[MRG] Add pprint for estimators - continued #11705

NicolasHug · 2018-07-29T21:02:33Z

Reference Issues/PRs

This is the continuation of #9099 that aims to have better printing representations for estimators

Close #9099
Close #9039
Close #7618
Fix #6323

What does this implement/fix? Explain your changes.

~~Fixed some line length issues~~
~~Factored code for Estimator and Pipeline into a single method~~
Completely refactored code to use extend the built-in PrettyPrinter class
Added some tests
Added some doc

Any other comments?

I marked this as WIP but a first review may be useful. I tried to make the code easier to understand but it's always a bit hard with those kind of functionalities.

Line length limit is still not perfect, see e.g. the test with RFE(...) that goes beyond 80 chars because the printed estimator is actually an argument.

Questions

Should the file _pprint.py be in utils instead of at the root? (test file would follow)
Should I remove the _pprint() method in base.py and use the Formatter class in __repr__ instead?

…into pr/9099

jnothman · 2018-07-31T09:34:31Z

best if you ping after the 0.20 RC is out

…

amueller · 2018-08-16T20:44:52Z

can you add a test with a pipeline inside a grid-search please? And the pipeline maybe has a ColumnTransformer? ;)

because callable() returns True also for class objects (which we want to reprensent with their name as well anyway)

jnothman · 2018-09-04T23:02:06Z

Don't worry about python 2 support. It will be dropped in the next release.

NicolasHug · 2018-09-11T15:47:09Z

So due to some limitations of the current method, I completely changed the implementation and relied on pprint.PrettyPrinter instead.

Here are some output examples:

LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
                   intercept_scaling=1, max_iter=100, multi_class='ovr',
                   n_jobs=1, penalty='l2', random_state=None,
                   solver='liblinear', tol=0.0001, verbose=0, warm_start=False)

Pipeline(memory=None,
         steps=['reduce_dim': PCA(copy=True, iterated_power='auto',
                                  n_components=None, random_state=None,
                                  svd_solver='auto', tol=0.0, whiten=False),
                'classify': LinearSVC(C=1.0, class_weight=None, dual=True,
                                      fit_intercept=True, intercept_scaling=1,
                                      loss='squared_hinge', max_iter=1000,
                                      multi_class='ovr', penalty='l2',
                                      random_state=None, tol=0.0001,
                                      verbose=0)])

RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=LogisticRegression(C=1.0,
                                                                                                                     class_weight=None,
                                                                                                                     dual=False,
                                                                                                                     fit_intercept=True,
                                                                                                                     intercept_scaling=1,
                                                                                                                     max_iter=100,
                                                                                                                     multi_class='ovr',
                                                                                                                     n_jobs=1,
                                                                                                                     penalty='l2',
                                                                                                                     random_state=None,
                                                                                                                     solver='liblinear',
                                                                                                                     tol=0.0001,
                                                                                                                     verbose=0,
                                                                                                                     warm_start=False),
                                                                                        n_features_to_select=None,
                                                                                        step=1,
                                                                                        verbose=0),
                                                                          n_features_to_select=None,
                                                                          step=1,
                                                                          verbose=0),
                                                            n_features_to_select=None,
                                                            step=1, verbose=0),
                                              n_features_to_select=None, step=1,
                                              verbose=0),
                                n_features_to_select=None, step=1, verbose=0),
                  n_features_to_select=None, step=1, verbose=0),
    n_features_to_select=None, step=1, verbose=0)

GridSearchCV(cv=5, error_score='raise-deprecating',
             estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
                           decision_function_shape='ovr', degree=3,
                           gamma='auto_deprecated', kernel='rbf', max_iter=-1,
                           probability=False, random_state=None, shrinking=True,
                           tol=0.001, verbose=False),
             fit_params=None, iid='warn', n_jobs=1,
             param_grid=[{'C': [1, 10, 100, 1000], 'gamma': [0.001, 0.0001],
                          'kernel': ['rbf']},
                         {'C': [1, 10, 100, 1000], 'kernel': ['linear']}],
             pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',
             scoring=None, verbose=0)

