Deepchecks is a Python package for comprehensively validating your machine learning models and data with minimal effort

Overview

Test Suites for Validating ML Models & Data

Join Slack   |   Documentation   |   Blog   |   Twitter

build Documentation Status pkgVersion pyVersions Maintainability Coverage Status

Deepchecks is a Python package for comprehensively validating your machine learning models and data with minimal effort. This includes checks related to various types of issues, such as model performance, data integrity, distribution mismatches, and more.

Installation

Using pip

pip install deepchecks -U --user

Using conda

conda install -c deepchecks deepchecks

Try it Out!

Head over to the Quickstart Notebook and choose the binder badge image or the colab badge image to have it up and running, and to then apply it on your own data and models.

Usage Examples

Running a Suite

A Suite runs a collection of Checks with optional Conditions added to them.

To see it in action, we recommend trying it out.

To run an existing suite all you need to do is to import the suite and run it with the required (suite-dependent) input parameters. The list of all built-in suites can be found here.

Let's take the "iris" dataset as an example

from sklearn.datasets import load_iris
iris_df = load_iris(return_X_y=False, as_frame=True)['frame']

and run the single_dataset_integrity suite, which requires only a single Dataset and can run also directly on a pd.DataFrame, like in the following example.

from deepchecks.suites import single_dataset_integrity
suite = single_dataset_integrity()
suite.run(iris_df)

Will result in printing the suite's output, that starts with a summary of the check conditions

Single Dataset Integrity Suite

The suite is composed of various checks such as: Mixed Data Types, Is Single Value, String Mismatch, etc...
Each check may contain conditions (which will result in pass / fail / warning, represented by / / ! ), as well as other outputs such as plots or tables.
Suites, checks and conditions can all be modified (see the Create a Custom Suite tutorial).


Conditions Summary

Status Check Condition More Info
Single Value in Column - Test Dataset Does not contain only a single value for all columns Columns containing a single value: ['target']
!
Data Duplicates - Test Dataset Duplicate data is not greater than 0% Found 2.00% duplicate data
Mixed Nulls - Test Dataset Not more than 1 different null types for all columns
Mixed Data Types - Test Dataset Rare data types in all columns are either more than 10.00% or less than 1.00% of the data
String Mismatch - Test Dataset No string variants for all columns
String Length Out Of Bounds - Test Dataset Ratio of outliers not greater than 0% string length outliers for all columns
Special Characters - Test Dataset Ratio of entirely special character samples not greater than 0.10% for all columns

Followed by the visual outputs of all of the checks that are in that suite, that isn't appended here for brevity. In the following section you can see an example of how the output of a single check may look.

Running a Check

To run a specific single check, all you need to do is import it and then to run it with the required (check-dependent) input parameters. More details about the existing checks and the parameters they can receive can be found in our API Reference.

from deepchecks.checks import TrainTestFeatureDrift
import pandas as pd

train_df = pd.read_csv('train_data.csv')
train_df = pd.read_csv('test_data.csv')
# Initialize and run desired check
TrainTestFeatureDrift().run(train_data, test_data)

Will produce output of the type:

Train Test Drift

The Drift score is a measure for the difference between two distributions, in this check - the test and train distributions.
The check shows the drift score and distributions for the features, sorted by feature importance and showing only the top 5 features, according to feature importance. If available, the plot titles also show the feature importance (FI) rank.

Key Concepts

Check

Each check enables you to inspect a specific aspect of your data and models. They are the basic building block of the deepchecks package, covering all kinds of common issues, such as:

  • Model Error Analysis
  • Label Ambiguity
  • Data Sample Leakage

and many more checks.

Each check can have two types of results:

  1. A visual result meant for display (e.g. a figure or a table).
  2. A return value that can be used for validating the expected check results (validations are typically done by adding a "condition" to the check, as explained below).

Condition

A condition is a function that can be added to a Check, which returns a pass ✓, fail or warning ! result, intended for validating the Check's return value. An example for adding a condition would be:

from deepchecks.checks import BoostingOverfit
BoostingOverfit().add_condition_test_score_percent_decline_not_greater_than(threshold=0.05)

which will return a check failure when running it if there is a difference of more than 5% between the best score achieved on the test set during the boosting iterations and the score achieved in the last iteration (the model's "original" score on the test set).

Suite

An ordered collection of checks, that can have conditions added to them. The Suite enables displaying a concluding report for all of the Checks that ran. See the list of predefined existing suites to learn more about the suites you can work with directly and also to see a code example demonstrating how to build your own custom suite. The existing suites include default conditions added for most of the checks. You can edit the preconfigured suites or build a suite of your own with a collection of checks and optional conditions.

What Do You Need in Order to Start Validating?

Environment

  • The deepchecks package installed
  • JupyterLab or Jupyter Notebook

Data / Model

Depending on your phase and what you wish to validate, you'll need a subset of the following:

  • Raw data (before pre-processing such as OHE, string processing, etc.), with optional labels
  • The model's training data with labels
  • Test data (which the model isn't exposed to) with labels
  • A model compatible with scikit-learn API that you wish to validate (e.g. RandomForest, XGBoost)

Deepchecks validation accompanies you from the initial phase when you have only raw data, through the data splits, and to the final stage of having a trained model that you wish to evaluate. Accordingly, each phase requires different assets for the validation.

See more about typical usage scenarios and the built-in suites in the docs.

Documentation

Community

  • Join our Slack Community to connect with the maintainers and follow users and interesting discussions
  • Post a Github Issue to suggest improvements, open an issue, or share feedback.
Comments
  • [FEAT] Tabular fairness suite

    [FEAT] Tabular fairness suite

    Describe the solution you'd like

    (As discussed briefly with @noamzbr.)

    A suite of checks for assessing the fairness of classification and regression models. Proposed fairness checks:

    • Performance Disparity: Check that there are no major performance disparities across a set of pre-defined subgroups (e.g., across sex and race). For certain choices of performance measures, this is equivalent to the equalized odds fairness metric in classification.
    • Demographic Parity: Check that there are no major differences in the response distribution across subgroups.
    • Representativity: Check that pre-defined subgroups are sufficiently well represented in the data (i.e., make sure that no subgroup of interest is underrepresented in the data).

    These checks should implement functionality comparable to what is other toolkits such as Microsoft's Responsible AI Fairness Dashboard. It can also go much beyond that by:

    1. Evaluating fairness at the intersection of multiple subgroups and allowing for control variables.
    2. Providing high-quality visualizations for evaluating fairness across many subgroups and control variables.
    3. Implementing a broad set of fairness metrics.
    4. Accounting for sampling uncertainty when comparing performance metrics across subgroups.

    I think (4) is out of score of deepchecks, but (1)-(3) are achievable.

    Describe alternatives you've considered

    The SegmentPerformance check looks at performance within subgroups. However, it is now deprecated and it is limited to looking at only two subgroups. Furthermore, it does not provide default conditions that are suited to fairness evaluation. Same problem with WeakSegmentsPerformance which isn't well-suited for fairness analysis.

    Additional context

    What could this look like?

    class PerformanceDisparityReport(SingleDatasetCheck):
    
        # List of protected attributes (e.g., sex and age)
        target_features: List[Hashable]
    
        # Optionally, a single variable to use as a control (e.g., education level)
        control_feature: Hashable = None
    
        # Scoring funtions
        scorers: Union[Mapping[str, Union[str, Callable]], List[str]] = None
        
        # For classification, whether to report scores as class averages.
        class_avg: bool = True
    
        # Each feature (in target_features and control_feature) is to be discretized in at most `max_segments` groups.
        max_segments: int = 10
    
        # Smallest group size on which a performance metric is going to be computed
        min_group_size: int = 5
    
        # Visual attributes to use for sensitive variables. Defaults to ["x", "color", "facet_col"]
        facets: List[str] = None
    
        # Visual attribute to use for performance dis-aggregated by class. Defaults to "facet_col".
        class_facet: str = "facet_col"
    

    How can we visualize performance at the intersection of many subgroups?

    I would recommend using a strip plot to visualize scores across subgroups. An histogram of subgroup size can be overlayed to provide context.

    For example, using the adult dataset and "sex", "race" and "age" as sensitive features, we can visualize performance by class and subgroup. Only the first two subgroups (sex and race) will be assigned distinct visual elements (on the x-axis and color), but there will be one point for each corresponding age group as well.

    from deepchecks.tabular.checks.model_evaluation.performance_disparity_report import PerformanceDisparityReport
    from deepchecks.tabular.datasets.classification import adult
    
    check = PerformanceDisparityReport(target_features=["sex", "race", "age"], max_segments=3, scorers=["f1_per_class"])
    
    dataset = adult.load_data()[0]
    model = adult.load_fitted_model()
    check.run(dataset=dataset, model=model)
    

    screenshot

    What conditions could be added?

    We can threshold the difference between min and max performance across subgroups, or the ratio between min and max performance across subgroups. We can also put a lower boundary on performance for all subgroups.

    Why only allow for a single control feature?

    Multivariate control variables can be reduced to a univariate variable by regressing on the outcome. I think this is sufficient for fairness analyses and it keeps things simple enough.

    What else could be considered?

    What I've proposed so far are group fairness checks (see https://en.wikipedia.org/wiki/Fairness_(machine_learning)). Individual or counterfactual fairness checks could also be considered. Additionally, more general definitions of group fairness based on independence and conditional independence between variables could be assessed using statistics such as a $\chi^2$ divergence, KL-divergence, or conditional entropy. It's less easily interpretable than what I proposed, but it's more principled and it's amenable to formal statistical tests.

    ds tabular 
    opened by OlivierBinette 29
  • [FEAT] Add option to save as md files - cml integration

    [FEAT] Add option to save as md files - cml integration

    Background cml is a ci tool which for our purposes sends md files as comments in PRs. (its features as a runner is irrelevant).

