modify export

model_zoo/official/cv/FCN8s/README.md

@@ -340,6 +340,27 @@ Dataset used:
 python export.py
 ```
 
+- 在modelarts上导出MindIR
+
+```Modelarts
+在ModelArts上导出MindIR示例
+数据集存放方式同Modelart训练
+# (1) 选择a(修改yaml文件参数)或者b(ModelArts创建训练作业修改参数)其中一种方式。
+#       a. 设置 "enable_modelarts=True"
+#          设置 "file_name=fcn8s"
+#          设置 "file_format=MINDIR"
+#          设置 "ckpt_file=/cache/data/checkpoint file name"
+
+#       b. 增加 "enable_modelarts=True" 参数在modearts的界面上。
+#          在modelarts的界面上设置方法a所需要的参数
+#          注意：路径参数不需要加引号
+# (2)设置网络配置文件的路径 "_config_path=/The path of config in default_config.yaml/"
+# (3) 在modelarts的界面上设置代码的路径 "/path/fcn8s"。
+# (4) 在modelarts的界面上设置模型的启动文件 "export.py" 。
+# (5) 在modelarts的界面上设置模型的数据路径 ".../VOC2012/checkpoint"(选择VOC2012/checkpoint文件夹路径) ,
+# MindIR的输出路径"Output file path" 和模型的日志路径 "Job log path" 。
+```
+
 ## 推理过程
 
 ### 推理

model_zoo/official/cv/FCN8s/default_config.yaml

@@ -10,8 +10,7 @@ output_path: "/cache/train"
 load_path: "/cache/checkpoint_path"
 device_target: "Ascend"
 enable_profiling: False
-checkpoint_path: "./checkpoint/"
-checkpoint_file: "./checkpoint/.ckpt"
+
 # ======================================================================================
 # common options
 

model_zoo/official/cv/FCN8s/export.py

@@ -13,9 +13,8 @@
 # limitations under the License.
 # ============================================================================
 """export FCN8s."""
-
+import os
 import numpy as np
-
 import mindspore as ms
 from mindspore import Tensor
 from mindspore import context
@@ -23,10 +22,18 @@ from mindspore.train.serialization import load_checkpoint, load_param_into_net,
 from src.nets.FCN8s import FCN8s
 from src.model_utils.config import config
 from src.model_utils.device_adapter import get_device_id
+from src.model_utils.moxing_adapter import moxing_wrapper
+
 
 context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target, device_id=get_device_id())
 
-if __name__ == '__main__':
+
+def modelarts_pre_process():
+    config.file_name = os.path.join(config.output_path, config.file_name)
+
+
+@moxing_wrapper(pre_process=modelarts_pre_process)
+def model_export():
     net = FCN8s(n_class=config.num_classes)
 
     # load model
@@ -35,3 +42,7 @@ if __name__ == '__main__':
 
     input_arr = Tensor(np.zeros([1, 3, config.crop_size, config.crop_size]), ms.float32)
     export(net, input_arr, file_name=config.file_name, file_format=config.file_format)
+
+
+if __name__ == '__main__':
+    model_export()

model_zoo/official/cv/cnn_direction_model/README.md

@@ -240,6 +240,27 @@ python export.py --ckpt_file [CKPT_PATH] --device_target [DEVICE_TARGET] --file_
 
 `EXPORT_FORMAT` should be in ["AIR", "MINDIR"]
 
+- Export MindIR on Modelarts
+
+```Modelarts
+Export MindIR example on ModelArts
+Data storage method is the same as training
+# (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
+#       a. set "enable_modelarts=True"
+#          set "file_name=cnn_direction_model"
+#          set "file_format=MINDIR"
+#          set "ckpt_file=/cache/data/checkpoint file name"
+
+#       b. Add "enable_modelarts=True" parameter on the interface of modearts。
+#          Set the parameters required by method a on the modelarts interface
+#          Note: The path parameter does not need to be quoted
+# (2)Set the path of the network configuration file "_config_path=/The path of config in default_config.yaml/"
+# (3) Set the code path on the modelarts interface "/path/crnn"。
+# (4) Set the model's startup file on the modelarts interface "export.py" 。
+# (5) Set the data path of the model on the modelarts interface ".../crnn_dataset/eval/checkpoint"(choices crnn_dataset/eval/checkpoint Folder path) ,