GridSearchCV(cv=3, error_score='raise-deprecating',
             estimator=Pipeline(memory=None,
                                steps=['reduce_dim': PCA(copy=True,
                                                         iterated_power='auto',
                                                         n_components=None,
                                                         random_state=None,
                                                         svd_solver='auto',
                                                         tol=0.0,
                                                         whiten=False),
                                       'classify': LinearSVC(C=1.0,
                                                             class_weight=None,
                                                             dual=True,
                                                             fit_intercept=True,
                                                             intercept_scaling=1,
                                                             loss='squared_hinge',
                                                             max_iter=1000,
                                                             multi_class='ovr',
                                                             penalty='l2',
                                                             random_state=None,
                                                             tol=0.0001,
                                                             verbose=0)]),
             fit_params=None, iid='warn', n_jobs=1,
             param_grid=[{'classify__C': [1, 10, 100, 1000],
                          'reduce_dim': [PCA(copy=True, iterated_power=7,
                                             n_components=None,
                                             random_state=None,
                                             svd_solver='auto', tol=0.0,
                                             whiten=False),
                                         NMF(alpha=0.0, beta_loss='frobenius',
                                             init=None, l1_ratio=0.0,
                                             max_iter=200, n_components=None,
                                             random_state=None, shuffle=False,
                                             solver='cd', tol=0.0001,
                                             verbose=0)],
                          'reduce_dim__n_components': [2, 4, 8]},
                         {'classify__C': [1, 10, 100, 1000],
                          'reduce_dim': [SelectKBest(k=10,
                                                     score_func=<function chi2 at 0x7f77d68c18c8>)],
                          'reduce_dim__k': [2, 4, 8]}],
             pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',
             scoring=None, verbose=0)

The default is to indent after the name of the estimator, but it's expensive in line width. Unfortunately, PrettyPrinter does not really support other forms of indentations:

The amount of indentation added for each recursive level is specified by indent; the default is one. Other values can cause output to look a little odd

From the docs. Example of discrepancy:

pp = PrettyPrinter(indent=4, compact=True) 
pp.pprint(list(range(30)))

will give

[   0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
    21, 22, 23, 24, 25, 26, 27, 28, 29]

Here is what the output looks like on the same examples when indenting with 4 characters:

LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
    intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,
    penalty='l2', random_state=None, solver='liblinear', tol=0.0001, verbose=0,
    warm_start=False)

Pipeline(memory=None,
    steps=[   'reduce_dim': PCA(copy=True, iterated_power='auto',
                                n_components=None, random_state=None,
                                svd_solver='auto', tol=0.0, whiten=False),
              'classify': LinearSVC(C=1.0, class_weight=None, dual=True,
                              fit_intercept=True, intercept_scaling=1,
                              loss='squared_hinge', max_iter=1000,
                              multi_class='ovr', penalty='l2',
                              random_state=None, tol=0.0001, verbose=0)])

RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=LogisticRegression(C=1.0,
                                                                                                      class_weight=None,
                                                                                                      dual=False,
                                                                                                      fit_intercept=True,
                                                                                                      intercept_scaling=1,
                                                                                                      max_iter=100,
                                                                                                      multi_class='ovr',
                                                                                                      n_jobs=1,
                                                                                                      penalty='l2',
                                                                                                      random_state=None,
                                                                                                      solver='liblinear',
                                                                                                      tol=0.0001,
                                                                                                      verbose=0,
                                                                                                      warm_start=False),
                                                                                        n_features_to_select=None,
                                                                                        step=1,
                                                                                        verbose=0),
                                                                          n_features_to_select=None,
                                                                          step=1,
                                                                          verbose=0),
                                                            n_features_to_select=None,
                                                            step=1, verbose=0),
                                              n_features_to_select=None, step=1,
                                              verbose=0),
                                n_features_to_select=None, step=1, verbose=0),
                  n_features_to_select=None, step=1, verbose=0),
    n_features_to_select=None, step=1, verbose=0)

