!7741 [ME] format code

Merge pull request !7741 from chenzhongming/zomi_master

mindspore/_extends/graph_kernel/model/model.py

@@ -257,10 +257,10 @@ class Value:
         self.data_format = data_format
 
     def __str__(self):
-        return self.name + str(list(self.shape)) + str(self.value)
+        return self.name + str(list(self.shape))
 
     def __repr__(self):
-        return "%s.%s%s%s" % (self.name, self.dtype, str(list(self.shape)), str(self.value))
+        return "%s.%s%s" % (self.name, self.dtype, str(list(self.shape)))
 
     def get_size(self):
         return 1

mindspore/common/parameter.py

@@ -158,16 +158,10 @@ class Parameter(MetaTensor_):
         return (Tensor, data)
 
     def __str__(self):
-        value_str = MetaTensor.__str__(self)
-        if isinstance(self, Tensor):
-            value_str = Tensor.__str__(self)
-        return f'Parameter (name={self._param_info.name}, value={value_str})'
+        return f'Parameter (name={self._param_info.name})'
 
     def __repr__(self):
-        value_str = MetaTensor.__repr__(self)
-        if isinstance(self, Tensor):
-            value_str = Tensor.__repr__(self)
-        return f'Parameter (name={self._param_info.name}, value={value_str})'
+        return f'Parameter (name={self._param_info.name})'
 
     def __parameter__(self):
         """For parse check."""

mindspore/nn/layer/basic.py

@@ -121,8 +121,7 @@ class Dropout(Cell):
         return self.dropout_do_mask(x, output, keep_prob)
 
     def extend_repr(self):
-        str_info = 'keep_prob={}, dtype={}'.format(self.keep_prob, self.dtype)
-        return str_info
+        return 'keep_prob={}, dtype={}'.format(self.keep_prob, self.dtype)
 
 
 class Flatten(Cell):
@@ -365,8 +364,7 @@ class Norm(Cell):
         return x
 
     def extend_repr(self):
-        str_info = 'axis={}, keep_dims={}'.format(self.axis, self.keep_dims)
-        return str_info
+        return 'axis={}, keep_dims={}'.format(self.axis, self.keep_dims)
 
 
 class OneHot(Cell):

mindspore/nn/layer/normalization.py

@@ -547,9 +547,8 @@ class LayerNorm(Cell):
 
     def extend_repr(self):
         """Display instance object as string."""
-        s = 'normalized_shape={}, begin_norm_axis={}, begin_params_axis={}, gamma{}, beta={}'.format(
+        return 'normalized_shape={}, begin_norm_axis={}, begin_params_axis={}, gamma{}, beta={}'.format(
             self.normalized_shape, self.begin_norm_axis, self.begin_params_axis, self.gamma, self.beta)
-        return s
 
 
 class GroupNorm(Cell):
@@ -642,5 +641,4 @@ class GroupNorm(Cell):
 
     def extend_repr(self):
         """Display instance object as string."""
-        s = 'num_groups={}, num_channels={}'.format(self.num_groups, self.num_channels)
-        return s
+        return 'num_groups={}, num_channels={}'.format(self.num_groups, self.num_channels)

mindspore/nn/layer/quant.py

@@ -1008,14 +1008,13 @@ class DenseQuant(Cell):
 
     def extend_repr(self):
         """A pretty print for Dense layer."""
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}'.format(
+        s = 'in_channels={}, out_channels={}, weight={}, has_bias={}'.format(
             self.in_channels, self.out_channels, self.weight, self.has_bias)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
+            s += ', bias={}'.format(self.bias)
         if self.activation_flag:
-            str_info = str_info + ', activation={}'.format(self.activation)
-
-        return str_info
+            s += ', activation={}'.format(self.activation)
+        return s
 
 
 class _QuantActivation(Cell):
@@ -1387,13 +1386,13 @@ class QuantBlock(Cell):
         return x
 
     def extend_repr(self):
-        str_info = f'quant={self.quant}, core_op={type(self.core_op)}, weight=shape[{self.weight.shape}]'
+        s = f'quant={self.quant}, core_op={type(self.core_op)}, weight=shape[{self.weight.shape}]'
         if self.has_bias:
-            str_info = str_info + f', bias=shape[{self.bias.shape}]'
+            s += f', bias=shape[{self.bias.shape}]'
         if self.has_act:
-            str_info = str_info + f', activation={self.activation}'
-        str_info = str_info + f', dequant={self.dequant}'
-        return str_info
+            s += f', activation={self.activation}'
+        s += f', dequant={self.dequant}'
+        return s
 
 
 class QuantMindirBlock(Cell):
@@ -1454,9 +1453,9 @@ class QuantMindirBlock(Cell):
         return x
 
     def extend_repr(self):
-        str_info = f'core_op={type(self.core_op)}, weight=shape[{self.weight.shape}]'
+        s = f'core_op={type(self.core_op)}, weight=shape[{self.weight.shape}]'
         if self.has_bias:
-            str_info = str_info + f', bias=shape[{self.bias.shape}]'
+            s += f', bias=shape[{self.bias.shape}]'
         if self.has_act:
-            str_info = str_info + f', activation={self.activation}'
-        return str_info
+            s += f', activation={self.activation}'
+        return s

