Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update environment #15

Merged
merged 2 commits into from
Oct 23, 2023
Merged

Conversation

fnattino
Copy link
Collaborator

@fnattino fnattino commented Sep 29, 2023

Update the environment file for two reasons:

  • the RF model is saved in a pickle file. Reading the model fails if the version of scikit-learn in the environment does not match the version used to write the file. For now, I suggest to pin the version of scikit learn, but we should maybe find a better format to store the model.
  • geopandas and shapely, used in the Europe notebook, are not part of the environment.

@fnattino fnattino changed the title Pin scikit-learn Update environment Sep 29, 2023
@geek-yang
Copy link
Member

  • the RF model is saved in a pickle file. Reading the model fails if the version of scikit-learn in the environment does not match the version used to write the file. For now, I suggest to pin the version of scikit learn, but we should maybe find a better format to store the model.

Good point @fnattino ! I would propose ONNX (we also have an issue for it #6! The conversion is very simple and here is an example notebook. @QianqianHan96 we can look into it together.

@QianqianHan96
Copy link
Collaborator

Update the environment file for two reasons:

  • the RF model is saved in a pickle file. Reading the model fails if the version of scikit-learn in the environment does not match the version used to write the file. For now, I suggest to pin the version of scikit learn, but we should maybe find a better format to store the model.
  • geopandas and shapely, used in the Europe notebook, are not part of the environment.

Thanks Francesco for updating the environment.yaml file.

@QianqianHan96
Copy link
Collaborator

  • the RF model is saved in a pickle file. Reading the model fails if the version of scikit-learn in the environment does not match the version used to write the file. For now, I suggest to pin the version of scikit learn, but we should maybe find a better format to store the model.

Good point @fnattino ! I would propose ONNX (we also have an issue for it #6! The conversion is very simple and here is an example notebook. @QianqianHan96 we can look into it together.

Thanks Yang for your advice, I will have a look at this example notebook and then try to convert the pickle file. I will let you know if I meet problems.

@geek-yang
Copy link
Member

geek-yang commented Oct 1, 2023

Thanks Yang for your advice, I will have a look at this example notebook and then try to convert the pickle file. I will let you know if I meet problems.

Great! Keep me posted if you got any issue. We could also do it together @QianqianHan96 .

@QianqianHan96
Copy link
Collaborator

Thanks Yang for your advice, I will have a look at this example notebook and then try to convert the pickle file. I will let you know if I meet problems.

Great! Keep me posted if you got any issue. We could also do it together @QianqianHan96 .

Thanks, Yang. I am wondering how could we do it together, in this pull request or create a new pull request?

@geek-yang
Copy link
Member

Thanks Yang for your advice, I will have a look at this example notebook and then try to convert the pickle file. I will let you know if I meet problems.

Great! Keep me posted if you got any issue. We could also do it together @QianqianHan96 .

Thanks, Yang. I am wondering how could we do it together, in this pull request or create a new pull request?

Sorry for the late reply @QianqianHan96 . For ONNX let's do it in another PR, please review this PR first. We could take about it in our next sprint.

@QianqianHan96 QianqianHan96 merged commit b32cc57 into EcoExtreML:main Oct 23, 2023
@QianqianHan96
Copy link
Collaborator

Thanks Yang, I have reviewed this PR and merged it. I agree, let's discuss about the ONNX on 27 Oct.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants