!1331 delete dropoutgenmask and dropoutdomask when dropout prob equals 0 to enhance performance and adjust ci script

Merge pull request !1331 from yoonlee666/master-deletedropout

example/bert_clue/dataset.py

@@ -42,7 +42,7 @@ def create_bert_dataset(epoch_size=1, device_num=1, rank=0, do_shuffle="true", e
     if enable_data_sink == "true":
         new_size = data_sink_steps * bert_net_cfg.batch_size
     ds.set_dataset_size(new_size)
-    repeat_count = int(repeat_count * ori_dataset_size // ds.get_dataset_size())
+    new_repeat_count = int(repeat_count * ori_dataset_size // ds.get_dataset_size())
     type_cast_op = C.TypeCast(mstype.int32)
     ds = ds.map(input_columns="masked_lm_ids", operations=type_cast_op)
     ds = ds.map(input_columns="masked_lm_positions", operations=type_cast_op)
@@ -55,4 +55,4 @@ def create_bert_dataset(epoch_size=1, device_num=1, rank=0, do_shuffle="true", e
     ds = ds.repeat(repeat_count)
     logger.info("data size: {}".format(ds.get_dataset_size()))
     logger.info("repeatcount: {}".format(ds.get_repeat_count()))
-    return ds
+    return ds, new_repeat_count

example/bert_clue/run_pretrain.py

@@ -24,7 +24,7 @@ from mindspore import context
 from mindspore.train.model import Model
 from mindspore.train.parallel_utils import ParallelMode
 from mindspore.nn.wrap.loss_scale import DynamicLossScaleUpdateCell
-from mindspore.train.callback import Callback, ModelCheckpoint, CheckpointConfig
+from mindspore.train.callback import Callback, ModelCheckpoint, CheckpointConfig, TimeMonitor
 from mindspore.train.serialization import load_checkpoint, load_param_into_net
 from mindspore.model_zoo.Bert_NEZHA import BertNetworkWithLoss, BertTrainOneStepCell, BertTrainOneStepWithLossScaleCell
 from mindspore.nn.optim import Lamb, Momentum, AdamWeightDecayDynamicLR
@@ -87,8 +87,9 @@ def run_pretrain():
         rank = 0
         device_num = 1
 
-    ds = create_bert_dataset(args_opt.epoch_size, device_num, rank, args_opt.do_shuffle, args_opt.enable_data_sink,
-                             args_opt.data_sink_steps, args_opt.data_dir, args_opt.schema_dir)
+    ds, new_repeat_count = create_bert_dataset(args_opt.epoch_size, device_num, rank, args_opt.do_shuffle,
+                                               args_opt.enable_data_sink, args_opt.data_sink_steps,
+                                               args_opt.data_dir, args_opt.schema_dir)
 
     netwithloss = BertNetworkWithLoss(bert_net_cfg, True)
 
@@ -112,7 +113,7 @@ def run_pretrain():
     else:
         raise ValueError("Don't support optimizer {}, only support [Lamb, Momentum, AdamWeightDecayDynamicLR]".
                          format(cfg.optimizer))
-    callback = [LossCallBack()]
+    callback = [TimeMonitor(ds.get_dataset_size()), LossCallBack()]
     if args_opt.enable_save_ckpt == "true":
         config_ck = CheckpointConfig(save_checkpoint_steps=args_opt.save_checkpoint_steps,
                                      keep_checkpoint_max=args_opt.save_checkpoint_num)
@@ -133,6 +134,6 @@ def run_pretrain():
         netwithgrads = BertTrainOneStepCell(netwithloss, optimizer=optimizer)
 
     model = Model(netwithgrads)
-    model.train(ds.get_repeat_count(), ds, callbacks=callback, dataset_sink_mode=(args_opt.enable_data_sink == "true"))
+    model.train(new_repeat_count, ds, callbacks=callback, dataset_sink_mode=(args_opt.enable_data_sink == "true"))
 if __name__ == '__main__':
     run_pretrain()

mindspore/nn/layer/basic.py

@@ -99,6 +99,9 @@ class Dropout(Cell):
             out, _ = self.dropout(x)
             return out
 
+        if self.keep_prob == 1:
+            return x
+
         shape = self.get_shape(x)
         dtype = P.DType()(x)
         keep_prob = self.cast(self.keep_prob, dtype)

tests/st/networks/models/bert/bert_tdt_lossscale.py

@@ -26,7 +26,7 @@ from mindspore import context
 from mindspore import log as logger
 from mindspore.common.tensor import Tensor
 from mindspore.model_zoo.Bert_NEZHA import BertConfig, BertNetworkWithLoss, BertTrainOneStepWithLossScaleCell
-from mindspore.nn.optim import Momentum
+from mindspore.nn.optim import Lamb
 from mindspore.train.callback import Callback
 from mindspore.train.loss_scale_manager import DynamicLossScaleManager
 from mindspore.train.model import Model
@@ -73,7 +73,7 @@ def get_config(version='base', batch_size=1):
             max_position_embeddings=512,
             type_vocab_size=2,
             initializer_range=0.02,
-            use_relative_positions=True,
+            use_relative_positions=False,
             input_mask_from_dataset=True,
             token_type_ids_from_dataset=True,
             dtype=mstype.float32,
@@ -138,7 +138,9 @@ def test_bert_tdt():
     batch_size = int(os.getenv('BATCH_SIZE', '16'))
     config = get_config(version=version, batch_size=batch_size)
     netwithloss = BertNetworkWithLoss(config, True)
-    optimizer = Momentum(netwithloss.trainable_params(), learning_rate=2e-5, momentum=0.9)
+    optimizer = Lamb(netwithloss.trainable_params(), decay_steps=ds.get_dataset_size()*ds.get_repeat_count(),
+                     start_learning_rate=5e-5, end_learning_rate=1e-9,
+                     power=10.0, warmup_steps=0, weight_decay=0.01)
     scale_window = 3
     scale_manager = DynamicLossScaleManager(2 ** 16, 2, scale_window)
     netwithgrads = BertTrainOneStepWithLossScaleCell(netwithloss, optimizer=optimizer,
@@ -169,10 +171,10 @@ def test_bert_tdt():
 
     # assertion occurs while the loss value, overflow state or loss_scale value is wrong
     loss_value = np.array(callback.loss_list)
-    expect_loss_value = [12.191826, 11.966009, 11.972208, 11.98216, 11.973932, 12.611078, 12.17554, 12.840299,
-                         12.403329, 12.621632]
+    expect_loss_value = [12.207201, 11.980862, 11.984737, 11.879344, 11.832838, 12.411388,
+                         12.009449, 12.621273, 12.223175, 12.427313]
     print("loss value: {}".format(loss_value))
-    assert np.allclose(loss_value, expect_loss_value, 0.00001, 0.00001)
+    assert np.allclose(loss_value, expect_loss_value, 0, 0.0005)
 
     overflow = np.array(callback.overflow_list)
     expect_overflow = [True, True, False, False, False, True, False, False, False, True]
@@ -182,7 +184,7 @@ def test_bert_tdt():
     loss_scale = np.array(callback.lossscale_list)
     expect_loss_scale = [32768.0, 16384.0, 16384.0, 16384.0, 32768.0, 16384.0, 16384.0, 16384.0, 32768.0, 16384.0]
     print("loss scale: {}".format(loss_scale))
-    assert np.allclose(loss_scale, expect_loss_scale, 0.00001, 0.00001)
+    assert np.allclose(loss_scale, expect_loss_scale, 0, 0)
 
 
 if __name__ == '__main__':