Remove estimators

[lionelkusch]

For the moment, the estimator are not necessary, they can be added latter.
See #57 and #167

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 97.77%. Comparing base (fff131f) to head (b12f917).
Report is 1 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #166       +/-   ##
===========================================
+ Coverage   75.08%   97.77%   +22.69%     
===========================================
  Files          24       23        -1     
  Lines        1533     1126      -407     
===========================================
- Hits         1151     1101       -50     
+ Misses        382       25      -357     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[bthirion]

I'm a bit confused: PR #167 only restaures a small fraction of the code... maybe it would be gould to already include the next bits ?

[bthirion]

LGTM otherwise

[lionelkusch]

I'm a bit confused: PR #167 only restaures a small fraction of the code... maybe it would be gould to already include the next bits ?

I separate the modified random forest and the DNN estimator in 2 pull requests. PR #167 is only about the modified random forest and requires a small modification of the random forest from scitkit learn.
We can talk about their utility in the associate PR.

[bthirion]

I think it would make sense to prepare the second PR on the DNN estimator before merging that one ?

[lionelkusch]

I think it would make sense to prepare the second PR on the DNN estimator before merging that one ?

This is done in the PR #57.

[jpaillard]

The file examples_not_exhibited/plot_residuals_sampling.py contains an example showing how to use the modified random forest for conditional sampling. I suggest removing it from the main and adding it to the PR related to the modified RF estimator.

[bthirion]

LGTM, thx.