    Is your feature request related to a problem? Please describe. We can currently save checks and suites as html, we need to add an option to save as md files too. as for now we can send just the condition table easily but it is not really helpful.

    Implementation Details Mostly we need to remove any style blocks in the html and save the plots as an image (this part might make us use some git giles api).

    feature 
    opened by JKL98ISR 23
  • [issue-1077] Display logic refactoring

    [issue-1077] Display logic refactoring

    resolves #1077 resolves #627

    As was said in the issue description currently, the code responsible for the check&suite displays is not clean and very hard to maintain. From my point of view in order to clean it up we need to do next things:

    1. Decouple display logic from CheckResult and SuiteResult
    2. Decouple different display approaches from each other (html, widget)

    I suggest the next set of the abstractions in order to achieve this

    
    # Serializer - class that implements serialization logic for the value of type `T` into some particular format 
    
    class Serializer(t.Protocol[T]):
        value: T
        def __init__(self, value: T, **kwargs):
            self.value = value
    
    class HtmlSerializer(Serializer[T]):
        def serialize(self, **kwargs) -> str:
            ...
    
    class JsonSerializer(Serializer[T]):
        def serialize(self, **kwargs) -> t.Dict[t.Any, t.Any]:
            ...
    
    class WidgetSerializer(Serializer[T]):
        def serialize(self, **kwargs) -> Widget:
            ...
    
    # ============
    
    # Presentation - a facade class for a set of serializers, it should not contain 
    # implementation of a serialization logic, its main task is to instantiate instances 
    # of serializers and to provide an interface to them. 
    
    class Presentation(t.Protocol[T]):
        value: T
    
        def __init__(self, value: T, **kwargs):
            self.value = value
    
        def to_json(self, **kwargs) -> t.Dict[t.Any, t.Any]:
            raise NotImplementedError()
    
        def to_html(self, **kwargs) -> str:
            raise NotImplementedError()
    
        def to_widget(self, **kwargs) -> Widget:
            raise NotImplementedError()
    
    # ============
    
    # It is also totally logical to add a `Deserializer` type that will contain 
    # logic opposite to the logic  `Serializer` but currently I am not sure about this idea
    

    For more details take a look at the code.

    Note: currently this PR is in the draft state but it already contains basic logic for CheckResult and SuiteResult

    feature 
    opened by yromanyshyn 16
  • Cml markdown

    Cml markdown

    Reference Issues/PRs

    Resolve #1290

    What does this implement/fix? Explain your changes.

    This PR implements saving a report as a GitHub- or GitLab- compliant markdown file for ease of use in CI Pipelines. The use case is to use the command:

    cml comment create report_made_by_deepchecks.md --publish --publish-native
    

    And have a nice summary being posted in a PR made by the data scientist that shows a summary table of the checks' conditions with an attachment to the full html report (and only this attachment includes nice plotly visualizations).

    Any other comments?

    As of the opening of this PR, the commit dd20560 is meant to kickstart the discussion around the code. Namely,

    • whether I should write a specific Serializer for this feature
    • tips for writing the test of this feature is greatly appreciated.

    TODO

    • [ ] ~~zip suite results .html for GitHub, as GitHub does not accept html file attachments.~~
      • CML is supposed to handle that.
    • [x] write documentation under "Integrations" in a new CML page (under docs/source/user-guide/integrations)
    • [x] make pylint
    feature 
    opened by bgalvao 13
  • [issue-1365] plotly figures width/height

    [issue-1365] plotly figures width/height

    resolves #1365

    I have experimented with height/width options, and noticed that:

    • yup, it is Indeed responsive to the window/screen size
    • it looks like by default (without explicitly provided width/height) it uses something like - width: 100%; height: max-content;. Added screenshots below to illustrate this.

    IMO, in some cases more compact figure (with explicit width/height) could look better/ It all depends on the situation

    also, should I attach screenshots of the affected checks?

    -- with width=800, height=500 -- image

    -- without explicit width/height -- image

    opened by yromanyshyn 13
  • [BUG] Some of the links in README file not working or redirecting properly

    [BUG] Some of the links in README file not working or redirecting properly

    Describe the bug Some of the links in the README file not working or redirecting properly.

    1. Below link does not redirect to the desired destination:

    Link Redirection Issue

    1. Below link does not work:

    Link not working

    Expected behavior

    1. Link should have redirected to the When Should You Use Deepchecks? section of the Getting Started page of deepchecks documentation. Instead it just redirects to the top of the Getting Started page.
    2. Link should redirect to either Which Types of Checks Exist? section of the Getting Started page of deep checks documentation or the README should be changed to point to Tabular Checks and Vision Checks page of deepchecks documentation.
    documentation bug 
    opened by kishore-s-15 11
  • [MLC-153] new dataset column type

    [MLC-153] new dataset column type

    There are two things that still need to be done (TODO):

    We need to verify all checks, that there is no function/method that tries to perform string operation over a column name without first checking that it is actually a string. Probably it is a good idea to add a test case to those checks for which this is actual.

    A lot of routines in our code-base use the next dtype for the columns list argument - Union[Hashable, Iterable[Hashable]]. For example LabelAmbiguity check.

    It raises a question - What if the provided value is hashable but also iterable? like tuple for example I think it is better to change those routines signature just to Iterable[Hashable]

    @noamzbr @matanper @ItayGabbay what do you think?

    opened by yromanyshyn 11
  • Implement PerformanceDisparityReport

    Implement PerformanceDisparityReport

    Reference Issues/PRs

    Implements PerformanceDisparityReport (see #2208).

    Any other comments?

    See https://github.com/OlivierBinette/PerformanceDisparityReport-test/blob/main/test_plots.ipynb for plot tests

    A few potential features have not yet been implemented:

    1. Alternative choice of baseline score (best possible score or best subgroup score).
    2. Alternative ranking by relative score difference
    3. Allowing lists of protected_feature and control_feature
    4. Allowing multiple scorers (there could be a tab for each scorer).

    Note: I wasn't able to run make test or make tox locally (I get the error "Could not find a version that satisfies the requirement torch==1.10.2+cpu"). I'm opening the PR to trigger CI/CD and there will probable be a few issues to fix with the tests.

    feature 
    opened by OlivierBinette 10
  • fix(docs): Minor grammar suggestions

    fix(docs): Minor grammar suggestions

    Reference Issues/PRs

    N/A

    What does this implement/fix? Explain your changes.

    Makes some suggestions to improve grammar in the documentation

    Any other comments?

    I think Deepchecks is super cool and am just trying to help, even if just a wee little bit😁


    Thanks for contributing a pull request! Please ensure you have taken a look at the contribution guidelines: https://github.com/deepchecks/deepchecks/blob/main/CONTRIBUTING.rst

    documentation feature 
    opened by jhwohlgemuth 10
  • QST: Question regarding documentation

    QST: Question regarding documentation

    Research

    Link to question on StackOverflow

    https://docs.deepchecks.com/en/stable/user-guide/when_should_you_use.html#when-should-you-use-new-data

    Question about deepchecks

    Hi, Based on your documentation page => https://docs.deepchecks.com/en/stable/user-guide/when_should_you_use.html#when-should-you-use-new-data

    There are 4 phases : 1)New Data: Single Dataset Validation -> For these purposes you can use the single_dataset_integrity() 2)After Splitting the Data: Train-Test Validation -> For these purposes you can use the -> train_test_validation() 3)After Training a Model: Analysis & Validation ->model_evaluation() 4)General Overview: Full Suite -> full_suite()

    The question where can I see the examples of using this 4 options ?

    Thanks, Boris

    question 
    opened by borisRa 10
  • [BUG] Format precentage in new category output

    [BUG] Format precentage in new category output

    Describe the bug CleanShot 2022-07-14 at 10 38 05@2x

    See "percent of new category in sample"

    To Reproduce Use to following dataset: https://github.com/AllonHammer/CPI_HRNN/blob/master/resources/cpi_us_dataset.csv Dataset definition: ds = Dataset(df, datetime_name='Date', label='Price', cat_features=['Category_id', 'Category', 'Indent', 'Parent', 'Parent_ID']) Split first 40000 samples to be train, and the rest to be test. And run the relevant checks (or the train-test-validation suite)

    bug 
    opened by shir22 9
  • [BUG] fix progress bar / remove it

    [BUG] fix progress bar / remove it

    Describe the bug Screen Shot 2023-01-02 at 16 48 54

    To reproduce: run the following `from deepchecks.vision.datasets.detection import mask from deepchecks.vision.suites import data_integrity, train_test_validation, model_evaluation, full_suite from deepchecks.vision import VisionData

    train_data = mask.load_dataset(object_type="VisionData", shuffle=False) test_data = mask.load_dataset(object_type="VisionData", shuffle=False)

    N_SAMPLES = 1000 suite = model_evaluation(n_samples = N_SAMPLES) suite.run(train_dataset=train_data,test_dataset=test_data,max_samples=N_SAMPLES) `

    bug vision needs triage 
    opened by Nadav-Barak 0
  • [FEAT] seperate conftest of vision, tabular, nlp

    [FEAT] seperate conftest of vision, tabular, nlp

    Currently all tests use all the conftests, which is a. un efficient b. causes error in some cases - for example we cannot test python 3.6 only on the tabular sub package c. ugly

    This is also partly related to https://github.com/deepchecks/deepchecks/issues/2240

    refactoring needs triage 
    opened by Nadav-Barak 0
  • Separate vision and tabular module to namespaced packages

    Separate vision and tabular module to namespaced packages

        So, we don't support vision on Python 3.6?
    