GridSearchCV(cv=5, error_score='raise-deprecating',
    estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
                  decision_function_shape='ovr', degree=3,
                  gamma='auto_deprecated', kernel='rbf', max_iter=-1,
                  probability=False, random_state=None, shrinking=True,
                  tol=0.001, verbose=False),
    fit_params=None, iid='warn', n_jobs=1,
    param_grid=[   {   'C': [1, 10, 100, 1000], 'gamma': [0.001, 0.0001],
                       'kernel': ['rbf']},
                   {'C': [1, 10, 100, 1000], 'kernel': ['linear']}],
    pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',
    scoring=None, verbose=0)

GridSearchCV(cv=3, error_score='raise-deprecating',
    estimator=Pipeline(memory=None,
                  steps=[   'reduce_dim': PCA(copy=True, iterated_power='auto',
                                              n_components=None,
                                              random_state=None,
                                              svd_solver='auto', tol=0.0,
                                              whiten=False),
                            'classify': LinearSVC(C=1.0, class_weight=None, dual=True,
                                            fit_intercept=True,
                                            intercept_scaling=1,
                                            loss='squared_hinge', max_iter=1000,
                                            multi_class='ovr', penalty='l2',
                                            random_state=None, tol=0.0001,
                                            verbose=0)]),
    fit_params=None, iid='warn', n_jobs=1,
    param_grid=[   {   'classify__C': [1, 10, 100, 1000],
                       'reduce_dim': [   PCA(copy=True, iterated_power=7,
                                             n_components=None,
                                             random_state=None,
                                             svd_solver='auto', tol=0.0,
                                             whiten=False),
                                         NMF(alpha=0.0, beta_loss='frobenius',
                                             init=None, l1_ratio=0.0,
                                             max_iter=200, n_components=None,
                                             random_state=None, shuffle=False,
                                             solver='cd', tol=0.0001,
                                             verbose=0)],
                       'reduce_dim__n_components': [2, 4, 8]},
                   {   'classify__C': [1, 10, 100, 1000],
                       'reduce_dim': [   SelectKBest(k=10,
                                             score_func=<function chi2 at 0x7f4a124bd8c8>)],
                       'reduce_dim__k': [2, 4, 8]}],
    pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',
    scoring=None, verbose=0)

I personally like the first output. It might render quite long strings when estimators are nested but I don't think that it's a major issue, and it's very easy to read.

ping @amueller

jnothman · 2018-09-12T10:19:18Z

Yes, I much prefer the first. Good idea to reuse PrettyPrinter. Will that be configurable for us wanting to hide default parameter settings, etc?

NicolasHug · 2018-09-12T13:28:21Z

Everything is encapsulated in BaseEstimator.__repr__() so we can hide the default parameters, if that what you mean?

I noticed that the steps is formatted as a list of key: values instead of a list of tuples. Is it OK like this or do we want to change it? I guess it's slightly clearer this way but it prevents the string from being evaluated with eval for example.

jnothman · 2018-09-13T07:55:12Z

I think it should be a list of tuples and am surprised that it is not.

NicolasHug · 2018-09-13T16:30:21Z

I fixed it, here is the new look:

LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
                   intercept_scaling=1, max_iter=100, multi_class='ovr',
                   n_jobs=1, penalty='l2', random_state=None,
                   solver='liblinear', tol=0.0001, verbose=0, warm_start=False)

Pipeline(memory=None,
         steps=[('reduce_dim',
                 PCA(copy=True, iterated_power='auto', n_components=None,
                     random_state=None, svd_solver='auto', tol=0.0,
                     whiten=False)),
                ('classify',
                 LinearSVC(C=1.0, class_weight=None, dual=True,
                           fit_intercept=True, intercept_scaling=1,
                           loss='squared_hinge', max_iter=1000,
                           multi_class='ovr', penalty='l2', random_state=None,
                           tol=0.0001, verbose=0))])

RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=RFE(estimator=LogisticRegression(C=1.0,
                                                                                                                     class_weight=None,
                                                                                                                     dual=False,
                                                                                                                     fit_intercept=True,
                                                                                                                     intercept_scaling=1,
                                                                                                                     max_iter=100,
                                                                                                                     multi_class='ovr',
                                                                                                                     n_jobs=1,
                                                                                                                     penalty='l2',
                                                                                                                     random_state=None,
                                                                                                                     solver='liblinear',
                                                                                                                     tol=0.0001,
                                                                                                                     verbose=0,
                                                                                                                     warm_start=False),
                                                                                        n_features_to_select=None,
                                                                                        step=1,
                                                                                        verbose=0),
                                                                          n_features_to_select=None,
                                                                          step=1,
                                                                          verbose=0),
                                                            n_features_to_select=None,
                                                            step=1, verbose=0),
                                              n_features_to_select=None, step=1,
                                              verbose=0),
                                n_features_to_select=None, step=1, verbose=0),
                  n_features_to_select=None, step=1, verbose=0),
    n_features_to_select=None, step=1, verbose=0)

GridSearchCV(cv=5, error_score='raise-deprecating',
             estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
                           decision_function_shape='ovr', degree=3,
                           gamma='auto_deprecated', kernel='rbf', max_iter=-1,
                           probability=False, random_state=None, shrinking=True,
                           tol=0.001, verbose=False),
             fit_params=None, iid='warn', n_jobs=1,
             param_grid=[{'C': [1, 10, 100, 1000], 'gamma': [0.001, 0.0001],
                          'kernel': ['rbf']},
                         {'C': [1, 10, 100, 1000], 'kernel': ['linear']}],
             pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',
             scoring=None, verbose=0)

GridSearchCV(cv=3, error_score='raise-deprecating',
             estimator=Pipeline(memory=None,
                                steps=[('reduce_dim',
                                        PCA(copy=True, iterated_power='auto',
                                            n_components=None,
                                            random_state=None,
                                            svd_solver='auto', tol=0.0,
                                            whiten=False)),
                                       ('classify',
                                        LinearSVC(C=1.0, class_weight=None,
                                                  dual=True, fit_intercept=True,
                                                  intercept_scaling=1,
                                                  loss='squared_hinge',
                                                  max_iter=1000,
                                                  multi_class='ovr',
                                                  penalty='l2',
                                                  random_state=None, tol=0.0001,
                                                  verbose=0))]),
             fit_params=None, iid='warn', n_jobs=1,
             param_grid=[{'classify__C': [1, 10, 100, 1000],
                          'reduce_dim': [PCA(copy=True, iterated_power=7,
                                             n_components=None,
                                             random_state=None,
                                             svd_solver='auto', tol=0.0,
                                             whiten=False),
                                         NMF(alpha=0.0, beta_loss='frobenius',
                                             init=None, l1_ratio=0.0,
                                             max_iter=200, n_components=None,
                                             random_state=None, shuffle=False,
                                             solver='cd', tol=0.0001,
                                             verbose=0)],
                          'reduce_dim__n_components': [2, 4, 8]},
                         {'classify__C': [1, 10, 100, 1000],
                          'reduce_dim': [SelectKBest(k=10,
                                                     score_func=<function chi2 at 0x7f046321a950>)],
                          'reduce_dim__k': [2, 4, 8]}],
             pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',
             scoring=None, verbose=0)

I'll put this as MRG for reviews, and if it gets approved I'll start updating the docs with the new formatting to fix the tests.

jnothman

I've realised the change is a bit weak on documentation. Not sure where to put it in the user guide, though. Or whether print_changed_only should be illustrated in the example gallery.

Otherwise, I'm keen to see this road tested.

jnothman · 2018-12-18T14:24:37Z

doc/whats_new/v0.21.rst

@@ -191,6 +191,10 @@ Support for Python 3.4 and below has been officially dropped.
 Multiple modules
 ................

