p60975123
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

modify if xxx is True to if xxx and modify if xxx is not True to if not xxx

This commit is contained in:
muchenjin 2022-01-12 22:31:49 +08:00
parent a743cc8d88
commit 718862fd6f
20 changed files with 38 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
mindspore/python/mindspore/mindrecord/filewriter.py
View File

@ -388,7 +388,7 @@ class FileWriter:
if not self._writer.get_shard_header():
self._writer.set_shard_header(self._header)
ret = self._writer.commit()
if self._index_generator is True:
if self._index_generator:
if self._append:
self._generator = ShardIndexGenerator(self._file_name, self._append)
elif len(self._paths) >= 1:
@ -470,7 +470,7 @@ class FileWriter:
return False, error
elif len(v) == 2 and 'type' in v:
res_1, res_2 = self._validate_array(k, v)
if res_1 is not True:
if not res_1:
return res_1, res_2
else:
error = "Field '{}' contains illegal attributes.".format(v)

4
mindspore/python/mindspore/nn/cell.py
View File

@ -483,7 +483,7 @@ class Cell(Cell_):
item.init_data()
elif isinstance(item, numpy.ndarray):
raise TypeError("For 'Cell', inputs should not be numpy array.")
if self.requires_grad is True:
if self.requires_grad:
_pynative_executor.set_grad_flag(True)
_pynative_executor.new_graph(self, *args, **kwargs)
cast_inputs = self.auto_cast_inputs(args)
@ -1518,7 +1518,7 @@ class Cell(Cell_):
def _set_recompute_scope(self, mode):
prefix = 'recompute_'
if mode is True:
if mode:
if self._scope is None:
self._scope = prefix
elif not self._scope.startswith(prefix):

2
mindspore/python/mindspore/nn/layer/normalization.py
View File

@ -221,7 +221,7 @@ class _BatchNorm(Cell):
self.moving_mean,
self.moving_variance)[0]
if self.use_batch_statistics is True:
if self.use_batch_statistics:
return self.bn_train(x,
self.gamma,
self.beta,

4
mindspore/python/mindspore/nn/metrics/confusion_matrix.py
View File

@ -239,7 +239,7 @@ class ConfusionMatrixMetric(Metric):
y_pred = self._convert_data(inputs[0])
y = self._convert_data(inputs[1])
if self.calculation_method is True:
if self.calculation_method:
score, not_nans = self.confusion_matrix(y_pred, y)
not_nans = int(not_nans.item())
self._total_num += score.item() * not_nans
@ -260,7 +260,7 @@ class ConfusionMatrixMetric(Metric):
ndarray, the computed result.
"""
if self.calculation_method is True:
if self.calculation_method:
if self._class_num == 0:
raise RuntimeError("The 'ConfusionMatrixMetric' can not be calculated, because the number of samples "
"is 0, please check whether your inputs(predicted value, true value) are empty, or "

2
mindspore/python/mindspore/nn/transformer/layers.py
View File

@ -376,7 +376,7 @@ class _Linear(Cell):
def construct(self, x):
out_shape = P.Shape()(x)[:-1] + (self.out_channels,)
x = P.Reshape()(x, (-1, self.in_channels))
if self.expert_flag is True:
if self.expert_flag:
x = P.Reshape()(x, (self.expert_num, -1, self.in_channels))
weight = self.cast(self.weight, self.dtype)
x = self.matmul(x, weight)

2
mindspore/python/mindspore/nn/transformer/moe.py
View File

@ -270,7 +270,7 @@ class Router(Cell):
def construct(self, input_tensor):
input_tensor = self.cast(input_tensor, mstype.float32)
if self.noisy_policy == "jitter" and self.training is True:
if self.noisy_policy == "jitter" and self.training:
# Here, we temporarily implement the multiplicative jitter this way,
# for the lack of UniforReal parallel operator.
input_tensor = self.mul(input_tensor, self.noise)