    That's a good reason to separate the package into namespaced packages.

    Originally posted by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/2234#pullrequestreview-1233539158

    breaking-backwards-compatabillity feature 
    opened by ItayGabbay 0
  • Multivariate drift does not show drift bar if feature importance was not calculated

    Multivariate drift does not show drift bar if feature importance was not calculated

    In multivariate drift (in all submodules), if feature importance was not calculated, the individual graphs should not be shown. However, the drift bar graph above them should show and now it doesn't.

    bug 
    opened by nirhutnik 0
  • [FEAT] Control charts for temporal datasets

    [FEAT] Control charts for temporal datasets

    Is your feature request related to a problem? Please describe.

    I'd like to be able to validate future data based on what old data looked like, and flag outlying values. The goal is to catch errors and process failures, not monitor slow drift.

    Examples:

    • Suppose I have a list of historical performance of a model. I'd like to check if this year's model performance is in the same range as before or if something has changed. I don't care much about gradual changes to the performance. Rather, I want to detect sudden changes that might indicate a bug or major failure.
    • Suppose I have a large dataset that is updated in batches (of size 1+) over time. At each new batch, I'd like to validate that the data falls within the range expected from past data. Again, the goal is to catch bugs, errors, and invalid data, not to detect slow drift over time.

    Describe the solution you'd like

    I'd like to use control charts to do this kind of monitoring. Learn a model from past data, get what you think new data should look like, and then compare the new data to the expectation.

    The implementation could look like this:

    check = ControlChart("performance_score", timedate="timedate_column", check_range="latest")
    check.add_condition_statistical_control(confidence=0.99) # Only flag data falling outside of the 99% confidence interval.
    check.add_condition_limits(range=(0.8, 0.9))  # Check that latest performance is between 0.8 and 0.9.
    check.run(dataset)  # Check that the latest performance score is as expected.
    

    The name "control chart" could be replaced by "process monitoring" or something more explicit.

    Describe alternatives you've considered

    Some of this is already done in train/test validation checks and there's already an outlier detection check. However, the goal is to detect drift rather than to check for errors given incremental changes.

    I think that some form of control charts or incremental outlier detection is necessary for Deepchecks to be used as part of a monitoring pipeline.

    needs triage 
    opened by OlivierBinette 0
Releases(0.10.0)
  • 0.10.0(Dec 11, 2022)

    What's Changed

    • Fix columns info data integrity call to return all columns by adding None to 'column_importance_sorter_dict' by @alphabetagamer in https://github.com/deepchecks/deepchecks/pull/2120
    • 1945 additional property texture level by @TheSolY in https://github.com/deepchecks/deepchecks/pull/2128
    • Enable getting model predictions using only the proba and the model_classes by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2126
    • 2095 deepcheck vision metrics access by @TheSolY in https://github.com/deepchecks/deepchecks/pull/2127
    • docs: add alphabetagamer as a contributor for code, and bug by @allcontributors in https://github.com/deepchecks/deepchecks/pull/2129
    • image properties linton by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/2123
    • 1721 update cv quickstarts remove reference to full suite by @TheSolY in https://github.com/deepchecks/deepchecks/pull/2132
    • Add mask dataset by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2130
    • change time_index to day_index by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2136
    • Fix the x zoom in edge cases of drift by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2138
    • [Fix: 2110] Pass feature_importance_timeout through to feature importance by @IngoStatworx in https://github.com/deepchecks/deepchecks/pull/2121
    • docs: add IngoStatworx as a contributor for code, and bug by @allcontributors in https://github.com/deepchecks/deepchecks/pull/2140
    • Add an util to display the wolf husky dataset by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2139
    • Ignore nans completely in new category check by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2143
    • Fix _do_static_prop by @matanper in https://github.com/deepchecks/deepchecks/pull/2145
    • Fix warning in task_inference.py by @matanper in https://github.com/deepchecks/deepchecks/pull/2149
    • Greater is better by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2155
    • Bug fix ROC report + some code cleanup by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/2154
    • docstring rule all by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/2157
    • Remove reduce from TrainTestPerformance by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2156
    • 1780 vision weak segment performance by @TheSolY in https://github.com/deepchecks/deepchecks/pull/2148
    • Add return statement to the validate_format function by @osw282 in https://github.com/deepchecks/deepchecks/pull/2133
    • docs: add osw282 as a contributor for code by @allcontributors in https://github.com/deepchecks/deepchecks/pull/2162
    • Make CI pipeline run also on major feature branches by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2167
    • Fix debugpy license check by @matanper in https://github.com/deepchecks/deepchecks/pull/2180
    • Cml markdown by @bgalvao in https://github.com/deepchecks/deepchecks/pull/2147
    • fixing test-windows by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/2178
    • Fix cat segmentation by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2163
    • Update all contributors badge by @matanper in https://github.com/deepchecks/deepchecks/pull/2183
    • docs: add bgalvao as a contributor for code, and doc by @allcontributors in https://github.com/deepchecks/deepchecks/pull/2182
    • Created junit serialization capability by @rcwoolston in https://github.com/deepchecks/deepchecks/pull/2159
    • docs: add rcwoolston as a contributor for code, doc, and ideas by @allcontributors in https://github.com/deepchecks/deepchecks/pull/2164
    • [DOCS] fix readme and update badge template by @shir22 in https://github.com/deepchecks/deepchecks/pull/2187
    • Fix tests runner by @matanper in https://github.com/deepchecks/deepchecks/pull/2188
    • Minor import and typo fixes by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/2191
    • feature_aggregation_method_argument is shared, so it shouldn't say drift by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2186
    • check config update by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/2192
    • add_include_defaults_to_other_checks by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/2194

    New Contributors

    • @alphabetagamer made their first contribution in https://github.com/deepchecks/deepchecks/pull/2120
    • @IngoStatworx made their first contribution in https://github.com/deepchecks/deepchecks/pull/2121
    • @osw282 made their first contribution in https://github.com/deepchecks/deepchecks/pull/2133
    • @bgalvao made their first contribution in https://github.com/deepchecks/deepchecks/pull/2147
    • @rcwoolston made their first contribution in https://github.com/deepchecks/deepchecks/pull/2159

    Full Changelog: https://github.com/deepchecks/deepchecks/compare/0.9.2...0.10.0

    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.10.0-py3-none-any.whl(3.41 MB)
    deepchecks-0.10.0.tar.gz(3.22 MB)
    LICENSE(34.03 KB)
  • 0.9.2(Oct 31, 2022)

    Commits

    • e275a0f: 1743 feat pass scorer list to model evaluation suite (#2101) (Sol) #2101
    • 79cbbf1: Plotly relax requirements (#2113) (JKL98ISR) #2113
    • d6e290c: 1852 single dataset static predictions (#2111) (Sol) #2111
    • 1b5bac9: regression_error_distribution makeover (#2112) (Nadav Barak) #2112
    • 49ffc01: Fix scorer per class bug (#2107) (matanper) #2107
    • e16e357: Merge branch 'main' of https://github.com/deepchecks/deepchecks into 0.9.x (JKL98ISR)
    • 9b3abe9: bump (JKL98ISR)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.9.2-py3-none-any.whl(3.39 MB)
    deepchecks-0.9.2.tar.gz(3.21 MB)
    LICENSE(34.03 KB)
  • 0.9.1(Oct 25, 2022)

