OS: macOS Monterey v12.2.1 Hardware: MacBook Pro (13-inch, M1, 2020) Chip: Apple M1 Python 3.9.10

AttributeError Traceback (most recent call last) Input In [22], in 10 warnings.simplefilter("ignore") # disable Keras warnings for this tutorial 11 import keras ---> 14 import eli5

File ~/miniforge3/envs/tf_gpu/lib/python3.9/site-packages/eli5/init.py:6, in 2 from future import absolute_import 4 version = '0.11.0' ----> 6 from .formatters import ( 7 format_as_html, 8 format_html_styles, 9 format_as_text, 10 format_as_dict, 11 ) 12 from .explain import explain_weights, explain_prediction 13 from .sklearn import explain_weights_sklearn, explain_prediction_sklearn

File ~/miniforge3/envs/tf_gpu/lib/python3.9/site-packages/eli5/formatters/init.py:9, in 2 """ 3 Functions to convert explanations to human-digestible formats. 4 5 TODO: IPython integration, customizability. 6 """ 8 from .text import format_as_text ----> 9 from .html import format_as_html, format_html_styles 10 try: 11 from .as_dataframe import ( 12 explain_weights_df, explain_weights_dfs, 13 explain_prediction_df, explain_prediction_dfs, 14 format_as_dataframe, format_as_dataframes, 15 )

File ~/miniforge3/envs/tf_gpu/lib/python3.9/site-packages/eli5/formatters/html.py:22, in 18 from .trees import tree2text 19 from .text_helpers import prepare_weighted_spans, PreparedWeightedSpans ---> 22 template_env = Environment( 23 loader=PackageLoader('eli5', 'templates'), 24 extensions=['jinja2.ext.with_']) 25 template_env.globals.update(dict(zip=zip, numpy=np)) 26 template_env.filters.update(dict( 27 weight_color=lambda w, w_range: format_hsl(weight_color_hsl(w, w_range)), 28 remaining_weight_color=lambda ws, w_range, pos_neg: (...) 33 format_decision_tree=lambda tree: _format_decision_tree(tree), 34 ))

File ~/miniforge3/envs/tf_gpu/lib/python3.9/site-packages/jinja2/environment.py:363, in Environment.init(self, block_start_string, block_end_string, variable_start_string, variable_end_string, comment_start_string, comment_end_string, line_statement_prefix, line_comment_prefix, trim_blocks, lstrip_blocks, newline_sequence, keep_trailing_newline, extensions, optimized, undefined, finalize, autoescape, loader, cache_size, auto_reload, bytecode_cache, enable_async) 360 self.policies = DEFAULT_POLICIES.copy() 362 # load extensions --> 363 self.extensions = load_extensions(self, extensions) 365 self.is_async = enable_async 366 _environment_config_check(self)

File ~/miniforge3/envs/tf_gpu/lib/python3.9/site-packages/jinja2/environment.py:117, in load_extensions(environment, extensions) 115 for extension in extensions: 116 if isinstance(extension, str): --> 117 extension = t.cast(t.Type["Extension"], import_string(extension)) 119 result[extension.identifier] = extension(environment) 121 return result

File ~/miniforge3/envs/tf_gpu/lib/python3.9/site-packages/jinja2/utils.py:149, in import_string(import_name, silent) 147 else: 148 return import(import_name) --> 149 return getattr(import(module, None, None, [obj]), obj) 150 except (ImportError, AttributeError): 151 if not silent:

AttributeError: module 'jinja2.ext' has no attribute 'with_'

Fix: fixes import error when importing jinja2 > 3.0.3, requirements & setup are also updated.

Tested it with jinja2 = 3.0.0

Also tested it with jinja2 < 3.0.0 (2.11.3), but this resulted in an error not related to the fix. So support for the lowest possible version of jinja2 is not impacted.

• drop py36-legacy, as now py27 works well enough with most libraries already dropping support for it
• restrict TF and Keras versions (we support only TF 1.x)
• test skelarn-crfsuite only under python 2.7, as it does not work with newer sklearn

Hello community,

I am using eli5 library for image explanations, I am referring this tutorial, but at the end I am facing estimator is not supported, Please help me to solve this error,

any help is highly appreciated, @lopuhin

Thanks!

hey, i'm trying to pair RFECV with permutation Importance, but without cross-validation. I tried several approaches but I get always a NotFittedError for the estimator_func while trying to fit the rfecv. Am i missing something?