+# The output path of the model "Output file path" and the log path of the model "Job log path"  。
+```
+
 ## [Inference Process](#contents)
 
 ### Usage

model_zoo/official/cv/cnn_direction_model/default_config.yaml

@@ -17,7 +17,6 @@ modelarts_dataset_unzip_name: "FSNS"
 # common options
 run_distribute: False
 
-
 # ======================================================================================
 # Training options
 
@@ -77,5 +76,3 @@ data_path: "The location of input data"
 output_pah: "The location of the output file"
 device_target: "device id of GPU or Ascend. (Default: None)"
 enable_profiling: "Whether enable profiling while training default: False"
-
-

model_zoo/official/cv/cnn_direction_model/export.py

@@ -13,24 +13,38 @@
 # limitations under the License.
 # ============================================================================
 """export script"""
-
+import os
 import numpy as np
 import mindspore as ms
 from mindspore import Tensor, context, load_checkpoint, export
 from src.cnn_direction_model import CNNDirectionModel
 from src.model_utils.config import config
 from src.model_utils.device_adapter import get_device_id
+from src.model_utils.moxing_adapter import moxing_wrapper
+
 
 context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target)
+
+
+def modelarts_pre_process():
+    config.file_name = os.path.join(config.output_path, config.file_name)
+
+
 device_id = get_device_id()
 context.set_context(device_id=device_id)
 
-if __name__ == '__main__':
+
+@moxing_wrapper(pre_process=modelarts_pre_process)
+def model_export():
     net = CNNDirectionModel([3, 64, 48, 48, 64], [64, 48, 48, 64, 64], [256, 64], [64, 512])
 
-    param_dict = load_checkpoint(config.ckpt_file, net=net)
+    load_checkpoint(config.ckpt_file, net=net)
     net.set_train(False)
 
     input_data = Tensor(np.zeros([1, 3, config.im_size_h, config.im_size_w]), ms.float32)
 
     export(net, input_data, file_name=config.file_name, file_format=config.file_format)
+
+
+if __name__ == '__main__':
+    model_export()

model_zoo/official/cv/cnnctc/README.md

@@ -164,7 +164,6 @@ The entire code structure is as following:
     |---src
         |---__init__.py    // init file
         |---cnn_ctc.py    // cnn_ctc network
-        |---config.py    // total config
         |---callback.py    // loss callback file
         |---dataset.py    // process dataset
         |---util.py    // routine operation
@@ -291,7 +290,7 @@ epoch: 1 step: 8698 , loss is 9.708542263610315, average time per step is 0.2184
 
 # (2) Set the path of the network configuration file  "_config_path=/The path of config in default_config.yaml/"
 # (3) Set the code path on the modelarts interface "/path/cnnctc"。
-# (4) Set the model's startup file on the modelarts interface "export.py" 。
+# (4) Set the model's startup file on the modelarts interface "train.py" 。
 # (5) Set the data path of the model on the modelarts interface ".../CNNCTC_Data/train"(choices CNNCTC_Data/train Folder path) ,
 # The output path of the model "Output file path" and the log path of the model "Job log path" 。
 # (6) start trainning the model。
@@ -309,7 +308,7 @@ epoch: 1 step: 8698 , loss is 9.708542263610315, average time per step is 0.2184
 
 # (3) Set the path of the network configuration file "_config_path=/The path of config in default_config.yaml/"
 # (4) Set the code path on the modelarts interface "/path/cnnctc"。
-# (5) Set the model's startup file on the modelarts interface "export.py" 。
+# (5) Set the model's startup file on the modelarts interface "train.py" 。
 # (6) Set the data path of the model on the modelarts interface ".../CNNCTC_Data/train"(choices CNNCTC_Data/train Folder path) ,
 # The output path of the model "Output file path" and the log path of the model "Job log path"  。
 # (7) Start model inference。
@@ -346,7 +345,7 @@ Export MindIR example on ModelArts