    Bug Fixes

    Documentation

    Commits

    • 98925d9: 0.9 hf (#2025) (Itay Gabbay) #2025
    • e1954e7: [DOCS] Added documentation for tabular Model Evaluation checks (#1999) (DanBasson) #1999
    • f07835a: [DOCS] Added documentation for tabular Model Evaluation checks (#1999) (DanBasson)
    • 93f4be1: [FEAT] new class imbalance check (#2015) (DanBasson) #2015
    • 30b93e3: Static properties enum fix (#2032) (Sol) #2032
    • 1e5057b: ability to pass classes argument + auto infer to context (#2026) (Nadav-Barak) #2026
    • 72d73e4: [DOCS] Test for adding "all-contributors" to readme (#2035) (shir22) #2035
    • fe53b50: [DOCS] all-contributors with first contributor (#2036) (allcontributors[bot]) #2036
    • df89962: [DOCS] README - add one line of maintainers to all-contributors [skip ci] (#2040) (allcontributors[bot]) #2040
    • 7b379e6: [DOCS] Omit double sentence in property label correlation check examples (#2033) (shir22) #2033
    • 90426dd: [DOCS] add second line of all-contributors [skip ci] (#2045) (allcontributors[bot]) #2045
    • ea612a8: deepchecks scorer 2.0 (#2050) (Nadav-Barak) #2050
    • e07a56a: Add favicon to suite html output (#2049) (Noam Bressler) #2049
    • ea1e839: Fix certifi (Noam Bressler)
    • b7b99f5: Add additional maintainers to readme with all-contributors (#2060) (allcontributors[bot]) #2060
    • 377dbee: Fix add_condition_train_test_relative_degradation_less_than for Regression (#2058) (Noam Bressler) #2058
    • 3b7fe83: [DOCS] Add contributors to readme with all-contributors, #1 out of 2 (allcontributors[bot]) #2071
    • 23a7258: image property drift static properties fix + properties conf fix (#2059) (JKL98ISR) #2059
    • 74131db: [DOCS] Add contributors to readme with all-contributors, #2 out of 2 (#2072) (allcontributors[bot]) #2072
    • 9f24c20: Fix typo in makefile (#2082) (Ikko Ashimine) #2082
    • 989eddb: Fix tabular class imbalance docs (#2085) (DanBasson) #2085
    • 69fe8d1: [DOC] add eltociear as a contributor for doc (#2086) (allcontributors[bot]) #2086
    • 4603318: 1984 detection scorers averaging (#2079) (Sol) #2079
    • 66fca52: [DOCS] Update readme to markdown [skip ci] (#2088) (shir22) #2088
    • 5a724e4: Make deepchecks work with classification tasks when no proba is given (#2081) (Noam Bressler) #2081
    • 86ac003: calc_default_image_properties (#2087) (Sol) #2087
    • 5ec7a99: Small changes in Class Imbalance texts (#2089) (Nir Hutnik) #2089
    • 4a55424: 2027 single dataset check select dataset (#2091) (Sol) #2091
    • 0ac815b: Remove warning (#2094) (Nadav-Barak) #2094
    • 210e774: [DOCS] Update .all-contributorsrc to readme.md (#2098) (shir22) #2098
    • f1cd58f: iou range parameter in mean average precision (#2096) (Sol) #2096
    • b7d5d1b: Nb/feat/airbnb dataset (#2100) (Nadav Barak) #2100
    • 00313ec: 1867 doc for static properties (#2097) (Sol) #2097
    • e40c536: adult dataset no split fixed issue with the label (#2102) (Sol) #2102
    • 0450488: Fix len() on None in single dataset performance (#2103) (matanper) #2103
    • 80e3659: updated warning comment (#2104) (Nadav Barak) #2104
    • 1b91244: Merge branch 'main' of https://github.com/deepchecks/MLChecks into 0.9.x (Itay Gabbay)
    • 5d5c35c: Bumping to 0.9.1 (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.9.1-py3-none-any.whl(3.39 MB)
    deepchecks-0.9.1.tar.gz(3.20 MB)
    LICENSE(34.03 KB)
  • 0.9.0(Sep 19, 2022)

    What's Changed

    • remove 'actual' kwarg from testing by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1891
    • fix feature order in TrainTestFeatureDrift if a column is selected by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1890
    • Add semantic segmentation by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1883
    • Fixing feat imp to skip running feature importance with static predictions by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1896
    • add details for columns_info check by @DanBasson in https://github.com/deepchecks/deepchecks/pull/1894
    • [Issue-1739] Suite/check instances serialization/deserialization to/from config by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1885
    • l2 combination feature drift by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1898
    • [Issue-1846] fixed CheckResult.priority property by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1915
    • Nb/bug/small fixes by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1916
    • bug fixed by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1911
    • Documentation/static predictions vision by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1910
    • Vision Properties v2 by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1919
    • Remove percent change from display of label correlation checks by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1924
    • Add option to sort columns in drift according to drift + importance by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1922
    • Fixed truncate_zero_percent by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1925
    • 1608 property label correlation check by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1897
    • Categorical drift methods update by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1923
    • [issue-1917] result.show_in_window fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1926
    • [DOC] Consistent structure for tabular and vision checks #1893 by @DanBasson in https://github.com/deepchecks/deepchecks/pull/1921
    • [FEAT] Add package installation explanation to vision quickstarts by @kishore-s-15 in https://github.com/deepchecks/deepchecks/pull/1886
    • Adding concurrency to CI jobs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1935
    • AveragePrecisionRecall._evaluate_image: Don't recompute on each min_iou. by @daavoo in https://github.com/deepchecks/deepchecks/pull/1933
    • run time improvements by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1936
    • 1863 resolve two ways of passing n samples to vision suites checks by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1937
    • fixing md5 coco by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1954
    • [DOCS] Many docs fixes (links, various warnings, and content updates) by @shir22 in https://github.com/deepchecks/deepchecks/pull/1956
    • make unused feature a single dataset check by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1950
    • Reduce mixin v2 by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1943
    • Dataset name by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1946
    • Changed gower distance to use actual given samples instead of only indices. by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1955
    • minor improvement in cramers-v runtime by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1966
    • added vision suite static properties kwarg by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1963
    • positive_regression_scorers by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1969
    • [issue-1917][v2] display_in_window fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1952
    • Fixed tqdm version for correct license by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1985
    • Small scorer update by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1977
    • top_5 aggregation method + bug fix in percent_of_nulls.py by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1987
    • Add roc_auc_per_class by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1989
    • Fixed dependencies check by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1996
    • feature: Task type warning for multiclass label containing integer by @DanBasson in https://github.com/deepchecks/deepchecks/pull/1953
    • Deprecate old modules by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1995
    • Changed owners: by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1998
    • Nb/feat/reduce performance by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/2001
    • use deepchecks metrics for precision recall f1 by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2006
    • Added duplicate column validation by @kishore-s-gitaa in https://github.com/deepchecks/deepchecks/pull/1994
    • Remove the ever-present warnings about direct imports from deepchecks by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2007
    • Added env var FAIL_ON_DEEPCHECKS_DEPRECATION_WARNINGS by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/2008
    • set_verbosity ERROR will ignore also regular warnings by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2009
    • Segmentation docs and tests 1888 by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1991
    • Add code to load the airbnb dataset by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2014
    • Feat/added deprecation warning to whole dataset drift check by @kishore-s-15 in https://github.com/deepchecks/deepchecks/pull/2011
    • Add n_samples to all tabular checks (except dataset size check) by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2019
    • Add all sklearn scorers by @noamzbr in https://github.com/deepchecks/deepchecks/pull/2020

    New Contributors

    • @daavoo made their first contribution in https://github.com/deepchecks/deepchecks/pull/1933
    • @kishore-s-gitaa made their first contribution in https://github.com/deepchecks/deepchecks/pull/1994

    Full Changelog: https://github.com/deepchecks/deepchecks/compare/0.8.3...0.9.0

    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.9.0-py3-none-any.whl(3.38 MB)
    deepchecks-0.9.0.tar.gz(3.20 MB)
    LICENSE(34.03 KB)
  • 0.8.3(Aug 23, 2022)

  • 0.8.2(Aug 8, 2022)

    What's Changed

    • fixing double output in docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1733
    • Suite/check conf serialize by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1736
    • fixing pps docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1729
    • Image property outliers display when there are no outliers by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1738
    • handle single value columns gower dist by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1741
    • Fixing a deprecated call in the create a custom suite tutorial by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1744
    • Separate gower distance function by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1746
    • [issue-1660] dataframe index display fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1740
    • [Issue 1654] added "connected" parameter to the "save_as_html" method by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1735
    • Scorers string by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1745
    • Nb/feat/new metrics by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1751
    • Nb/feat/performance report update by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1755
    • fixed edge cases with weird data + tested by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1762
    • Fixed docs links in data classes by @kishore-s-15 in https://github.com/deepchecks/deepchecks/pull/1763
    • Fixed README links by @kishore-s-15 in https://github.com/deepchecks/deepchecks/pull/1765
    • Nb/feat/drift scores by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1773
    • Adding the run function of the suite to the suite api docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1779
    • [Monitoring][issue-52] Added PercentOfNull check / added ReduceMixin to checks by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1758
    • Update README.rst by @shir22 in https://github.com/deepchecks/deepchecks/pull/1784
    • [Issues: 1715, 1730] added 'read more' link to the check short description / fixed 'display_in_gui' func by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1771
    • 1637 SingleDatasetPerformance update to support multiple scorers by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1778
    • tabular single dataset performance by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1788
    • [Issue-1646] renamed 'FeatureLabelCorrelationChange' check by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1787
    • Adapting tabular SimpleModelComparison API by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1785
    • Adding methods and attributes to API ref classes by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1793
    • Nb/feat/update custom check guide by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1792
    • SingleDatasetPerformance Minor adjustment to the scorers signature and display by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1791
    • fix_static_model_bug by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/1797
    • updated unused features plot example by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1805
    • updated classification_dataset_from_directory by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1803
    • small fixes to drift and gower distance by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1794
    • Add indices to conflicting labels display by @matanper in https://github.com/deepchecks/deepchecks/pull/1811
    • [DOCS] Guide for Viewing Results (show) and for Exporting them (json & html) by @shir22 in https://github.com/deepchecks/deepchecks/pull/1812
    • add model_evaluation_gif in README by @shir22 in https://github.com/deepchecks/deepchecks/pull/1817
    • Removing Similar image leakage from the suites because of performance issues by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1806
    • fixed license dependency by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1826
    • bugfix - label in evaluation suite example inferred incorrectly by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1818
    • Added text for drift differentiation by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1815
    • FI new function for users by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1822
    • run checks without labels if they are not needed by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1834
    • Add probability drift on proba by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1809
    • Nb/doc/small fixes by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1837
    • Fix issues in date overlap and duplicate checks by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1839
    • Fixing dependencies by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1844
    • Added flag for users to not ignore nulls in categorical columns by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1843
    • Shir was right, I was wrong by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1840
    • Unify supported models and static predictions by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1835
    • Update plot_create_a_custom_suite.py by @cemalgurpinar in https://github.com/deepchecks/deepchecks/pull/1853
    • [Doc] Documentation contains a typo in "Using Deepchecks Vision With a Few Lines of Code" tutorial. by @cemalgurpinar in https://github.com/deepchecks/deepchecks/pull/1855
    • [suggestion] small fixes to the result display guide by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/1827
    • 1770 seperate quickstarts in user guide by @kishore-s-15 in https://github.com/deepchecks/deepchecks/pull/1798
    • Nb/bug/drift na by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1851
    • Fix the deprecated import warning by importing directly from source by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1857
    • Correctly count the ratio of new categories by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1860
    • Add FeatureFeatureCorrelation to suite and fix python 3.6 bug by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1866
    • aggregation methods for object detection by @Nadav-Barak in https://github.com/deepchecks/deepchecks/pull/1870
    • Adding a section for evaluating a suite in pytest by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1873
    • Added max_num_categories_for_display and show_categories_by to ImageDatasetDrift by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/1876
    • Metrics Guide by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1799
    • Fixing bugs in the docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1881
    • Fix links by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1877
    • Fix issues with image docs by @noamzbr in https://github.com/deepchecks/deepchecks/pull/1879
    • Adding warning if the user is not on the latest version by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/1882
    • 1706 static properties by @TheSolY in https://github.com/deepchecks/deepchecks/pull/1864

