aws / aws/aws-step-functions-data-science-sdk-python
Make sagemaker an optional dependency
- Dominant language
- Python
- Stars
- 299
- Forks
- 84
- PR merge metrics
- No merged PRs in 30d
Description
I'd like to see sagemaker marked as an optional dependency, as it is not always required.
### Use Case
Not all users of this package are using the sagemaker dependency, and it adds a lot of bloat to overall installations by forcing users to have gcc and g++ to be able to install numpy and pandas.
### Proposed Solution
It's quite simple to actually do this.
In `setup.py` you'd have to change `install_requires` and add `extras_require`:
```python
setup(
# [...]
install_requires = [
"boto3>=1.14.38",
"pyyaml"
]
extras_require = {
"sagemaker": ["sagemaker>=2.1.0"],
"test": [] # [...]
}
# [...]
)
```
In addition to this, some extra documentation should probably be added to the README to make it easier for users to understand how to install the extra dependency (`pip install stepfunctions[sagemaker]`).
### Other
If this is implemented, the major version probably has to be bumped because it's a change in how the package is installed.
---
* [x] :wave: I may be able to implement this feature request
* [x] :warning: This feature might incur a breaking change
---
This is a :rocket: Feature Request
Contributor guide
Assessment
This issue has not been assessed yet.