Summary

For every image in LightGBM's documentation, alternative text that describes the image should be added.

Motivation

People with limited vision use screen readers to view web pages. When these screen readers encounter an image, they will only read some placeholder like "IMAGE" by default. However, if the "alt" tag is specified,

Not a high-priority at all, but it'd be more sensible for such a tutorial/testing utility corpus to be implemented elsewhere - maybe under /test/ or some other data- or doc- related module – rather than in gensim.models.word2vec.

Originally posted by @gojomo in RaRe-Technologies/gensim#2939 (comment)

Problem: the approximate method can still be slow for many trees
catboost version: master
Operating System: ubuntu 18.04
CPU: i9
GPU: RTX2080

Would be good to be able to specify how many trees to use for shapley. The model.predict and prediction_type versions allow this. lgbm/xgb allow this.

Right now, pages created by in-memory http driver does not support navigation between pages due to their static nature.
It makes it problematic to "move" from one static page to another within a same domain simply because we cannot reuse current page. In order to do so, we have to create a new page every time we need to move from A to B.
Let's add a possibility to reuse existing page by imp

The official instructions say to use joblib for pickling PyOD models.

This fails for AutoEncoders, or any other TensorFlow-backed model as far as I can tell. The error is:

>>> dump(model, 'model.joblib')
...
TypeError: can't pickle _thread.RLock objects

Note that it's not sufficient to save the underlying Keras S

Functions for forecasting intermittent demand time series.

Conventional methods for modeling(like ARIMA) are not suited for time series which is Intermittent in nature.
We need specialized methods to model such Demands/Time Series

Methods like Croston's can be used to model such series
A rough sketch of the API

croston=Croston(...)
cro