26
mindspore/python/mindspore/nn/transformer/transformer.py
View File

@ -1309,7 +1309,7 @@ class TransformerEncoderLayer(Cell):
parallel_config=parallel_config)
_check_moe_config(moe_config, parallel_config)
self.use_moe = (moe_config.expert_num > 1)
if self.use_moe is True:
if self.use_moe:
self.output = MoE(hidden_size=hidden_size,
dropout_rate=hidden_dropout_rate,
ffn_hidden_size=ffn_hidden_size,
@ -1378,7 +1378,7 @@ class TransformerEncoderLayer(Cell):
output_x = self.layernorm2(x)
output_x = F.cast(output_x, self.dtype)
aux_loss = None
if self.use_moe is True:
if self.use_moe:
mlp_logit, aux_loss = self.output(output_x)
else:
mlp_logit = self.output(output_x)
@ -1416,7 +1416,7 @@ class TransformerEncoderLayer(Cell):
output = self.add(x, mlp_logit)
output = F.reshape(output, x_shape)
if self.use_moe is True:
if self.use_moe:
return output, layer_present, aux_loss
return output, layer_present
@ -1588,7 +1588,7 @@ class TransformerDecoderLayer(Cell):
"divisibled by 'parallel_config.model_parallel', but got the ffn_hidden_size is {} "
"and parallel_config.model_parallel is {}."
.format(ffn_hidden_size, parallel_config.model_parallel))
if use_past is True:
if use_past:
raise ValueError(f"The {self.cls_name} does not support use_past=True.")
self.batch_size = batch_size
self.use_past = use_past
@ -1632,7 +1632,7 @@ class TransformerDecoderLayer(Cell):
self.cross_attention_layernorm.shard(((parallel_config.data_parallel, 1),))
_check_moe_config(moe_config, parallel_config)
self.use_moe = (moe_config.expert_num > 1)
if self.use_moe is True:
if self.use_moe:
self.output = MoE(hidden_size=hidden_size,
dropout_rate=hidden_dropout_rate,
ffn_hidden_size=ffn_hidden_size,
@ -1718,7 +1718,7 @@ class TransformerDecoderLayer(Cell):
output_x = self.layernorm2(x)
output_x = F.cast(output_x, self.dtype)
aux_loss = None
if self.use_moe is True:
if self.use_moe:
mlp_logit, aux_loss = self.output(output_x)
else:
mlp_logit = self.output(output_x)
@ -1756,7 +1756,7 @@ class TransformerDecoderLayer(Cell):
output = self.add(x, mlp_logit)
output = F.reshape(output, hidden_shape)
if self.use_moe is True:
if self.use_moe:
return output, layer_present, aux_loss
return output, layer_present
@ -2044,7 +2044,7 @@ class TransformerEncoder(Cell):
def construct(self, hidden_states, attention_mask, init_reset=True, batch_valid_length=None):
present_layer = ()
if self.use_moe is True:
if self.use_moe:
accum_loss = self.aux_loss
for i in range(self.num_layers):
hidden_states, present, aux_loss = self.blocks[i](hidden_states,
@ -2242,7 +2242,7 @@ class TransformerDecoder(Cell):
def construct(self, hidden_states, attention_mask, encoder_output=None, memory_mask=None,
init_reset=True, batch_valid_length=None):
present_layer = ()
if self.use_moe is True:
if self.use_moe:
accum_loss = self.aux_loss
for i in range(self.num_layers):
hidden_states, present, aux_loss = self.blocks[i](hidden_states,
@ -2433,7 +2433,7 @@ class Transformer(Cell):
if encoder_layers <= 0 < decoder_layers:
raise ValueError(f"Transformer doest support encoder layer {encoder_layers} and decoder"
f"layer {decoder_layers}, please use TransformerDecoder")
if encoder_layers > 0 and decoder_layers > 0 and use_past is True:
if encoder_layers > 0 and decoder_layers > 0 and use_past:
raise ValueError(f"The {self.cls_name} with encoder and decoder does not support use_past=True.")
if _get_parallel_mode() in (ParallelMode.AUTO_PARALLEL,):
raise RuntimeError(f"The {self.cls_name} does not support auto parallel mode now.")
@ -2503,7 +2503,7 @@ class Transformer(Cell):
decoder_layer_present = None
accum_loss = self.aux_loss
if self.encoder is not None:
if self.use_moe is True:
if self.use_moe:
encoder_output, encoder_layer_present, encoder_aux_loss = self.encoder(encoder_inputs, encoder_masks,
init_reset, batch_valid_length)
accum_loss = self.add(accum_loss, encoder_aux_loss)
@ -2514,7 +2514,7 @@ class Transformer(Cell):
if self.decoder is not None:
# decoder mask should be created outside of the model
if self.use_moe is True:
if self.use_moe:
decoder_output, decoder_layer_present, decoder_aux_loss = self.decoder(decoder_inputs, decoder_masks,
encoder_output, memory_mask,
init_reset, batch_valid_length)
@ -2526,6 +2526,6 @@ class Transformer(Cell):
memory_mask, init_reset,
batch_valid_length)
output = decoder_output
if self.use_moe is True:
if self.use_moe:
return output, encoder_layer_present, decoder_layer_present, accum_loss
return output, encoder_layer_present, decoder_layer_present