mindspore/nn/probability/bijector/gumbel_cdf.py

@@ -100,8 +100,7 @@ class GumbelCDF(Bijector):
         return self._parameter_type
 
     def extend_repr(self):
-        str_info = f'loc = {self.loc}, scale = {self.scale}'
-        return str_info
+        return f'loc = {self.loc}, scale = {self.scale}'
 
     def shape_mapping(self, shape):
         return shape

mindspore/nn/probability/bijector/power_transform.py

@@ -81,8 +81,7 @@ class PowerTransform(Bijector):
         return self._power
 
     def extend_repr(self):
-        str_info = f'power = {self.power}'
-        return str_info
+        return f'power = {self.power}'
 
     def shape_mapping(self, shape):
         return shape

mindspore/nn/probability/bijector/scalar_affine.py

@@ -90,8 +90,7 @@ class ScalarAffine(Bijector):
         return self._shift
 
     def extend_repr(self):
-        str_info = f'scale = {self.scale}, shift = {self.shift}'
-        return str_info
+        return f'scale = {self.scale}, shift = {self.shift}'
 
     def shape_mapping(self, shape):
         return shape

mindspore/nn/probability/bijector/softplus.py

@@ -118,8 +118,7 @@ class Softplus(Bijector):
         return self._sharpness
 
     def extend_repr(self):
-        str_info = f'sharpness = {self.sharpness}'
-        return str_info
+        return f'sharpness = {self.sharpness}'
 
     def shape_mapping(self, shape):
         return shape

mindspore/nn/probability/bnn_layers/conv_variational.py

@@ -141,15 +141,15 @@ class _ConvVariational(_Conv):
         return outputs
 
     def extend_repr(self):
-        str_info = 'in_channels={}, out_channels={}, kernel_size={}, stride={},  pad_mode={}, ' \
-                    'padding={}, dilation={}, group={}, weight_mean={}, weight_std={}, has_bias={}'\
+        s = 'in_channels={}, out_channels={}, kernel_size={}, stride={},  pad_mode={}, ' \
+            'padding={}, dilation={}, group={}, weight_mean={}, weight_std={}, has_bias={}'\
             .format(self.in_channels, self.out_channels, self.kernel_size, self.stride, self.pad_mode, self.padding,
                     self.dilation, self.group, self.weight_posterior.mean, self.weight_posterior.untransformed_std,
                     self.has_bias)
         if self.has_bias:
-            str_info = str_info + ', bias_mean={}, bias_std={}'\
+            s += ', bias_mean={}, bias_std={}'\
                 .format(self.bias_posterior.mean, self.bias_posterior.untransformed_std)
-        return str_info
+        return s
 
     def _apply_variational_bias(self, inputs):
         bias_posterior_tensor = self.bias_posterior("sample")

mindspore/nn/probability/bnn_layers/dense_variational.py

@@ -107,16 +107,15 @@ class _DenseVariational(Cell):
         return outputs
 
     def extend_repr(self):
-        str_info = 'in_channels={}, out_channels={}, weight_mean={}, weight_std={}, has_bias={}' \
+        s = 'in_channels={}, out_channels={}, weight_mean={}, weight_std={}, has_bias={}' \
             .format(self.in_channels, self.out_channels, self.weight_posterior.mean,
                     self.weight_posterior.untransformed_std, self.has_bias)
         if self.has_bias:
-            str_info = str_info + ', bias_mean={}, bias_std={}' \
+            s += ', bias_mean={}, bias_std={}' \
                 .format(self.bias_posterior.mean, self.bias_posterior.untransformed_std)
-
         if self.activation_flag:
-            str_info = str_info + ', activation={}'.format(self.activation)
-        return str_info
+            s += ', activation={}'.format(self.activation)
+        return s
 
     def _apply_variational_bias(self, inputs):
         bias_posterior_tensor = self.bias_posterior("sample")

mindspore/nn/probability/distribution/bernoulli.py

@@ -141,10 +141,10 @@ class Bernoulli(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'probs = {self.probs}'
+            s = f'probs = {self.probs}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     @property
     def probs(self):

mindspore/nn/probability/distribution/categorical.py

@@ -157,10 +157,10 @@ class Categorical(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'probs = {self.probs}'
+            s = f'probs = {self.probs}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     @property
     def probs(self):

mindspore/nn/probability/distribution/exponential.py

@@ -145,10 +145,10 @@ class Exponential(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'rate = {self.rate}'
+            s = f'rate = {self.rate}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     @property
     def rate(self):

mindspore/nn/probability/distribution/geometric.py

@@ -150,10 +150,10 @@ class Geometric(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'probs = {self.probs}'
+            s = f'probs = {self.probs}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     @property
     def probs(self):

mindspore/nn/probability/distribution/log_normal.py

@@ -163,10 +163,10 @@ class LogNormal(msd.TransformedDistribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'loc = {self._mean_value}, scale = {self._sd_value}'
+            s = f'loc = {self._mean_value}, scale = {self._sd_value}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     def _mean(self, loc=None, scale=None):
         """

mindspore/nn/probability/distribution/logistic.py

@@ -156,10 +156,10 @@ class Logistic(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'location = {self._loc}, scale = {self._scale}'
+            s = f'location = {self._loc}, scale = {self._scale}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     @property
     def loc(self):

mindspore/nn/probability/distribution/normal.py

@@ -149,10 +149,10 @@ class Normal(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'mean = {self._mean_value}, standard deviation = {self._sd_value}'
+            s = f'mean = {self._mean_value}, standard deviation = {self._sd_value}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     def _mean(self, mean=None, sd=None):
         """

mindspore/nn/probability/distribution/uniform.py

@@ -154,10 +154,10 @@ class Uniform(Distribution):
 
     def extend_repr(self):
         if self.is_scalar_batch:
-            str_info = f'low = {self.low}, high = {self.high}'
+            s = f'low = {self.low}, high = {self.high}'
         else:
-            str_info = f'batch_shape = {self._broadcast_shape}'
-        return str_info
+            s = f'batch_shape = {self._broadcast_shape}'
+        return s
 
     @property
     def low(self):

model_zoo/official/cv/resnet_thor/src/thor_layer.py

@@ -392,15 +392,12 @@ class Dense_Thor_GPU(Cell):
 
     def extend_repr(self):
         """extend_repr"""
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}' \
-            .format(self.in_channels, self.out_channels, self.weight, self.has_bias)
+        s = 'in_channels={}, out_channels={}'.format(self.in_channels, self.out_channels)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
-
+            s += ', has_bias={}'.format(self.has_bias)
         if self.activation_flag:
-            str_info = str_info + ', activation={}'.format(self.activation)
-
-        return str_info
+            s += ', activation={}'.format(self.activation)
+        return s
 
 
 class Conv2d_Thor(_Conv):
@@ -775,12 +772,9 @@ class Dense_Thor(Cell):
 
     def extend_repr(self):
         """extend_repr"""
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}' \
-            .format(self.in_channels, self.out_channels, self.weight, self.has_bias)
+        s = 'in_channels={}, out_channels={}'.format(self.in_channels, self.out_channels)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
-
+            s += ', has_bias={}'.format(self.has_bias)
         if self.activation_flag:
-            str_info = str_info + ', activation={}'.format(self.activation)
-
-        return str_info
+            s += ', activation={}'.format(self.activation)
+        return s

model_zoo/official/gnn/gat/src/gat.py

@@ -102,12 +102,10 @@ class GNNFeatureTransform(nn.Cell):
         return output
 
     def extend_repr(self):
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}' \
-                .format(self.in_channels, self.out_channels, self.weight, self.has_bias)
+        s = 'in_channels={}, out_channels={}'.format(self.in_channels, self.out_channels)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
-
-        return str_info
+            s += ', has_bias={}'.format(self.has_bias)
+        return s
 
 
 class _BaseAggregator(nn.Cell):

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

@@ -270,12 +270,9 @@ class Dense_Thor(Cell):
 
     def extend_repr(self):
         """extend_repr"""
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}' \
-            .format(self.in_channels, self.out_channels, self.weight, self.has_bias)
+        s = 'in_channels={}, out_channels={}'.format(self.in_channels, self.out_channels)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
-
+            s += ', bias={}'.format(self.bias)
         if self.activation_flag:
-            str_info = str_info + ', activation={}'.format(self.activation)
-
-        return str_info
+            s += ', activation={}'.format(self.activation)
+        return s

tests/st/gnn/aggregator.py

@@ -104,12 +104,10 @@ class GNNFeatureTransform(nn.Cell):
         return output
 
     def extend_repr(self):
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}' \
-            .format(self.in_channels, self.out_channels, self.weight, self.has_bias)
+        s = 'in_channels={}, out_channels={}'.format(self.in_channels, self.out_channels)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
-
-        return str_info
+            s += ', has_bias={}'.format(self.has_bias)
+        return s
 
 
 class _BaseAggregator(nn.Cell):

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

@@ -470,12 +470,10 @@ class Dense_Thor(Cell):
 
     def extend_repr(self):
         """extend_repr"""
-        str_info = 'in_channels={}, out_channels={}, weight={}, has_bias={}' \
-            .format(self.in_channels, self.out_channels, self.weight, self.has_bias)
+        s = 'in_channels={}, out_channels={}'.format(self.in_channels, self.out_channels)
         if self.has_bias:
-            str_info = str_info + ', bias={}'.format(self.bias)
-
+            s += ', has_bias={}'.format(self.has_bias)
         if self.activation_flag:
-            str_info = str_info + ', activation={}'.format(self.activation)
+            s += ', activation={}'.format(self.activation)
 
         return str_info