    New Contributors

    • @kishore-s-15 made their first contribution in https://github.com/deepchecks/deepchecks/pull/1763
    • @cemalgurpinar made their first contribution in https://github.com/deepchecks/deepchecks/pull/1853

    Full Changelog: https://github.com/deepchecks/deepchecks/compare/0.8.1...0.8.2

    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.8.2-py3-none-any.whl(3.36 MB)
    deepchecks-0.8.2.tar.gz(3.19 MB)
    LICENSE(34.03 KB)
  • 0.8.1(Jul 7, 2022)

    Commits

    • 379db4b: Kaggle display fix (#1659) (Itay Gabbay) #1659
    • 902ee4b: Static pred vision to use dicts (#1653) (JKL98ISR) #1653
    • e107fd4: Adding 0.8 release (Itay Gabbay)
    • 5330f17: [Issue-1545] progress bar text (#1658) (Yurii Romanyshyn) #1658
    • 4478be7: [Issue-1613] cv2 import fix (#1657) (Yurii Romanyshyn) #1657
    • 06bef87: 1586 add condition to image dataset drift (#1665) (Sol) #1665
    • 78593c4: Add tabular benchmark tests (#1668) (matanper) #1668
    • a738df2: Update pr.yaml (#1663) (Nir Hutnik) #1663
    • c4e84b7: Adding benchmarks to CI (#1664) (Itay Gabbay) #1664
    • cb88549: Add new custom check templates (#1633) (matanper) #1633
    • 37fb754: [Issue-1561] features_importance parameter deprecation (#1667) (Yurii Romanyshyn) #1667
    • d14b56f: 1585 rename identifier leakage (#1670) (Sol) #1670
    • 4c3b023: Weak segments performance (#1671) (Nadav-Barak) #1671
    • 19c387b: Update docs index.rst (#1673) (matanper) #1673
    • 8601969: Display flag (#1662) (JKL98ISR) #1662
    • 5c28d02: [DOCS] Update how to use package in IDE (#1682) (Sol) #1682
    • 1ca38ad: Fix nullable int bug (#1676) (JKL98ISR) #1676
    • fbfcaa0: added "result =" to all of the checks documentation + required adjustments (#1684) (Sol) #1684
    • d1ac880: [Issue-1542] MixedNull display table order (#1679) (Yurii Romanyshyn) #1679
    • 236b16b: Deprecating the DominantFrequencyChange check (#1685) (Itay Gabbay) #1685
    • f2ef42b: [issue-1674] mixed nulls bug fix (#1677) (Yurii Romanyshyn) #1677
    • 7cf93ae: [Issue-1560] fix effect of display control parameters on CheckResult.value (#1680) (Yurii Romanyshyn) #1680
    • 7bd8d22: display map widget fix (#1696) (Yurii Romanyshyn) #1696
    • b0bf20b: [issue-1514] conflicting labels check fix (#1694) (Yurii Romanyshyn) #1694
    • 482664f: Shared docstrings (#1669) (Yurii Romanyshyn) #1669
    • bac1be9: Vision benchmark (#1681) (JKL98ISR) #1681
    • d57bb29: Suite api docs (#1693) (Itay Gabbay) #1693
    • ac32522: New vis metrics (#1698) (JKL98ISR) #1698
    • d211cf1: custom vision scorer (#1704) (JKL98ISR) #1704
    • fc70ca5: Nb/feat/weak segments performance (#1678) (Nadav-Barak) #1678
    • a2d8231: Class performance / performance report condition details upgrade (#1720) (JKL98ISR) #1720
    • 54ca182: Ignore NaNs in FeatureFeatureCorrelation (#1718) (Sol) #1718
    • cd967d3: [suggestion] html progress bar (#1717) (Yurii Romanyshyn) #1717
    • 8f07ffb: Update label-new-issue.yml (#1727) (Nir Hutnik) #1727
    • 6a33c73: deprecation warnings (#1732) (Nadav-Barak) #1732
    • eef382e: Quick start model evaluation + new dataset (#1726) (Nadav-Barak) #1726
    • 524595f: Merge branch 'main' of https://github.com/deepchecks/MLChecks into 0.8.x (Itay Gabbay)
    • b0f3fc2: 0.8.1 version bump (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.8.1-py3-none-any.whl(482.08 KB)
    deepchecks-0.8.1.tar.gz(323.14 KB)
    LICENSE(34.03 KB)
  • 0.8.0(Jun 19, 2022)

    Commits

    • b92ae5c: Update dataset exceptions (#1520) (matanper) #1520
    • 7a0b2d3: Mixed data types examples (#1528) (JKL98ISR) #1528
    • 541b119: Fixing doc copyubtton (Itay Gabbay)
    • 22f59bd: Enable validation of empty predictions (#1530) (Noam Bressler) #1530
    • b4b620c: [issue-1448] IsSingleValue check - ignore nan (#1522) (Yurii Romanyshyn) #1522
    • d3c51ac: Deprecation error 1536 (#1548) (Nir Hutnik) #1548
    • b07b308: Update switcher.json (Itay Gabbay)
    • 1cb5be3: reduce_image_size (#1551) (JKL98ISR) #1551
    • d0a49c4: added reduce mixin to performance report (#1550) (JKL98ISR) #1550
    • 8f0f6e2: 1433 feat support scikit learn 1.1.0 (#1554) (matanper) #1554
    • 1343ac8: Vision conditions "more details" update (#1546) (matanper) #1546
    • 0cb97cb: Fixing API reference (#1557) (Itay Gabbay) #1557
    • defcbeb: static_model_suites_fix (#1558) (JKL98ISR) #1558
    • 5387e3a: [Issue 1445] properties output type (#1547) (Yurii Romanyshyn) #1547
    • f71ec33: sticking pylint to 2.13.5 (Itay Gabbay)
    • b461d40: Fixing dependencies license check (Itay Gabbay)
    • 83719b1: Create label-new-issue.yml (Itay Gabbay)
    • c5f15bb: Update label-new-issue.yml (Itay Gabbay)
    • 026780a: Update label-new-issue.yml (Itay Gabbay)
    • 8fa20cc: Centralize Requirements dir and file to variable (#1563) (DanArlowski) #1563
    • 0617ca3: [Issue-1443] tabular Dataset repr (#1549) (Yurii Romanyshyn) #1549
    • 659b6d1: Performance report: add reduce parameter (#1559) (JKL98ISR) #1559
    • 26d7a7b: Fixed bug in cramers V for single value (#1570) (Nir Hutnik) #1570
    • 456a6b0: Add display in python gui (#1568) (matanper) #1568
    • 3f2e686: Nb/infa/pylint violations fix (#1579) (Nadav-Barak) #1579
    • 1e77878: improved condition argument for feature drift (#1578) (Nadav-Barak) #1578
    • d36be62: base form operates only if at least one alphanumeric char exist (#1581) (Nadav-Barak) #1581
    • 677f6f5: Refactor all conditions names (#1583) (matanper) #1583
    • 34adcfe: Lending club dataset (#1588) (Nadav-Barak) #1588
    • 07b1f03: Bugfix/seg infinite loop (#1590) (Noam Bressler) #1590
    • 2e40cca: adding lending club dataset to init (#1595) (shir22) #1595
    • 8b33000: Add passed function to suite result (#1594) (matanper) #1594
    • de91c9b: [DOCS] train test validation quickstart (#1531) (shir22) #1531
    • 8aae4ef: static_model_tab_docs (#1562) (JKL98ISR) #1562
    • Fix display bug on numeric column names (#1615) #1615 (matanper)
    • Replace dataset type with dataset kind (#1616) #1616 (matanper)
    • f827333: Fixed broken links (#1623) (Nadav-Barak)
    • 77823e4: standardize the sampling footnote (#1621) (matanper) #1621
    • 918cc5e: added logger util (#1610) (JKL98ISR) #1610
    • 29e21a3: Improved categorical feature inference + label column description (#1612) (Nadav-Barak) #1612
    • 9dbb763: Fix bug in plots caused by new albumentations version (#1636) (matanper) #1636
    • 19fec99: no_lru (#1631) (JKL98ISR) #1631
    • eda1a42: fixing databricks tutorial (#1644) (Itay Gabbay) #1644
    • d06cb62: Update Readme (#1639) (shir22) #1639
    • 3f899a1: Docs Welcome Page Update (#1641) (shir22) #1641
    • 3d532f7: Update city_arrogance test dataset (#1624) (matanper) #1624
    • a4c07dc: [Issue-1501] new suite output design (#1627) (Yurii Romanyshyn) #1627
    • 01d0c0d: skip timeout/calculation msg in our fi calculations (#1645) (JKL98ISR) #1645
    • f411999: Remove upper level titles from the image dataset drift guide (#1648) (Noam Bressler) #1648
    • 3f9627b: [Issue-1574] new display item type - DisplayMap (#1638) (Yurii Romanyshyn) #1638
    • aed8c9c: Add new check - correlation between features (Sol) #1606
    • bd550b9: Fixed bug (#1651) (Nir Hutnik) #1651
    • a6894a3: fixed link in getting started (#1655) (shir22) #1655
    • d9d9a19: [suggestion] new suite/check result display api (#1611) (Yurii Romanyshyn) #1611
    • 7761cb7: Version bump (Itay Gabbay)
    • 9593171: docs 0.8 bump (Itay Gabbay)
    • 914385e: Kaggle display fix (#1659) (Itay Gabbay)
    • baea2cb: Fix (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.8.0-py3-none-any.whl(455.83 KB)
    deepchecks-0.8.0.tar.gz(307.18 KB)
    LICENSE(34.03 KB)
  • 0.7.2(Jun 7, 2022)