2
mindspore/python/mindspore/ops/_grad_experimental/grad_array_ops.py
View File

@ -37,7 +37,7 @@ def get_bprop_masked_select(self):
dinput, dvalue = binop_grad_common(input_data, mask, dinput, dvalue)
dvalue = sum_op(dvalue)
dinput = F.cast(dinput, F.dtype(input_data))
if is_instance_op(value, mstype.number) is True:
if is_instance_op(value, mstype.number):
dvalue = 0
else:
dvalue = F.cast(dvalue, F.dtype(value))

2
mindspore/python/mindspore/ops/_grad_experimental/grad_math_ops.py
View File

@ -70,7 +70,7 @@ def get_bprop_index_lerp(self):
dend = mul_op(dout, weight)
dweight = mul_op(dout, sub_op(end, start))
dstart, dend = binop_grad_common(start, end, dstart, dend)
if is_instance_op(weight, mstype.number) is True:
if is_instance_op(weight, mstype.number):
dweight = 0
else:
_, dweight = binop_grad_common(start, weight, dstart, dweight)

4
mindspore/python/mindspore/ops/bprop_mindir/generate_mindir.py
View File

@ -66,7 +66,7 @@ if __name__ == "__main__":
# If the specified bprop source directory is not on the mindspore installed path,
# copy the bprop source files to the installed path.
backup_suffix = "_generate_bak"
if copy_flag is True:
if copy_flag:
shutil.rmtree(bprop_installed_dir + backup_suffix, ignore_errors=True)
os.rename(bprop_installed_dir, bprop_installed_dir + backup_suffix)
os.mkdir(bprop_installed_dir)
@ -81,7 +81,7 @@ if __name__ == "__main__":
# If the specified bprop source directory is not on the mindspore installed path,
# copy the generated mindir files to the mindir directory relative to the specified path.
if copy_flag is True:
if copy_flag:
shutil.rmtree(bprop_installed_dir)
os.rename(bprop_installed_dir + backup_suffix, bprop_installed_dir)
ls = os.listdir(bprop_mindir_export_dir)

2
mindspore/python/mindspore/ops/operations/math_ops.py
View File

@ -5683,7 +5683,7 @@ class IsClose(Primitive):
validator.check_value_type('rtol', rtol, [float], self.name)
validator.check_value_type('atol', atol, [float], self.name)
validator.check_value_type('equal_nan', equal_nan, [bool], self.name)
if equal_nan is not True:
if not equal_nan:
raise ValueError("For IsClose, the `equal_nan` must be True, but got False.")
validator.check_non_negative_float(rtol, 'rtol', self.name)
validator.check_non_negative_float(atol, 'atol', self.name)

2
mindspore/python/mindspore/profiler/common/util.py
View File

@ -55,7 +55,7 @@ def fwrite_format(output_data_path, data_source=None, is_print=False, is_start=F
is_start (bool): Whether is the first line of the output file, will remove the old file if True."
"""
if is_start is True and os.path.exists(output_data_path):
if is_start and os.path.exists(output_data_path):
os.remove(output_data_path)
if isinstance(data_source, str) and data_source.startswith("title:"):

2
mindspore/python/mindspore/train/callback/_checkpoint.py
View File

@ -446,7 +446,7 @@ class ModelCheckpoint(Callback):
return True
elif self._config.save_checkpoint_seconds and self._config.save_checkpoint_seconds > 0:
self._cur_time = time.time()
if (self._cur_time - self._last_time) > self._config.save_checkpoint_seconds or force_to_save is True:
if (self._cur_time - self._last_time) > self._config.save_checkpoint_seconds or force_to_save:
self._last_time = self._cur_time
return True

