Ascend":
         net.to_float(mstype.float16)

tests/st/networks/models/resnet50/src_thor/model_thor.py

@@ -128,7 +128,7 @@ class Model:
         >>>         return out
         >>>
         >>> net = Net()
-        >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+        >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
         >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
         >>> model = Model(net, loss_fn=loss, optimizer=optim, metrics=None)
         >>> dataset = get_dataset()
@@ -295,7 +295,7 @@ class Model:
             >>> train_dataset = get_train_dataset()
             >>> valid_dataset = get_valid_dataset()
             >>> net = Net()
-            >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
             >>> model = Model(net, loss_fn=loss, optimizer=optim, metrics={'acc'})
             >>> model.init(train_dataset, valid_dataset)
@@ -566,7 +566,7 @@ class Model:
         Examples:
             >>> dataset = get_dataset()
             >>> net = Net()
-            >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> loss_scale_manager = FixedLossScaleManager()
             >>> optim = Momentum(params=net.trainable_params(), learning_rate=0.1, momentum=0.9)
             >>> model = Model(net, loss_fn=loss, optimizer=optim, metrics=None, loss_scale_manager=loss_scale_manager)
@@ -678,7 +678,7 @@ class Model:
         Examples:
             >>> dataset = get_dataset()
             >>> net = Net()
-            >>> loss = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=True)
+            >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
             >>> model = Model(net, loss_fn=loss, optimizer=None, metrics={'acc'})
             >>> model.eval(dataset)
         """