 Data storage method is the same as training
 # (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
 #       a. set "enable_modelarts=True"
-#          set "file_name=/cache/train/cnnctc"
+#          set "file_name=cnnctc"
 #          set "file_format=MINDIR"
 #          set "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/cnnctc/README_CN.md

@@ -347,7 +347,7 @@ python export.py --ckpt_file [CKPT_PATH] --file_format [EXPORT_FORMAT] --TEST_BA
 数据集存放方式同Modelart训练
 # (1) 选择a(修改yaml文件参数)或者b(ModelArts创建训练作业修改参数)其中一种方式。
 #       a. 设置 "enable_modelarts=True"
-#          设置 "file_name=/cache/train/cnnctc"
+#          设置 "file_name=cnnctc"
 #          设置 "file_format=MINDIR"
 #          设置 "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/cnnctc/default_config.yaml

@@ -11,10 +11,6 @@ load_path: "/cache/checkpoint_path"
 device_target: "Ascend"
 enable_profiling: False
 
-# ======================================================================================
-# common options
-
-
 # ======================================================================================
 # Training options
 CHARACTER: "0123456789abcdefghijklmnopqrstuvwxyz"
@@ -73,4 +69,3 @@ enable_profiling: "Whether enable profiling while training default: False"
 file_name: "CNN&CTC output air name"
 file_format: "choices [AIR, MINDIR]"
 ckpt_file: "CNN&CTC ckpt file"
-

model_zoo/official/cv/cnnctc/export.py

@@ -15,6 +15,7 @@
 """export checkpoint file into air, onnx, mindir models
    suggest run as python export.py --filename cnnctc --file_format MINDIR --ckpt_file [ckpt file path]
 """
+import os
 import numpy as np
 from mindspore import Tensor, context, load_checkpoint, export
 import mindspore.common.dtype as mstype
@@ -29,7 +30,7 @@ if config.device_target == "Ascend":
 
 
 def modelarts_pre_process():
-    pass
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
 @moxing_wrapper(pre_process=modelarts_pre_process)

model_zoo/official/cv/crnn/README.md

@@ -307,7 +307,7 @@ Export MindIR example on ModelArts
 Data storage method is the same as training
 # (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
 #       a. set "enable_modelarts=True"
-#          set "file_name=/cache/train/crnn"
+#          set "file_name=crnn"
 #          set "file_format=MINDIR"
 #          set "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/crnn/export.py

@@ -14,6 +14,7 @@
 # ============================================================================
 
 """ export model for CRNN """
+import os
 import numpy as np
 import mindspore as ms
 from mindspore import Tensor, context, load_checkpoint, export
@@ -24,7 +25,7 @@ from src.model_utils.device_adapter import get_device_id
 
 
 def modelarts_pre_process():
-    pass
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
 @moxing_wrapper(pre_process=modelarts_pre_process)

model_zoo/official/cv/ctpn/README.md

@@ -338,7 +338,7 @@ Export MindIR example on ModelArts
 Data storage method is the same as training
 # (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
 #       a. set "enable_modelarts=True"
-#          set "file_name=/cache/train/cnnctc"
+#          set "file_name=ctpn"
 #          set "file_format=MINDIR"
 #          set "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/ctpn/export.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 # ============================================================================
 """export checkpoint file into air, onnx, mindir models"""
+import os
 import numpy as np
 import mindspore as ms
 from mindspore import Tensor, load_checkpoint, load_param_into_net, export, context
@@ -29,7 +30,7 @@ if config.device_target == "Ascend":
 
 