  • 0.7.1(May 31, 2022)

  • 0.7.0(May 29, 2022)

    Commits

    • 84a7e2c: Fix string mismatch performance (#1432) (matanper) #1432
    • 6059c0a: Create about.RST (#1434) (Itay Gabbay) #1434
    • 2c6acd6: Rename ABOUT.rst (shir22) #1435
    • cfeaf97: Update about (#1436) (shir22) #1436
    • 9ae6f18: Airflow integration (#1412) (Itay Gabbay) #1412
    • 9fcb58e: Update checks_and_conditions image in docs (#1441) (shir22) #1441
    • b738310: Add datasets tests (#1439) (matanper) #1439
    • 5cc5db6: Doc mainpage redirect (#1450) (Itay Gabbay) #1450
    • 747a4c0: Catch warnings in detection metric (#1451) (matanper) #1451
    • d80f851: Cramers v (#1446) (JKL98ISR) #1446
    • fca4e27: better format validation for detection data class (#1428) (Nadav-Barak) #1428
    • 14abff1: Segment Performance: fix figure and unnecessary feature importance usage (#1456) (matanper) #1456
    • 27e3e05: Check/Suite Result repr_html fix (#1458) (Yurii Romanyshyn) #1458
    • 98adbfa: Docs get started update (#1466) (shir22) #1466
    • 05e8013: Hide checks without display and remove links (#1470) (matanper) #1470
    • 4bd7fe0: Replace logger.warnings with warnings.warn (#1460) (matanper) #1460
    • 8a325c9: Restructure of the tabular submodule (#1469) (Itay Gabbay) #1469
    • 524ac70: 1409 feat single dataset performance check (#1442) (Sol) #1442
    • 6a5bade: Fix grouping nan types in MixedNulls for pandas < 1.4 (#1473) (matanper) #1473
    • af2f9d7: Vision new structure (#1481) (Itay Gabbay) #1481
    • f945c0d: [issue-1340][BUGFIX] fixed TrainTestSamplesMix fillna issue (#1486) (Yurii Romanyshyn) #1486
    • 4871b77: Images merge (#1488) (Itay Gabbay) #1488
    • ae2c3c4: [Docs] Avocado Sales - Integrity Suite Quickstart (#1474) (shir22) #1474
    • f374742: Fixes for running prediction drift & segment performance on various datasets (Noam Bressler) #1482
    • 0a7d474: Add details to conditions for tabular model evaluation (#1489) (matanper) #1489
    • 3942d0e: Improve sfc 1399 (#1478) (Nir Hutnik) #1478
    • ab84485: [Issue 1410] Progress Bar fix (#1471) (Yurii Romanyshyn) #1471
    • ddc6c0e: Update conditions details for tabular integrity checks (#1485) (matanper) #1485
    • 6861918: import order fix (#1492) (Yurii Romanyshyn) #1492
    • 0679a72: Fixing copybutton (#1494) (Itay Gabbay) #1494
    • e8e85fc: [issue-1467] suite widgets serializer fix (#1493) (Yurii Romanyshyn) #1493
    • 712f824: Nb/feat/correlation methods (#1484) (Nadav-Barak) #1484
    • edfecb4: Add the hugging face transformer example (#1477) (Noam Bressler) #1477
    • 7d693ba: Tabular train-test checks conditions details (#1495) (matanper) #1495
    • 2909a99: Fix H2O subsection title (Noam Bressler) #1499
    • d9029eb: Fixing avocado (#1502) (Itay Gabbay) #1502
    • 067170e: Rename single feature contribution to feature label correlation 1399 (#1498) (Nir Hutnik) #1498
    • 95d8481: bug fix in inf calculations gower method (#1505) (Nadav-Barak) #1505
    • 8be63cc: Warn if cuda present but not used (Itay Gabbay) #1506
    • fecc341: Static model (#1491) (JKL98ISR) #1491
    • 826ae3a: Mixed nulls bug (#1507) (matanper) #1507
    • 48ecb47: fix links to tabular and vision checks (#1504) (shir22) #1504
    • 0959405: [Issue-1365][V2] Plotly figures width&height (#1483) (Yurii Romanyshyn) #1483
    • ac6412d: [Refactoring] add wandb context manager (Yurii Romanyshyn) #1496
    • 9659b7f: Added kwargs for tabular suite + fixed missing kwargs in checks (#1515) (JKL98ISR) #1515
    • f1ddff5: [issue-1516] fixed suite output display order (#1521) (Yurii Romanyshyn) #1521
    • eeadd80: remove DominantFrequencyChange from default suite (#1519) (Noam Bressler) #1519
    • bac80e5: Fixing right nav scroll (#1526) (Itay Gabbay) #1526
    • 40df64d: Fixing the release pipeline (#1529) (Itay Gabbay) #1529
    • e87b2cd: Drift guide 1066 (#1447) (Nir Hutnik) #1447
    • f76ba17: Static vision model (#1525) (JKL98ISR) #1525
    • 2bcf0e5: Adding device to missing tests (#1524) (Itay Gabbay) #1524
    • b4bd746: 0.7.x (JKL98ISR)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.7.0-py3-none-any.whl(432.03 KB)
    deepchecks-0.7.0.tar.gz(289.77 KB)
    LICENSE(34.03 KB)
  • 0.6.4(May 12, 2022)

    Commits

    • 4065254: Simple model global metric (#1352) (Itay Gabbay) #1352
    • c7043bb: Nb/documentation and autocasting pixels (#1345) (Nadav-Barak) #1345
    • 040b8ff: Modify import order (#1353) (Itay Gabbay) #1353
    • 310ae95: Fix adult dataset fitted model (#1354) (matanper) #1354
    • 351f118: Sklearn 0.23.2 support (#1333) (JKL98ISR) #1333
    • 3e19f63: Explaining boosting overfit notebook (#1350) (Itay Gabbay) #1350
    • 1806a5c: Rename check label ambiguity (#1359) (matanper) #1359
    • f606ff0: Fixing tutorials and checks examples (#1362) (Itay Gabbay) #1362
    • bc9dbeb: adding submodule to each doc page (#1366) (Itay Gabbay) #1366
    • d4d7ec9: Update plot model error analysis (#1361) (matanper) #1361
    • b708e14: Torch transformers support (#1357) (JKL98ISR) #1357
    • 3f09614: Added margin_quantile_filter to EMD drift score (#1368) (Nir Hutnik) #1368
    • 3528633: [issue-892] modified VisionData ctor (#1356) (Yurii Romanyshyn) #1356
    • ad112a3: [issue-811] TrainTestFeatureDrift - fi order bug (#1372) (Yurii Romanyshyn) #1372
    • fc35b1e: Update mixed data types plot (#1363) (matanper) #1363
    • 59c42ac: [issue-1248] features inference for passed into check dataframes (#1355) (Yurii Romanyshyn) #1355
    • 3852c62: redo_plotly_ver (#1373) (JKL98ISR) #1373
    • 6114d8e: Fix minor link issue & cast to float in log loss (#1374) (Noam Bressler) #1374
    • 15c6f37: Nb/vision/new label (#1369) (Nadav-Barak) #1369
    • a5fcf3e: [issue-1077] Display logic refactoring (#1279) (Yurii Romanyshyn) #1279
    • ce9d4b2: Add mean and median to dist plots (#1376) (matanper) #1376
    • 6876380: Serialization subpackage fix - added init.py module (#1389) (Yurii Romanyshyn) #1389
    • 6de5e8c: [issue-1370] fixed coco labels map (#1380) (Yurii Romanyshyn) #1380
    • 6929f74: Fix emd drift 1276 revised (#1390) (Nir Hutnik) #1390
    • e3fc8ce: Nb/vision/map output + doc (#1388) (Nadav-Barak) #1388
    • d2d2992: Confusion matrix normalized display (#1377) (JKL98ISR) #1377
    • 9133ae6: Added feature_importance.rst (#1391) (Nir Hutnik) #1391
    • 11b9e52: Adding spark databricks integration (#1395) (Itay Gabbay) #1395
    • 01ec9ff: Fixing examples README.rst, removing OutlierSampleDetection from suite (#1405) (Itay Gabbay) #1405
    • e8205cf: From json (#1402) (JKL98ISR) #1402
    • 8454128: Fixing validations on GPU (#1396) (Itay Gabbay) #1396
    • c6b64d2: hot-fix for the SuiteResult.save_as_html (#1413) (Yurii Romanyshyn) #1413
    • edf9ce2: Adding pytest integration (#1403) (Itay Gabbay) #1403
    • 46ad6e6: Run the phishing integrity suite on a Dataset instead of a DataFrame. (#1417) (Noam Bressler) #1417
    • d8b2885: Minor fixes to the vision documentation: task_type clarification and typo in DetectionData (#1422) (Sol) #1422
    • e09ddd5: Removing the extra display info when nothing found (#1423) (matanper) #1423
    • fb884cc: Docs improve (#1424) (Itay Gabbay) #1424
    • 969be4d: Add example for using H2O with deepchecks (#1400) (Noam Bressler) #1400
    • f4d1e6d: Nb/tabular/increase speed outlier detection + timeout (#1414) (Nadav-Barak) #1414
    • 56c77c0: [BUGFIX] Serialization logic - additional fixes. (#1420) (Yurii Romanyshyn) #1420
    • 0b05fca: Single feature fixes (#1419) (JKL98ISR) #1419
    • 6ba8629: Fixes to rst files in integration examples. (Noam Bressler) #1426
    • 0fe0db3: new png (#1430) (shir22) #1430
    • c3df4f2: Add note to string mismatch checks (#1429) (Noam Bressler) #1429
    • c4a2deb: Merge branch 'main' of https://github.com/deepchecks/MLChecks into 0.6.x (Itay Gabbay)
    • 8fd0895: 0.6.4 version bump 🚀 (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.6.4-py3-none-any.whl(412.20 KB)
    deepchecks-0.6.4.tar.gz(271.79 KB)
    LICENSE(34.03 KB)
  • 0.6.3(Apr 28, 2022)

