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

Added RFC for Feast Python SDK #14

Merged
merged 2 commits into from
Jan 2, 2019
Merged

Added RFC for Feast Python SDK #14

merged 2 commits into from
Jan 2, 2019

Conversation

woop
Copy link
Member

@woop woop commented Dec 24, 2018

This pull request adds an RFC for a Python SDK

@woop woop changed the title Added python-sdk RFC Added RFC for Feast Python SDK Dec 24, 2018
Copy link
Collaborator

@zhilingc zhilingc left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we change the sample workflow to follow more closely to what @pradithya and I have done? other than that, looks great

@woop
Copy link
Member Author

woop commented Dec 25, 2018

Can we change the sample workflow to follow more closely to what @pradithya and I have done? other than that, looks great

Absolutely.

The feature set is simply an object that locally tracks which entity, granularity, and features you are interested in.

```python
feature_set = fs.create_feature_set(entity='driver', granularity='minute', features=['latitude', 'longitude', 'event_time'])
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you think it make more sense to have FeatureSet creation outside of client, and then to get training/serving data we pass the instance into the corresponding client method?

# create feature set
feature_set = FeatureSet(entity='driver', granularity='minute', features=['latitude', 'longitude', 'event_time'])

# training dataset
dataset_info = fs.create_training_dataset(feature_set, start_date='2018-01-01', end_date='2018-02-01')
dataset_info.download(dest=file_path, type='feather')
df = dataset_info.download_to_df()

# serving
feature_data = fs.get_serving_data(feature_set, keys, type='last')

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we want to pass feature_set into fs methods, or do we want to pass the client (fs) into feature_set methods?

@zhilingc @tims

rfcs/0000-python-sdk.md Outdated Show resolved Hide resolved
@woop
Copy link
Member Author

woop commented Jan 1, 2019

I've incorporated your feedback @pradithya, @zhilingc. I have also written out a reference level explanation of the important classes and methods. Can you please have a look?

https://github.com/gojek/feast/blob/ccf6ace51b1df920d870d83c133167e5e5e5ac68/rfcs/0000-python-sdk.md

@pradithya pradithya merged commit 6323b00 into feast-dev:master Jan 2, 2019
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