6
mindspore/python/mindspore/train/model.py
View File

@ -756,7 +756,7 @@ class Model:
>>> model.train(2, dataset)
"""
dataset_sink_mode = Validator.check_bool(dataset_sink_mode)
if isinstance(self._train_network, nn.GraphCell) and dataset_sink_mode is True:
if isinstance(self._train_network, nn.GraphCell) and dataset_sink_mode:
raise ValueError("Dataset sink mode is currently not supported when training with a GraphCell.")
if hasattr(train_dataset, '_warmup_epoch') and train_dataset._warmup_epoch != epoch:
@ -942,7 +942,7 @@ class Model:
if not self._metric_fns:
raise ValueError("The model argument 'metrics' can not be None or empty, "
"you should set the argument 'metrics' for model.")
if isinstance(self._eval_network, nn.GraphCell) and dataset_sink_mode is True:
if isinstance(self._eval_network, nn.GraphCell) and dataset_sink_mode:
raise ValueError("Sink mode is currently not supported when evaluating with a GraphCell.")
cb_params = _InternalCallbackParam()
@ -1019,7 +1019,7 @@ class Model:
dataset_sink_mode = Validator.check_bool(dataset_sink_mode)
if not dataset_sink_mode:
raise ValueError("Only dataset sink mode is supported for now.")
if isinstance(self._train_network, nn.GraphCell) and dataset_sink_mode is True:
if isinstance(self._train_network, nn.GraphCell) and dataset_sink_mode:
raise ValueError("Dataset sink mode is currently not supported when training with a GraphCell.")
Validator.check_is_int(sink_size)
dataset_size = train_dataset.get_dataset_size()

2
tests/st/auto_parallel/multifieldembeddinglookup_parallel.py
View File

@ -97,7 +97,7 @@ class FakeData:
self.is_onehot = True
self.fakedata_mode = fakedata_mode
if use_parallel is True:
if use_parallel:
init(backend_name='nccl')
self.rank_size = get_group_size()
self.rank_id = get_rank()

2
tests/st/auto_parallel/optimizer_parallel.py
View File

@ -99,7 +99,7 @@ class FakeData:
self.is_onehot = True
self.fakedata_mode = fakedata_mode
if use_parallel is True:
if use_parallel:
init(backend_name='hccl')
self.rank_size = get_group_size()
self.rank_id = get_rank()

2
tests/st/auto_parallel/parallel_strategy_search.py
View File

@ -102,7 +102,7 @@ class FakeData:
self.is_onehot = True
self.fakedata_mode = fakedata_mode
if use_parallel is True:
if use_parallel:
init(backend_name='hccl')
self.rank_size = get_group_size()
self.rank_id = get_rank()

2
tests/st/mix_precision/utils.py
View File

@ -81,7 +81,7 @@ class FakeData:
self.is_onehot = True
self.fakedata_mode = fakedata_mode
if use_parallel is True:
if use_parallel:
init()
self.rank_size = get_group_size()
self.rank_id = get_rank()

2
tests/st/networks/models/resnet50/src_thor/model_thor.py
View File

@ -574,7 +574,7 @@ class Model:
>>> model.train(2, dataset)
"""
repeat_count = train_dataset.get_repeat_count()
if epoch != repeat_count and dataset_sink_mode is True:
if epoch != repeat_count and dataset_sink_mode:
logger.warning(f"The epoch_size {epoch} is not the same with dataset repeat_count {repeat_count}")
dataset_sink_mode = Validator.check_bool(dataset_sink_mode)
_device_number_check(self._parallel_mode, self._device_number)

2
tests/vm_impl/nn_ops_vm_impl.py
View File

@ -346,7 +346,7 @@ def vm_impl_momentum(self):
learning_rate = np.full(shape, learning_rate.asnumpy())
momentum = np.full(shape, momentum.asnumpy())
accumulation = accumulation * momentum + gradient
if use_nesterov is True:
if use_nesterov:
variable -= gradient * learning_rate + accumulation * momentum * learning_rate
else:
variable -= accumulation * learning_rate