+- The `__repr__()` method of all estimators (used when calling
+  `print(estimator)`) has been entirely re-written. :issue:`11705` by
+  :user:`Nicolas Hug <NicolasHug>`.


print_changed_only needs to be mentioned in what's new. I think it should also be mentioned somewhere in the documentation more explicitly.

jnothman · 2018-12-18T14:25:27Z

doc/whats_new/v0.21.rst

@@ -191,6 +191,10 @@ Support for Python 3.4 and below has been officially dropped.
 Multiple modules
 ................

+- The `__repr__()` method of all estimators (used when calling
+  `print(estimator)`) has been entirely re-written. :issue:`11705` by


"building on Python's pretty printing standard library."

NicolasHug · 2018-12-18T15:20:40Z

doc/whats_new/v0.21.rst

@@ -185,12 +185,19 @@ Support for Python 3.4 and below has been officially dropped.
  ``max_depth`` by 1 while expanding the tree if ``max_leaf_nodes`` and
  ``max_depth`` were both specified by the user. Please note that this also
  affects all ensemble methods using decision trees.
-  :pr:`12344` by :user:`Adrin Jalali <adrinjalali>`.
+  :issue:`12344` by :user:`Adrin Jalali <adrinjalali>`.


@adrinjalali I took the liberty to fix this

NicolasHug · 2018-12-18T15:21:40Z

I've added an example in set_config(). There's one in config_context() about assume_finite so I figured why not... But it's not necessarily ideal.

EDIT: well, actually, I removed it. It broke the other doctests. Could have made it work be setting again set_config(print_changed_only=False) but I guess that's an indication that this example does not belong here.

jnothman · 2018-12-19T02:19:56Z

We don't need another example of how to use config, we need somewhere in the user guide that explicitly describes what this feature can do for the user. An example in the gallery might be appropriate.

NicolasHug · 2018-12-19T12:04:12Z

Where do I reference this example from in the doc?

amueller · 2018-12-19T21:40:19Z

link it from the set_config docs maybe? I don't think there's a single section on that in the user guide, though...

NicolasHug · 2018-12-19T21:44:01Z

Isn't it automatically linked?

amueller · 2018-12-19T21:44:38Z

to the API docs, yes.

NicolasHug · 2018-12-19T22:03:46Z

set_config is referenced in in doc/modules/computing.rst but I doubt it makes sense there

jnothman

User guide could consider adding a section on working with scikit-learn in a terminal? But not in this PR.

jnothman · 2018-12-20T04:15:27Z

Note that the example is automatically referenced at the bottom of https://41116-843222-gh.circle-artifacts.com/0/doc/modules/generated/sklearn.set_config.html

jnothman · 2018-12-20T10:15:05Z

Yay! Thanks and well done @NicolasHug!!

GaelVaroquaux · 2018-12-20T11:04:01Z

Hurray! Thanks heaps!! Sorry for not giving feedback on this. Anyhow, I love it.

Any plans to add back the color work, that @amueller had originally implemented? If it could be enabled by default, it would help usability.

amueller · 2018-12-20T17:38:52Z

@GaelVaroquaux I'd love to see that, but I figured "small steps" would be better. This is already a big improvement imho.
I guess one question is whether we want something jupyter specific or something that also works on the console. I would probably go with jupyter and use html.

GaelVaroquaux · 2018-12-20T20:10:14Z

@GaelVaroquaux I'd love to see that, but I figured "small steps" would be better. This is already a big improvement imho.

+1!

