Data Science

This issue is reserved for the Data Umbrella Africa & Middle East Sprint on October 23, 2021 as a good first issue. For non-sprint contributors, please wait till October 30, 2021 to work on this issue.

Note

For those who are running into "YD01: No Yields section found", it could be the cv parameter. Update `An iterable yielding (train, test) sp

Currently, the funnel report percentage is calculated using:
The number at a given funnel step /
Sum(everything in the funnel)

Example from blog:

Here, the Discussed Pricing (900) gets divided by 11900 (sum of all ev

When users starts Serve cluster with a set of options (http options, checkpoint path) and then connects to it with a different set of options, we should either update it, or error out.

Problem

I am just exploring streamlit and having a great time so far! I have video data that I want to visually inspect, but they are short videos of ~5 seconds. Currently it seems st.video does not have a loop option.

Solution

Add a loop option to st.video.

Not sure how difficult this would be. If not, I would be happy to work on a PR if someone could give me some quick po

Proposed refactoring or deprecation

Remove deprecated code after the 1.5 release

Pitch

All the deprecated features we have are tested here

https://github.com/PyTorchLightning/pytorch-lightning/blob/master/tests/deprecated_api/test_remove_1-6.py

If you are interested in taking care of one item, post a comment here asking to take it. This avoids multiple people working on the s

In recent versions (can't say from exactly when), there seems to be an off-by-one error in dcc.DatePickerRange. I set max_date_allowed = datetime.today().date(), but in the calendar, yesterday is the maximum date allowed. I see it in my apps, and it is also present in the first example on the DatePickerRange documentation page.

E

in #13206 there is a test that seem to always be skipped. Once merged we could delete it.

Originally posted by @Carreau in ipython/ipython#13206 (comment)

In gensim/models/fasttext.py:

    model = FastText(
        vector_size=m.dim,
        vector_size=m.dim,
        window=m.ws,
        window=m.ws,
        epochs=m.epoch,
        epochs=m.epoch,
        negative=m.neg,
        negative=m.neg,
        # FIXME: these next 2 lines read in unsupported FB FT modes (loss=3 softmax or loss=4 onevsall,
        # or model=3 supervi

Is your feature request related to a problem? Please describe.
I typically used compressed datasets (e.g. gzipped) to save disk space. This works fine with AllenNLP during training because I can write my dataset reader to load the compressed data. However, the predict command opens the file and reads lines for the Predictor. This fails when it tries to load data from my compressed files.