 def modelarts_pre_process():
-    pass
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
 @moxing_wrapper(pre_process=modelarts_pre_process)

model_zoo/official/cv/densenet/README.md

@@ -399,6 +399,27 @@ python export.py --net [NET_NAME] --ckpt_file [CKPT_PATH] --device_target [DEVIC
 
 `EXPORT_FORMAT` should be in ["AIR", "MINDIR"]
 
+- Export MindIR on Modelarts
+
+```Modelarts
+Export MindIR example on ModelArts
+Data storage method is the same as training
+# (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
+#       a. set "enable_modelarts=True"
+#          set "file_name=densenet121"
+#          set "file_format=MINDIR"
+#          set "ckpt_file=/cache/data/checkpoint file name"
+
+#       b. Add "enable_modelarts=True" parameter on the interface of modearts。
+#          Set the parameters required by method a on the modelarts interface
+#          Note: The path parameter does not need to be quoted
+# (2)Set the path of the network configuration file "_config_path=/The path of config in densenet121_config.yaml/"
+# (3) Set the code path on the modelarts interface "/path/densenet121"。
+# (4) Set the model's startup file on the modelarts interface "export.py" 。
+# (5) Set the data path of the model on the modelarts interface ".../ImageNet_Original/checkpoint"(choices ImageNet_Original/checkpoint Folder path) ,
+# The output path of the model "Output file path" and the log path of the model "Job log path"  。
+```
+
 ## [Inference Process](#contents)
 
 ### Inference

model_zoo/official/cv/densenet/README_CN.md

@@ -398,6 +398,27 @@ python export.py --net [NET_NAME] --ckpt_file [CKPT_PATH] --device_target [DEVIC
 
 `EXPORT_FORMAT` 可选 ["AIR", "MINDIR"]
 
+- 在modelarts上导出MindIR
+
+```Modelarts
+在ModelArts上导出MindIR示例
+数据集存放方式同Modelart训练
+# (1) 选择a(修改yaml文件参数)或者b(ModelArts创建训练作业修改参数)其中一种方式。
+#       a. 设置 "enable_modelarts=True"
+#          设置 "file_name=densenet121"
+#          设置 "file_format=MINDIR"
+#          设置 "ckpt_file=/cache/data/checkpoint file name"
+
+#       b. 增加 "enable_modelarts=True" 参数在modearts的界面上。
+#          在modelarts的界面上设置方法a所需要的参数
+#          注意：路径参数不需要加引号
+# (2)设置网络配置文件的路径 "_config_path=/The path of config in densenet121_config.yaml/"
+# (3) 在modelarts的界面上设置代码的路径 "/path/densenet"。
+# (4) 在modelarts的界面上设置模型的启动文件 "export.py" 。
+# (5) 在modelarts的界面上设置模型的数据路径 ".../ImageNet_Original/eval/checkpoint"(选择ImageNet_Original/eval/checkpoint文件夹路径) ,
+# MindIR的输出路径"Output file path" 和模型的日志路径 "Job log path" 。
+```
+
 ## 推理过程
 
 ### 推理

model_zoo/official/cv/densenet/densenet100_config.yaml

@@ -10,9 +10,6 @@ output_path: "/cache/train"
 load_path: "/cache/checkpoint_path/"
 device_target: 'Ascend'
 enable_profiling: False
-checkpoint_path: "./checkpoint/"
-checkpoint_file: "./checkpoint/.ckpt"
-
 
 # ==============================================================================
 # Common options
@@ -22,7 +19,6 @@ is_distributed: 0
 rank: 0
 group_size: 1
 
-
 # ==============================================================================
 # Training options
 train_data_dir: ""
@@ -50,7 +46,6 @@ ckpt_interval: 3124
 save_ckpt_path: "./"
 is_save_on_master: 1
 
-
 # Eval options
 eval_data_dir: ""
 backbone: "resnet50"
@@ -58,7 +53,6 @@ ckpt_files: ""
 log_path: ""
 eval_url: ""
 
-
 # export options
 device_id: 0
 batch_size: 32

model_zoo/official/cv/densenet/densenet121_config.yaml

@@ -10,8 +10,6 @@ output_path: "/cache/train"
 load_path: "/cache/checkpoint_path/"
 device_target: 'Ascend'
 enable_profiling: False
-checkpoint_path: "./checkpoint/"
-checkpoint_file: "./checkpoint/.ckpt"
 
 # ==============================================================================
 # Common options
@@ -21,7 +19,6 @@ is_distributed: 0
 rank: 0
 group_size: 1
 
-
 # ==============================================================================
 # Training options
 train_data_dir: ""
@@ -49,7 +46,6 @@ ckpt_interval: 50000
 save_ckpt_path: "./"
 is_save_on_master: 1
 
-
 # Eval options
 eval_data_dir: ""
 backbone: "resnet50"
@@ -57,7 +53,6 @@ ckpt_files: ""
 log_path: ""
 eval_url: ""
 
-
 # export options
 device_id: 0
 batch_size: 32
@@ -95,4 +90,3 @@ backbone: "backbone"
 device_id: "Device id"
 file_name: "output file name"
 file_format: "file format choices [AIR MINDIR ONNX]"
-

model_zoo/official/cv/densenet/export.py

@@ -16,27 +16,32 @@
 """export checkpoint file into air, onnx, mindir models
    Suggest run as python export.py --file_name [file_name] --ckpt_files [ckpt path] --file_format [file format]
 """
-
+import os
 import numpy as np
 from mindspore.common import dtype as mstype
 from mindspore import context, Tensor
 from mindspore.train.serialization import export, load_checkpoint, load_param_into_net
 from src.model_utils.config import config
+from src.model_utils.moxing_adapter import moxing_wrapper
 
 
 context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target)
 
 
-if config.device_target == "Ascend":
-    context.set_context(device_id=config.device_id)
-
-if config.net == "densenet100":
-    from src.network.densenet import DenseNet100 as DenseNet
-else:
-    from src.network.densenet import DenseNet121 as DenseNet
+def modelarts_pre_process():
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
-if __name__ == "__main__":
+@moxing_wrapper(pre_process=modelarts_pre_process)
+def model_export():
+    if config.device_target == "Ascend":
+        context.set_context(device_id=config.device_id)
+
+    if config.net == "densenet100":
+        from src.network.densenet import DenseNet100 as DenseNet
+    else:
+        from src.network.densenet import DenseNet121 as DenseNet
+
     network = DenseNet(config.num_classes)
 
     param_dict = load_checkpoint(config.ckpt_files)
@@ -59,3 +64,7 @@ if __name__ == "__main__":
     input_data = Tensor(np.zeros(shape), mstype.float32)
 
     export(network, input_data, file_name=config.file_name, file_format=config.file_format)
+
+
+if __name__ == '__main__':
+    model_export()

model_zoo/official/cv/dpn/README.md

@@ -337,6 +337,27 @@ python export.py --config_path [CONFIG_PATH] --ckpt_file [CKPT_PATH] --file_name
 The ckpt_file parameter is required,
 `FILE_FORMAT` should be in ["AIR", "MINDIR"]
 
+- Export MindIR on Modelarts
+
+```Modelarts
+Export MindIR example on ModelArts
+Data storage method is the same as training
+# (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
+#       a. set "enable_modelarts=True"
+#          set "file_name=dpn"
+#          set "file_format=MINDIR"
+#          set "ckpt_file=/cache/data/checkpoint file name"
+
+#       b. Add "enable_modelarts=True" parameter on the interface of modearts。
+#          Set the parameters required by method a on the modelarts interface
+#          Note: The path parameter does not need to be quoted
+# (2)Set the path of the network configuration file "_config_path=/The path of config in default_config.yaml/"
+# (3) Set the code path on the modelarts interface "/path/dpn"。
+# (4) Set the model's startup file on the modelarts interface "export.py" 。
+# (5) Set the data path of the model on the modelarts interface ".../ImageNet_Original/eval/checkpoint"(choices ImageNet_Original/eval/checkpoint Folder path) ,
+# The output path of the model "Output file path" and the log path of the model "Job log path"  。
+```
+
 ### [Infer on Ascend310](#contents)
 
 Before performing inference, the mindir file must be exported by `export.py` script. We only provide an example of inference using MINDIR model.