  • 0.6.2(Apr 28, 2022)

    Commits

    • 8a1b87c: Remove difference from heatmap_comparison.py (#1269) (matanper) #1269
    • c7d1390: 1168 feat single dataset outlier detection (#1237) (Nadav-Barak) #1237
    • 00fceed: improve performance_report notebook (#1271) (DBI) #1271
    • eb13d00: Get exception traceback in suite easily (#1270) (DBI) #1270
    • d02251d: Manually set iris label type (#1275) (matanper) #1275
    • f0aa52e: Check to html (#1277) (JKL98ISR) #1277
    • 16a9014: Link examples with apis (#1264) (Shiv Shankar Dayal) #1264
    • 442b87c: Prevent empty datasets, as that can create bugs down the line (#1273) (Noam Bressler) #1273
    • ff79ce0: fix over 255 bug in whole dataset drift (#1278) (DBI) #1278
    • a43a41a: Add similar image leakage to default suite (#1241) (Noam Bressler) #1241
    • a5aee41: Segment Performance display fix (#1282) (matanper) #1282
    • 95fb347: make show function work on sphinx + fixed double output (#1281) (JKL98ISR) #1281
    • 6c12a8e: Add option to pass FI in tabular to single run (#1285) (matanper) #1285
    • 00843e9: Update warning on categorical infer (#1293) (matanper) #1293
    • 4fd87f9: fix_check_html_style (#1291) (JKL98ISR) #1291
    • 718b7b9: dissable tqdm globaly in tests (#1289) (DBI) #1289
    • 06fb730: Changed calculation of similar image leakage to only take one similar image per test image (#1297) (Nir Hutnik) #1297
    • 0e0827c: New categorical drift method 1138 (#1288) (Nir Hutnik) #1288
    • 60ee6b5: Improve dominant frequency change example (#1296) (Itay Gabbay) #1296
    • b5409ad: Adding examples validation (#1298) (Itay Gabbay) #1298
    • 80f1b37: v0 (#1304) (JKL98ISR) #1304
    • d026204: fix inconsistent fi behavior (#1306) (JKL98ISR) #1306
    • e7841a2: Dc 1238 found nothing (#1316) (DBI) #1316
    • 30dc402: Add test for depr warn (#1318) (Nir Hutnik) #1318
    • 6474add: Tutorial custom check quickfixes (#1321) (Levi Bard) #1321
    • 6b68aaa: Update unused features plot (#1317) (matanper) #1317
    • 0e8bb3d: Fi pipeline to always use permutation (#1315) (JKL98ISR) #1315
    • bac0043: Simple classification data example using EuroSAT to find Simple Feature Contribution issue (#1327) (Noam Bressler) #1327
    • 79d705a: Condition image property fix (#1329) (JKL98ISR) #1329
    • 54e052f: fix_cpu_detach (#1334) (JKL98ISR) #1334
    • 161059c: Update read more generation and add test (#1326) (matanper) #1326
    • 93046ae: Various single feature contribution fixes (#1324) (Noam Bressler) #1324
    • fc2ffee: Similar image small fixes in docs (#1331) (Nir Hutnik) #1331
    • 6359c7a: Update train test samples mix docs (#1325) (matanper) #1325
    • c700e58: Update single feature contribution plot (#1337) (matanper) #1337
    • f442813: Examples restructure (#1335) (Itay Gabbay) #1335
    • e679b25: image_segment_mem_fix (#1341) (JKL98ISR) #1341
    • df80eb1: Merge branch 'main' of https://github.com/deepchecks/MLChecks into 0.6.x (Itay Gabbay)
    • aba238d: 0.6.2 bump (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.6.2-py3-none-any.whl(360.19 KB)
    deepchecks-0.6.2.tar.gz(241.85 KB)
    LICENSE(34.03 KB)
  • 0.6.1(Apr 12, 2022)

    Commits

    • fd8671a: adding 0.6 (Itay Gabbay)
    • ddf6d57: Using abstraction for vision average precision (#1160) (matanper) #1160
    • 3fe0c83: added texlive-latex-extra for missing style files (#1176) (Shiv Shankar Dayal) #1176
    • c5e78f3: fix image array dtype (#1178) (matanper) #1178
    • a93c326: custom metrics reuse bug (#1180) (matanper) #1180
    • c824e77: Fix ModelErrorAnalysis task type error (#1190) (matanper) #1190
    • e229d89: fi_random_state_fix (#1182) (JKL98ISR) #1182
    • 000fd77: Update robustness test (#1192) (matanper) #1192
    • 826fef9: Add Similar image leakage check (#1135) (Noam Bressler) #1135
    • fc2a52b: fix_train_yest_links (#1181) (JKL98ISR) #1181
    • 22d4a72: fixed tooltips and getting started content on sidebar (#1185) (Shiv Shankar Dayal) #1185
    • 0e7ec77: fix_tensor_gpu (#1197) (JKL98ISR) #1197
    • 0514348: change image property drift to raise error instead of returning None (#1193) (JKL98ISR) #1193
    • c4996e5: add_condition_type and test suite condition category results (#1194) (JKL98ISR) #1194
    • 28c72d5: Add sampling to checks and suite run (#1195) (matanper) #1195
    • 8936eac: Update tabular drift checks to raise correct error if there are no features defined (#1205) (matanper) #1205
    • c8a9949: v1 (#1207) (JKL98ISR) #1207
    • 9630d75: Fix the logic finding the weak segment in categorical features. (#1209) (Noam Bressler) #1209
    • 00aa059: Fixing performance bugs (#1212) (Itay Gabbay) #1212
    • 6793e13: apply device to validators (#1216) (Noam Bressler) #1216
    • 2419033: 1141 bug cv unclear if the prediction was correct in classification data validation (#1214) (matanper) #1214
    • b6824ff: Sending anonymous analytics (#1204) (Itay Gabbay) #1204
    • 904e010: Fix read more (#1217) (matanper) #1217
    • bad6c69: Fixing wandb example redirect(#1224) (Itay Gabbay) #1224
    • 576efbd: Fixing wandb example redirect(#1224) (Itay Gabbay)
    • 6b5ec9c: fix (Itay Gabbay)
    • 94c93eb: wandb train/test fix plus better value formatting (#1213) (JKL98ISR) #1213
    • 4fa9ae0: Adding examples to conditions (#1226) (Itay Gabbay) #1226
    • d9c1c29: Disable CI telemetry (#1227) (Itay Gabbay) #1227
    • 5c1dcac: Fix drift plots for few unique numerical values (#1199) (matanper) #1199
    • a64bb51: fixed simple model new classes in test behavior (#1230) (JKL98ISR) #1230
    • 817d4bf: 1140 bug cv add label map to conditions (#1228) (matanper) #1228
    • c9ce8b9: Update README to accommodate beta release of vision package (#1231) (shir22) #1231
    • 8108684: reduce runtime of jaccard_iou by more than 50 percent (#1223) (DBI) #1223
    • 6fa04e6: Upgrade gpu torch vision (#1234) (Itay Gabbay) #1234
    • 7a0b9de: fix suite progress bar in single checks (#1232) (JKL98ISR) #1232
    • 0fc6e9c: RobustnessReport & ClassPerformance work on custom task (#1235) (matanper) #1235
    • 1f10d97: Add tests for docs plots on GPU (#1229) (matanper) #1229
    • e49958b: Add progress bars to single checks runs (#1236) (matanper) #1236
    • 3b74680: Fix model error plot (#1239) (matanper) #1239
    • 0475894: Align the label index in LABEL_MAP to match cls_id (#1244) (Noam Bressler) #1244
    • 6de3e96: Update outliers & robustness image grid (#1245) (matanper) #1245
    • 195c7fd: Model error analysis gallery (#1225) (Itay Gabbay) #1225
    • fd0e0bb: mixed nulls on non string columns (#1211) (DBI) #1211
    • b4f1227: Update width: inherit to fill-available (#1249) (matanper) #1249
    • b09450b: Update string mismatch docs (#1247) (matanper) #1247
    • d6ce78f: allow checks to return CheckFailure (#1252) (DBI) #1252
    • b8898ef: Image segment graph improvements (#1251) (JKL98ISR) #1251
    • 85ad712: Merge from main (Itay Gabbay)
    • 86b67f8: versio bump (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.6.1-py3-none-any.whl(347.70 KB)
    deepchecks-0.6.1.tar.gz(233.79 KB)
    LICENSE(34.03 KB)
  • 0.6.0(Apr 1, 2022)

