Azure / Azure/MachineLearningNotebooks
python package azureml-contrib-pipeline-steps 1.20.0 not working
- 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.