!2388 GPU update bert scripts

Merge pull request !2388 from VectorSL/bert-eval

model_zoo/bert/evaluation.py

@@ -18,9 +18,11 @@ Bert evaluation script.
 """
 
 import os
+import argparse
 import numpy as np
 import mindspore.common.dtype as mstype
 from mindspore import context
+from mindspore import log as logger
 from mindspore.common.tensor import Tensor
 import mindspore.dataset as de
 import mindspore.dataset.transforms.c_transforms as C
@@ -105,8 +107,17 @@ def bert_predict(Evaluation):
     '''
     prediction function
     '''
-    devid = int(os.getenv('DEVICE_ID'))
-    context.set_context(mode=context.GRAPH_MODE, device_target="Ascend", device_id=devid)
+    target = args_opt.device_target
+    if target == "Ascend":
+        devid = int(os.getenv('DEVICE_ID'))
+        context.set_context(mode=context.GRAPH_MODE, device_target="Ascend", device_id=devid)
+    elif target == "GPU":
+        context.set_context(mode=context.GRAPH_MODE, device_target="GPU")
+        if bert_net_cfg.compute_type != mstype.float32:
+            logger.warning('GPU only support fp32 temporarily, run with fp32.')
+            bert_net_cfg.compute_type = mstype.float32
+    else:
+        raise Exception("Target error, GPU or Ascend is supported.")
     dataset = get_dataset(bert_net_cfg.batch_size, 1)
     if cfg.use_crf:
         net_for_pretraining = Evaluation(bert_net_cfg, False, num_labels=len(tag_to_index), use_crf=True,
@@ -147,6 +158,9 @@ def test_eval():
                                                                       callback.acc_num / callback.total_num))
         print("==============================================================")
 
+parser = argparse.ArgumentParser(description='Bert eval')
+parser.add_argument('--device_target', type=str, default='Ascend', help='Device target')
+args_opt = parser.parse_args()
 if __name__ == "__main__":
     num_labels = cfg.num_labels
     test_eval()

model_zoo/bert/finetune.py

@@ -23,6 +23,7 @@ from src.utils import BertFinetuneCell, BertCLS, BertNER, BertSquad, BertSquadCe
 from src.finetune_config import cfg, bert_net_cfg, tag_to_index
 import mindspore.common.dtype as mstype
 from mindspore import context
+from mindspore import log as logger
 import mindspore.dataset as de
 import mindspore.dataset.transforms.c_transforms as C
 from mindspore.nn.wrap.loss_scale import DynamicLossScaleUpdateCell
@@ -105,6 +106,9 @@ def test_train():
         context.set_context(mode=context.GRAPH_MODE, device_target="Ascend", device_id=devid)
     elif target == "GPU":
         context.set_context(mode=context.GRAPH_MODE, device_target="GPU")
+        if bert_net_cfg.compute_type != mstype.float32:
+            logger.warning('GPU only support fp32 temporarily, run with fp32.')
+            bert_net_cfg.compute_type = mstype.float32
     else:
         raise Exception("Target error, GPU or Ascend is supported.")
     #BertCLSTrain for classification

tests/st/networks/test_gpu_alexnet.py

@@ -86,6 +86,6 @@ def test_trainTensor(num_classes=10, epoch=15, batch_size=32):
     for i in range(0, epoch):
         data = Tensor(np.ones([batch_size, 3, 227, 227]).astype(np.float32) * 0.01)
         label = Tensor(np.ones([batch_size]).astype(np.int32))
-        loss = train_network(data, label)
+        loss = train_network(data, label).asnumpy()
         losses.append(loss)
-    assert (losses[-1].asnumpy() < 0.01)
+    assert losses[-1] < 0.01

tests/st/networks/test_gpu_lenet.py

@@ -144,9 +144,9 @@ def test_train_lenet():
     for i in range(epoch):
         data = Tensor(np.ones([net.batch_size, 3, 32, 32]).astype(np.float32) * 0.01)
         label = Tensor(np.ones([net.batch_size]).astype(np.int32))
-        loss = train_network(data, label)
+        loss = train_network(data, label).asnumpy()
         losses.append(loss)
-    print(losses)
+    assert losses[-1] < 0.01
 
 
 def create_dataset(data_path, batch_size=32, repeat_size=1,