   ``` 

estimator_funct = estimator_funct.fit(extract_relevant_features, choosen_target)

pi = PermutationImportance(estimator_funct,  scoring='r2', n_iter=10, random_state=1).fit(extract_relevant_features, choosen_target)

rfecv = RFECV(
    estimator=pi,
    step=1,
    cv=cv_func,
    scoring=score,
    min_features_to_select=min_features_to_select,
)
#
rfecv.fit(extract_relevant_features,
          choosen_target,
          groups=extract_relevant_features.index) ```python

When i pair RFECV with CV- PermutationImportance, everything works fine, but i would like to have the non-cv version.

I just pip installed eli5 0.13.0 on a new AzureML instance and it is throwing an exception during "import eli5"

The sklearn.feature_selection.base module is deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.feature_selection. Anything that cannot be imported from sklearn.feature_selection is now part of the private API. Using TensorFlow backend.

AttributeError Traceback (most recent call last) Input In [59], in <cell line: 1>() ----> 1 import eli5

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/eli5/init.py:93, in 89 pass 92 try: ---> 93 from .keras import ( 94 explain_prediction_keras 95 ) 96 except ImportError: 97 # keras is not available 98 pass

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/eli5/keras/init.py:3, in 1 # -- coding: utf-8 -- ----> 3 from .explain_prediction import explain_prediction_keras 4 from .gradcam import gradcam, gradcam_backend

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/eli5/keras/explain_prediction.py:8, in 5 import PIL 7 import numpy as np ----> 8 import keras 9 import keras.backend as K 10 from keras.models import Model

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/init.py:25, in 22 from keras import distribute 24 # See b/110718070#comment18 for more details about this import. ---> 25 from keras import models 27 from keras.engine.input_layer import Input 28 from keras.engine.sequential import Sequential

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/models.py:19, in 16 """Code for model cloning, plus model-related API entries.""" 18 import tensorflow.compat.v2 as tf ---> 19 from keras import backend 20 from keras import metrics as metrics_module 21 from keras import optimizer_v1

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/backend/init.py:1, in ----> 1 from .load_backend import epsilon 2 from .load_backend import set_epsilon 3 from .load_backend import floatx

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/backend/load_backend.py:90, in 88 elif _BACKEND == 'tensorflow': 89 sys.stderr.write('Using TensorFlow backend.\n') ---> 90 from .tensorflow_backend import * 91 else: 92 # Try and load external backend. 93 try:

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/backend/tensorflow_backend.py:25, in 22 import numpy as np 23 from distutils.version import StrictVersion ---> 25 from ..utils.generic_utils import transpose_shape 27 py_all = all 28 py_any = any

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/utils/generic_utils.py:415, in 410 else: 411 return obj.name 414 @tf_contextlib.contextmanager --> 415 def skip_failed_serialization(): 416 global _SKIP_FAILED_SERIALIZATION 417 prev = _SKIP_FAILED_SERIALIZATION

File /anaconda/envs/azureml_py38/lib/python3.8/site-packages/keras/utils/tf_contextlib.py:33, in contextmanager(target) 23 """A tf_decorator-aware wrapper for contextlib.contextmanager. 24 25 Usage is identical to contextlib.contextmanager. (...) 30 A callable that can be used inside of a with statement. 31 """ 32 context_manager = _contextlib.contextmanager(target) ---> 33 return tf.internal.decorator.make_decorator(target, context_manager, 'contextmanager')

AttributeError: module 'tensorflow.compat.v2' has no attribute 'internal'

Hi folks,

I just started looking into this really library. One thing that makes me curious is, why isn't there an api likeeli5.explain_prediction but for multiple predictions? It would be nice to be able to provide a slice of inference data and have the method tell you the overall contribution of of the features in the sample

Hi,

I would like to make Lime colouring a bit less sensitive: highlight only words that have weights above a specific threshold.

None of the arguments of show_prediction enables to fix this threshold.

@kmike, @lopuhin @zzz4zzz. Is there a way?

A possible solution could be to add more opacity to the words where weights are not too large.