model_zoo/official/cv/dpn/export.py

@@ -15,18 +15,27 @@
 """Export DPN
 suggest run as python export.py --file_name [filename] --file_format [file format] --checkpoint_path [ckpt path]
 """
-
+import os
 import numpy as np
 from mindspore import Tensor, context, load_checkpoint, load_param_into_net, export
 from src.dpn import dpns
 from src.model_utils.config import config
+from src.model_utils.moxing_adapter import moxing_wrapper
 
 
 context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target)
+
+
+def modelarts_pre_process():
+    config.file_name = os.path.join(config.output_path, config.file_name)
+
+
 if config.device_target == "Ascend":
     context.set_context(device_id=config.device_id)
 
-if __name__ == "__main__":
+
+@moxing_wrapper(pre_process=modelarts_pre_process)
+def model_export():
     # define net
     backbone = config.backbone
     num_classes = config.num_classes
@@ -39,3 +48,7 @@ if __name__ == "__main__":
 
     image = Tensor(np.zeros([config.batch_size, 3, config.height, config.width], np.float32))
     export(net, image, file_name=config.file_name, file_format=config.file_format)
+
+
+if __name__ == '__main__':
+    model_export()

model_zoo/official/cv/openpose/README.md

@@ -269,7 +269,7 @@ Export MindIR example on ModelArts
 Data storage method is the same as training
 # (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
 #       a. set "enable_modelarts=True"
-#          set "file_name=/cache/train/openpose"
+#          set "file_name=openpose"
 #          set "file_format=MINDIR"
 #          set "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/openpose/default_config.yaml

@@ -55,7 +55,6 @@ max_epoch_train: 60
 max_epoch_train_NP: 80
 loss_scale: 16384
 
-
 # default param
 batch_size: 10
 min_keypoints: 5
@@ -74,7 +73,6 @@ max_scale: 2.0
 max_rotate_degree: 40
 center_perterb_max: 40
 
-
 # ======================================================================================
 # Eval options
 is_distributed: 0
@@ -84,7 +82,6 @@ imgpath_val: ""
 ann: ""
 output_img_path: "./output_imgs/"
 
-
 # inference params
 inference_img_size: 368
 inference_scales: [0.5, 1, 1.5, 2]

model_zoo/official/cv/openpose/export.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 # ============================================================================
 """export"""
-
+import os
 import numpy as np
 from mindspore import Tensor
 from mindspore import context
@@ -27,7 +27,7 @@ context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target,
 
 
 def modelarts_pre_process():
-    pass
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
 @moxing_wrapper(pre_process=None)

model_zoo/official/cv/psenet/README.md

@@ -270,7 +270,7 @@ Export MindIR example on ModelArts
 Data storage method is the same as training
 # (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
 #       a. set "enable_modelarts=True"
-#          set "file_name=/cache/train/psenet"
+#          set "file_name=psenet"
 #          set "file_format=MINDIR"
 #          set "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/psenet/README_CN.md

@@ -269,7 +269,7 @@ python export.py --ckpt [CKPT_PATH] --file_name [FILE_NAME] --file_format [FILE_
 数据集存放方式同Modelart训练
 # (1) 选择a(修改yaml文件参数)或者b(ModelArts创建训练作业修改参数)其中一种方式。
 #       a. 设置 "enable_modelarts=True"
-#          设置 "file_name=/cache/train/psenet"
+#          设置 "file_name=psenet"
 #          设置 "file_format=MINDIR"
 #          设置 "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/official/cv/psenet/default_config.yaml

@@ -15,7 +15,6 @@ checkpoint_file: "./checkpoint/.ckpt"
 modelarts_home: "/home/work/user-job-dir"
 object_name: "psenet"
 
-
 # ======================================================================================
 # Training options
 pre_trained: ""
@@ -46,7 +45,6 @@ TRAIN_REPEAT_NUM: 1800
 TRAIN_DROP_REMAINDER: True
 TRAIN_MODEL_SAVE_PATH: "./"
 
-
 # ======================================================================================
 # Eval options
 ckpt: ""

model_zoo/official/cv/psenet/export.py

@@ -15,7 +15,7 @@
 """
 ##############export checkpoint file into air models#################
 """
-
+import os
 import numpy as np
 import mindspore as ms
 from mindspore import Tensor, load_checkpoint, load_param_into_net, export, context
@@ -25,11 +25,17 @@ from src.model_utils.moxing_adapter import moxing_wrapper
 
 
 context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target)
+
+
+def modelarts_pre_process():
+    config.file_name = os.path.join(config.output_path, config.file_name)
+
+
 if config.device_target == "Ascend":
     context.set_context(device_id=config.device_id)
 
 
-@moxing_wrapper(pre_process=None)
+@moxing_wrapper(pre_process=modelarts_pre_process)
 def model_export():
     net = ETSNet(config)
     param_dict = load_checkpoint(config.ckpt)

model_zoo/official/cv/retinanet/README_CN.md

@@ -346,7 +346,7 @@ python export.py  --file_name retinanet --file_format MINDIR --checkpoint_path /
 在ModelArts上导出MindIR示例
 # (1) 选择a(修改yaml文件参数)或者b(ModelArts创建训练作业修改参数)其中一种方式。
 #       a. 设置 "enable_modelarts=True"
-#          设置 "file_name=/cache/train/cnnctc"
+#          设置 "file_name=retinanet"
 #          设置 "file_format=MINDIR"
 #          设置 "checkpoint_path=/cache/data/checkpoint/checkpoint file name"
 

model_zoo/official/cv/retinanet/export.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 # ============================================================================
 """export for retinanet"""
+import os
 import numpy as np
 import mindspore.common.dtype as mstype
 from mindspore import context, Tensor
@@ -24,7 +25,7 @@ from src.model_utils.moxing_adapter import moxing_wrapper
 
 
 def modelarts_pre_process():
-    pass
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
 @moxing_wrapper(pre_process=modelarts_pre_process)

model_zoo/official/cv/shufflenetv1/README_CN.md

@@ -227,6 +227,27 @@ python export.py --ckpt_path [CKPT_PATH] --device_target [DEVICE_TARGET] --file_
 
 `EXPORT_FORMAT` 可选 ["AIR", "MINDIR"]
 
+- Export MindIR on Modelarts
+
+```Modelarts
+Export MindIR example on ModelArts
+Data storage method is the same as training
+# (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
+#       a. set "enable_modelarts=True"
+#          set "file_name=shufflenetv1"
+#          set "file_format=MINDIR"
+#          set "ckpt_file=/cache/data/checkpoint file name"
+
+#       b. Add "enable_modelarts=True" parameter on the interface of modearts。
+#          Set the parameters required by method a on the modelarts interface
+#          Note: The path parameter does not need to be quoted
+# (2)Set the path of the network configuration file "_config_path=/The path of config in default_config.yaml/"
+# (3) Set the code path on the modelarts interface "/path/shufflenetv1"。
+# (4) Set the model's startup file on the modelarts interface "export.py" 。
+# (5) Set the data path of the model on the modelarts interface ".../ImageNet_Original/eval/checkpoint"(choices ImageNet_Original/eval/checkpoint Folder path) ,
+# The output path of the model "Output file path" and the log path of the model "Job log path"  。
