Azure / Azure/MachineLearningNotebooks

python package azureml-contrib-pipeline-steps 1.20.0 not working

Đang mở
#1,417 3 bình luận 0 reaction 0 người được giao Xem trên GitHub
ADO bug Compute
Ngôn ngữ chính
Jupyter Notebook
Star
4.4k
Fork
2.6k
Chỉ số merge pull request
Không có pull request nào được merge trong 30 ngày

Mô tả

## Describe the issue

version 1.20.0 of python package azureml-contrib-pipeline-steps throws (works fine on version 1.19 or 1.18)

File "C:/Users/v-songshanli/projects/ashexplore/object_identification/obj_segmentation_azure_2_steps.py", line 88, in run
pipeline = Pipeline(workspace=ws, steps=pipeline_steps)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\core\_experiment_method.py", line 97, in wrapper
return init_func(self, *args, **kwargs)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\pipeline.py", line 177, in __init__
self._graph = self._graph_builder.build(self._name, steps, finalize=False)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\builder.py", line 1481, in build
graph = self.construct(name, steps)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\builder.py", line 1503, in construct
self.process_collection(steps)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\builder.py", line 1539, in process_collection
builder.process_collection(collection)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\builder.py", line 1830, in process_collection
self._base_builder.process_collection(item)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\builder.py", line 1533, in process_collection
return self.process_step(collection)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\builder.py", line 1577, in process_step
node = step.create_node(self._graph, self._default_datastore, self._context)
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\steps\python_script_step.py", line 243, in create_node
return super(PythonScriptStep, self).create_node(
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\_python_script_step_base.py", line 140, in create_node
self._set_compute_params_to_node(node,
File "C:\Users\v-songshanli\Anaconda3\envs\pytouchEnv\lib\site-packages\azureml\pipeline\core\_python_script_step_base.py", line 229, in _set_compute_params_to_node
self._module_param_provider.set_params_to_node(
TypeError: _set_params_to_node_hook() got an unexpected keyword argument 'command'

## Minimal example

```python
from azureml.core import Workspace

ws = Workspace.from_config()

split_step = PythonScriptStep(
name="Train Test Split",
script_name="obj_segment_step_data_process.py",
arguments=["--data-path", dataset.as_named_input('pennfudan_data').as_mount(),
"--train-split", train_split_data, "--test-split", test_split_data,
"--test-size", 50],
compute_target=compute_target,
runconfig=aml_run_config,
source_directory=source_directory,
allow_reuse=False
)

pipeline_steps = [split_step ]

pipeline = Pipeline(workspace=ws, steps=pipeline_steps)
```

## Additional context
I am using aml sdk 1.20. no type errors with version 1.19/1.18 of azureml-contrib-pipeline-steps.
-

Hướng dẫn đóng góp

Chưa lập chỉ mục được hướng dẫn đóng góp cho kho mã nguồn này

Đánh giá

Issue này chưa được đánh giá.

Nhận issue mới trong hộp thư của bạn

Bản tóm tắt ngắn những issue GitHub phù hợp với người mới.