* add pprint for estimators * strip color from length, add color option * Minor cleaning, fixes, factoring and docs * Added some basic tests * Fixed line length issue * fixed flake8 and added visual test for review * Fixed test * Fixed Python 2 issues (inspect.signature import) * Trying to fix flake8 again * Added special repr for functions * Added some other visual tests * Changed _format_function in to _format_callable because callable() returns True also for class objects (which we want to reprensent with their name as well anyway) * Consistent output in Python 2 and 3 * WIP * Now using the builtin pprint module * pep8 * Added changed_only param * Fixed printing when string would fit in less than line width * Fixed printing of steps parameter * Fixed changed_only param for short estimators * fixed pep8 * Added some more description in docstring * changed_only is now an option from set_config() * Put _pprint.py into sklearn/utils, added tests * Added doctest NORMALIZE_WHITESPACE where needed * Fixed tests * fix test-doc * fixing test that passed before.... * Fixed tests * Added test for changed_only and long lines * typo * Added authors names * Added license file * Added ellipsis based on number of elements in sequence + added increasinly aggressive repr strategies * Updated whatsnew * dont use increaingly aggressive strategy * Fixed tests * Removed LICENSE file and put license text in _pprint.py * fixed test_base * Sorted parameters dictionary for consistent output in 3.5 * Actually using OrderedDict... * Addressed comments * Added test for NaN changed parameter * Update whatsnew * Added example to set_config() * Removed example * Added example in gallery * Spelling

…)" This reverts commit 5f34303.

* add pprint for estimators * strip color from length, add color option * Minor cleaning, fixes, factoring and docs * Added some basic tests * Fixed line length issue * fixed flake8 and added visual test for review * Fixed test * Fixed Python 2 issues (inspect.signature import) * Trying to fix flake8 again * Added special repr for functions * Added some other visual tests * Changed _format_function in to _format_callable because callable() returns True also for class objects (which we want to reprensent with their name as well anyway) * Consistent output in Python 2 and 3 * WIP * Now using the builtin pprint module * pep8 * Added changed_only param * Fixed printing when string would fit in less than line width * Fixed printing of steps parameter * Fixed changed_only param for short estimators * fixed pep8 * Added some more description in docstring * changed_only is now an option from set_config() * Put _pprint.py into sklearn/utils, added tests * Added doctest NORMALIZE_WHITESPACE where needed * Fixed tests * fix test-doc * fixing test that passed before.... * Fixed tests * Added test for changed_only and long lines * typo * Added authors names * Added license file * Added ellipsis based on number of elements in sequence + added increasinly aggressive repr strategies * Updated whatsnew * dont use increaingly aggressive strategy * Fixed tests * Removed LICENSE file and put license text in _pprint.py * fixed test_base * Sorted parameters dictionary for consistent output in 3.5 * Actually using OrderedDict... * Addressed comments * Added test for NaN changed parameter * Update whatsnew * Added example to set_config() * Removed example * Added example in gallery * Spelling

amueller and others added 10 commits June 10, 2017 14:56

add pprint for estimators

acc3732

strip color from length, add color option

c99c85a

Minor cleaning, fixes, factoring and docs

a2ef6dc

Added some basic tests

fd543e8

Fixed line length issue

64932cd

fixed flake8 and added visual test for review

df281fb

Fixed test

05df6ad

Merge branch 'master' of https://github.com/scikit-learn/scikit-learn …

01db0c6

…into pr/9099

Fixed Python 2 issues (inspect.signature import)

89ee958

Trying to fix flake8 again

f6450db

NicolasHug added 4 commits August 16, 2018 18:37

Added special repr for functions

2079d78

Added some other visual tests

1a3c380

Changed _format_function in to _format_callable

cd39e8d

because callable() returns True also for class objects (which we want to reprensent with their name as well anyway)

Consistent output in Python 2 and 3

d45bd0a

NicolasHug added 4 commits September 10, 2018 10:24

WIP

30431a7

Now using the builtin pprint module

6a547d6

pep8

d82afee

Added changed_only param

4a98b5a

NicolasHug added 4 commits September 13, 2018 10:31

Fixed printing when string would fit in less than line width

5a64453

Fixed printing of steps parameter

4f8c450

Fixed changed_only param for short estimators

68d1806

fixed pep8

9afcd0b

jnothman approved these changes Dec 18, 2018

View reviewed changes

jnothman reviewed Dec 18, 2018

View reviewed changes

NicolasHug added 2 commits December 18, 2018 10:18

Update whatsnew

f2808a1

Added example to set_config()

ab639ae

NicolasHug commented Dec 18, 2018

View reviewed changes

Removed example

c48d713

Added example in gallery

4e06804

Spelling

affaae5

jnothman approved these changes Dec 19, 2018

View reviewed changes

amueller merged commit a50c03f into scikit-learn:master Dec 20, 2018

NicolasHug deleted the pr/9099 branch December 20, 2018 20:42

NicolasHug mentioned this pull request Mar 21, 2019

RFE/RFECV step enhancements #13470

Closed

xhluca pushed a commit to xhluca/scikit-learn that referenced this pull request Apr 28, 2019

