MLBazaar/MLPrimitives

Typo in the primitive keras.Sequential.LSTMTimeSeriesRegressor

DanielCalvoCerezo opened this issue · 4 comments

  • MLPrimitives version:0.1.10
  • Python version:N/A
  • Operating System:N/A

Description

There is a typo in line 74 of MLPrimitives/mlprimitives/jsons/keras.Sequential.LSTMTimeSeriesRegressor.json
It says bastch_size and should say batch_size. Otherwise, the batch size of the training cannot be tuned

What I Did

Simply correct the typo and it works :)
csala commented

Thanks for pointing it out @DanielCalvoCerezo !
Would you mind doing a PR with the change?

Hi @csala
I am trying to create a new branch to add a PR with the fix but I cannot. I do not know if I need to be added to the project as collaborator or it is a matter of my setup here (our proxy is really picky)
Sorry for the trouble

csala commented

Hi @DanielCalvoCerezo
You probably need to create a fork of the project under your account and then make the PR from there.

You can see the exact instructions about the development setup here: https://hdi-project.github.io/MLPrimitives/readme.html#install-for-development

You can also have a look at the contributing step-by-step guide for this scenario: https://hdi-project.github.io/MLPrimitives/community/annotations.html#modifying-an-existing-annotation
Since we have already discussed this and the issue is already assigned to you, you can jump directly to point 4 of this list.

Also, thanks for willing to contribute! :-)

csala commented

Closed via #178