+```
+
 ## 推理过程
 
 ### 推理

model_zoo/official/cv/shufflenetv1/default_config.yaml

@@ -10,8 +10,7 @@ output_path: "/cache/train"
 load_path: "/cache/checkpoint_path"
 device_target: "Ascend"
 enable_profiling: False
-checkpoint_path: "./checkpoint/"
-checkpoint_file: "./checkpoint/.ckpt"
+
 # ======================================================================================
 # common options
 num_classes: 1000
@@ -46,13 +45,11 @@ loss_scale: 1024
 weight_decay: 0.00004
 momentum: 0.9
 
-
 # ======================================================================================
 # Eval options
 ckpt_path: ""
 eval_dataset_path: ""
 
-
 # ======================================================================================
 # export options
 file_name: "shufflenetv1"

model_zoo/official/cv/shufflenetv1/export.py

@@ -16,21 +16,28 @@
 ##############export checkpoint file into air, onnx, mindir models#################
 suggest run as python export.py --file_name [file name] --ckpt_path [ckpt path] --file_format [file format]
 """
-
+import os
 import numpy as np
 import mindspore as ms
 from mindspore import Tensor, load_checkpoint, load_param_into_net, export, context
 from src.model_utils.config import config
 from src.shufflenetv1 import ShuffleNetV1
+from src.model_utils.moxing_adapter import moxing_wrapper
 
 
 context.set_context(mode=context.GRAPH_MODE, device_target=config.device_target)
+
+
+def modelarts_pre_process():
+    config.file_name = os.path.join(config.output_path, config.file_name)
+
+
 if config.device_target == "Ascend":
     context.set_context(device_id=config.device_id)
 
 
-if __name__ == '__main__':
-
+@moxing_wrapper(pre_process=modelarts_pre_process)
+def model_export():
     net = ShuffleNetV1(model_size=config.model_size)
 
     param_dict = load_checkpoint(config.ckpt_path)
@@ -39,3 +46,7 @@ if __name__ == '__main__':
     image_height, image_width = (224, 224)
     input_arr = Tensor(np.ones([config.batch_size, 3, image_height, image_width]), ms.float32)
     export(net, input_arr, file_name=config.file_name, file_format=config.file_format)
+
+
+if __name__ == '__main__':
+    model_export()

model_zoo/research/nlp/dscnn/README.md

@@ -397,7 +397,7 @@ Export MindIR example on ModelArts
 Data storage method is the same as training
 # (1) Choose either a (modify yaml file parameters) or b (modelArts create training job to modify parameters)。
 #       a. set "enable_modelarts=True"
-#          set "file_name=/cache/train/dscnn"
+#          set "file_name=dscnn"
 #          set "file_format=MINDIR"
 #          set "ckpt_file=/cache/data/checkpoint file name"
 

model_zoo/research/nlp/dscnn/default_config.yaml

@@ -28,9 +28,6 @@ per_batch_size: 100
 model_size_info: [6, 276, 10, 4, 2, 1, 276, 3, 3, 2, 2, 276, 3, 3, 1, 1, 276, 3, 3, 1, 1, 276, 3, 3, 1,
                                  1, 276, 3, 3, 1, 1
 ]
-
-
-
 # ======================================================================================
 # Training options
 amp_level: "O0"

model_zoo/research/nlp/dscnn/export.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 # ===========================================================================
 """DSCNN export."""
+import os
 import numpy as np
 from mindspore import Tensor
 from mindspore.train.serialization import export
@@ -23,7 +24,7 @@ from src.model_utils.moxing_adapter import moxing_wrapper
 
 
 def modelarts_pre_process():
-    pass
+    config.file_name = os.path.join(config.output_path, config.file_name)
 
 
 @moxing_wrapper(pre_process=None)