Revert "[MRG] Add pprint for estimators - continued (scikit-learn#11705…

6353084

…)" This reverts commit 5f34303.

xhluca pushed a commit to xhluca/scikit-learn that referenced this pull request Apr 28, 2019

Revert "[MRG] Add pprint for estimators - continued (scikit-learn#11705…

cc402ec

…)" This reverts commit 5f34303.

RituRajSingh878 mentioned this pull request Mar 2, 2020

added repr in BaseEstimator from sklearn VowpalWabbit/vowpal_wabbit#2307

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[MRG] Add pprint for estimators - continued #11705

[MRG] Add pprint for estimators - continued #11705

NicolasHug commented Jul 29, 2018 •

edited

Loading

jnothman commented Jul 31, 2018 via email

amueller commented Aug 16, 2018

jnothman commented Sep 4, 2018 via email

NicolasHug commented Sep 11, 2018 •

edited

Loading

jnothman commented Sep 12, 2018

NicolasHug commented Sep 12, 2018

jnothman commented Sep 13, 2018 via email

NicolasHug commented Sep 13, 2018

jnothman left a comment

jnothman Dec 18, 2018

jnothman Dec 18, 2018

NicolasHug Dec 18, 2018

NicolasHug commented Dec 18, 2018 •

edited

Loading

jnothman commented Dec 19, 2018 via email

NicolasHug commented Dec 19, 2018

amueller commented Dec 19, 2018

NicolasHug commented Dec 19, 2018

amueller commented Dec 19, 2018

NicolasHug commented Dec 19, 2018

jnothman left a comment

jnothman commented Dec 20, 2018

jnothman commented Dec 20, 2018 via email

GaelVaroquaux commented Dec 20, 2018

amueller commented Dec 20, 2018

GaelVaroquaux commented Dec 20, 2018 via email

[MRG] Add pprint for estimators - continued #11705

[MRG] Add pprint for estimators - continued #11705

Conversation

NicolasHug commented Jul 29, 2018 • edited Loading

Reference Issues/PRs

What does this implement/fix? Explain your changes.

Any other comments?

Questions

jnothman commented Jul 31, 2018 via email

amueller commented Aug 16, 2018

jnothman commented Sep 4, 2018 via email

NicolasHug commented Sep 11, 2018 • edited Loading

jnothman commented Sep 12, 2018

NicolasHug commented Sep 12, 2018

jnothman commented Sep 13, 2018 via email

NicolasHug commented Sep 13, 2018

jnothman left a comment

Choose a reason for hiding this comment

jnothman Dec 18, 2018

Choose a reason for hiding this comment

jnothman Dec 18, 2018

Choose a reason for hiding this comment

NicolasHug Dec 18, 2018

Choose a reason for hiding this comment

NicolasHug commented Dec 18, 2018 • edited Loading

jnothman commented Dec 19, 2018 via email

NicolasHug commented Dec 19, 2018

amueller commented Dec 19, 2018

NicolasHug commented Dec 19, 2018

amueller commented Dec 19, 2018

NicolasHug commented Dec 19, 2018

jnothman left a comment

Choose a reason for hiding this comment

jnothman commented Dec 20, 2018

jnothman commented Dec 20, 2018 via email

GaelVaroquaux commented Dec 20, 2018

amueller commented Dec 20, 2018

GaelVaroquaux commented Dec 20, 2018 via email

NicolasHug commented Jul 29, 2018 •

edited

Loading

NicolasHug commented Sep 11, 2018 •

edited

Loading

NicolasHug commented Dec 18, 2018 •

edited

Loading