    Commits

    • 0bbcfed: New version release (github-actions[bot])
    • 2b017f3: Move responsibilities from context to dataset (#767) (matanper) #767
    • 9a54486: [issue-650] functionality split (#751) (Yurii Romanyshyn) #751
    • 3defd1b: small fix (#770) (Yurii Romanyshyn) #770
    • 65739b7: links-checks github action fix (#775) (Yurii Romanyshyn) #775
    • 08a350f: added_series_predict_support (#771) (JKL98ISR) #771
    • 2543d85: updated makefile (#780) (Yurii Romanyshyn) #780
    • 08b9d75: Adding base vision package (#773) (Itay Gabbay) #773
    • 0fdb33d: Add links to suite in where to use docs (#783) (matanper) #783
    • 39bfef7: Making the notebooks gen multi-procces (#785) (Itay Gabbay) #785
    • e6e36c3: Vision docs placeholder (#788) (Itay Gabbay) #788
    • 142524d: Fixing rtd build (#790) (Itay Gabbay) #790
    • 0b2886f: Add sample data loader & sample predictions and label caching (#786) (matanper) #786
    • 7995a74: Fix pandas warning (#792) (JKL98ISR) #792
    • e2ba947: [issue-781] moved checks base classes to core (#789) (Yurii Romanyshyn) #789
    • e739cc5: Fixing nb generation (#795) (Itay Gabbay) #795
    • c2bca9a: [issue-757] updated MNIST dataset api (#798) (Yurii Romanyshyn) #798
    • 8eb0187: Adding 2022 to license (#802) (Itay Gabbay) #802
    • d4d33a6: Label drift fixes (#797) (Nir Hutnik) #797
    • 7c9ae4e: Adding labels and detection encoder classes (#800) (Itay Gabbay) #800
    • 3d84a3e: fix_widget_plotly_bug (#807) (JKL98ISR) #807
    • 0abec77: New version release (github-actions[bot])
    • bcd6de2: Updated Readme (#814) (shir22) #814
    • 0edb5b5: rename VisionDataset to VisionData (#820) (Noam Bressler) #820
    • 4bcd1d7: [DOCS] Updated main page and when should you use page (shir22) #821
    • ec71505: add DataFormatter (#812) (Noam Bressler) #812
    • 66b7f28: Fixing label drift in vision submodule (#823) (Itay Gabbay) #823
    • b3bf954: rename to ImageFormatter and fix bug in VisionData init (#824) (Noam Bressler) #824
    • d546f38: [issue-799] coco split into train/test; updated coco dataset module api (#805) (Yurii Romanyshyn) #805
    • f9b1ef6: change checks to init - update - compute structure (#826) (Noam Bressler) #826
    • ec307ea: Fixing CI (#838) (Itay Gabbay) #838
    • 0833fd8: Label drift fixes (#808) (Nir Hutnik) #808
    • 11230f6: Label drift refactor (#843) (Nir Hutnik) #843
    • 0a68f7f: Adding performance report nb (#844) (Itay Gabbay) #844
    • 8e852a5: Update README.rst (#849) (JKL98ISR) #849
    • f911fe8: Added typehints where they were missing in CheckResult's methods. Also slightly improved english in docs for those functions (#848) (RonItay) #848
    • 77735cd: Formatter docs (#837) (Itay Gabbay) #837
    • b2f0c20: [issue-787] added device parameter to the vision checks context (#832) (Yurii Romanyshyn) #832
    • 1f9f6b4: Added properties in image formatter (#855) (Nir Hutnik) #855
    • 29302fa: Vision quickstart notebook (#853) (Noam Bressler) #853
    • de7bab1: Robustness Report (#801) (matanper) #801
    • 2ace56c: Image properties (#858) (Nir Hutnik) #858
    • aaa22c7: suite fixes (#857) (Yurii Romanyshyn) #857
    • 595b58d: Vision docs (#860) (Itay Gabbay) #860
    • ba13a32: add export to html to quickstart (#865) (Noam Bressler) #865
    • 754b25a: Adding class display limit in ClassPerformance check (#866) (Itay Gabbay) #866
    • c7ccd5b: Add title to drift plot (#867) (Noam Bressler) #867
    • eb5ecd8: Improved metric (#825) (JKL98ISR) #825
    • 69a0f65: Added imgaug-equivalent augmentations. (#869) (Nir Ben-Zvi) #869
    • 1e62372: add wrap_run to ModelComparisonCheck (#868) (Noam Bressler) #868
    • fee9fd5: remove warnings from main (#870) (JKL98ISR) #870
    • 44e2551: Modifying make notebook (#873) (Itay Gabbay) #873
    • 0ffe328: Add recall to default metrics (#872) (JKL98ISR) #872
    • fbd8e46: Fixing deployment (#861) (Itay Gabbay) #861
    • 8dcb676: Vision domain classifier issue 841 (#879) (Nir Hutnik) #879
    • 988eed7: MeanAveragePrecisionReport check (#880) (JKL98ISR) #880
    • d25643e: Formatters validation (#875) (matanper) #875
    • dccb198: v1 (#885) (JKL98ISR) #885
    • 28e5eb5: Fix version read in init (#891) (matanper) #891
    • 45e379a: fix_cells (JKL98ISR)
    • 3f98ce3: Revert "fix_cells" (JKL98ISR)
    • 148e0df: Adding label_map to dataset (#877) (Itay Gabbay) #877
    • d31a0cb: Vision Suite refactor (#890) (JKL98ISR) #890
    • 547b225: Fixed (maybe?) the tests (#893) (Nir Hutnik) #893
    • ec6122b: Make image drift tests consistent (#902) (matanper) #902
    • a9595a4: fixing docs (#898) (Itay Gabbay) #898
    • 1220e7f: Vision model comparison (#897) (Itay Gabbay) #897
    • 61b30a0: Save mnist model for tests - Fix robustness test (#905) (matanper) #905
    • b68f44e: Set seed in vision context (#904) (matanper) #904
    • 5c3f663: [Issue 834] label formatters (#882) (Yurii Romanyshyn) #882
    • 12f5e34: Changing alternative_metrics to Dict (#907) (Itay Gabbay) #907
    • 9ec023e: set default prediction formatters per task type (#909) (Noam Bressler) #909
    • e196de6: The notebook is back (#900) (Nir Hutnik) #900
    • 580a12c: Robustness + VisionData: copy generator when copying dataset (#924) (matanper) #924
    • b801321: Update robustness report notebook (#913) (matanper) #913
    • f0e45e8: Remove state copying and setting seed in robustness_report (#926) (matanper) #926
    • fc854e0: Gpu fixes (#912) (Itay Gabbay) #912
    • 2cecf54: Heatmap comp (#901) (Noam Bressler) #901
    • b7fa71c: add confusion matrix (#828) (DBI) #828
    • d347792: [Issue 755] Property Drift Check (#908) (Yurii Romanyshyn) #908
    • c89c4c3: version bump (#927) (Itay Gabbay) #927
    • e3336a0: Add new checks to vision full suite (Noam Bressler) #929
    • 3ab5dbd: Wandb integration (#906) (JKL98ISR) #906
    • e7f6149: Fix bugs when task_type is None and remove print (#937) (Noam Bressler) #937
    • 0e0fba2: Adding gpu tests (#939) (Itay Gabbay) #939
    • f767889: Adding explanations to label drift notebook. (#936) (Itay Gabbay) #936
    • 55b7b85: Add segment performance check (#935) (matanper) #935
    • 5109c78: Fixing quickstart (#953) (Itay Gabbay) #953
    • f79caa3: Add latest version check to package init (#956) (matanper) #956
    • e9d89a5: Refactor label drift (#958) (Noam Bressler) #958
    • faa527c: New version release (github-actions[bot])
    • f8e6adb: Add DEEPCHECKS_DISABLE_LATEST to build (#962) (matanper) #962
    • c35bac6: Prediction drift check (#963) (Noam Bressler) #963
    • 20e08a4: Add prediction and label drift conditions (#968) (Noam Bressler) #968
    • Fix sphinx errors (#970) #970 (Shiv Shankar Dayal)
    • f5ec928: Update robustness report display (#969) (matanper) #969
    • 5d8cd2f: Update drift plot title (#971) (matanper) #971
    • 50e2859: Fix order of results in vision suite (#977) (matanper) #977
    • 7cc184f: Split VisionData and remove Formatters (#966) (JKL98ISR) #966
    • d676fde: Fix inverted confusion matrix axis (#980) (matanper) #980
    • 0d48726: Fix vision suite run (#982) (matanper) #982
    • 2e63c00: Fixing heatmap_comparison.py (#981) (Itay Gabbay) #981
    • 44ef310: Import warnings fix/ import backward compatibility (#972) (Yurii Romanyshyn) #972
    • 38e185a: Confusion matrix - detection classes ids bug (#983) (matanper) #983
    • 2c3abe0: Update validate image data error message to be clearer and include range (#986) (matanper) #986
    • f482107: Add progress bars to vision suite run (#984) (matanper) #984
    • f0dd066: Split many files in core and base (#987) (JKL98ISR) #987
    • 599c014: Make sure cv2 IO is unit8 (#992) (Noam Bressler) #992
    • f358feb: Dc 941 test load dataset use downloaded dataset (#978) (DBI) #978
    • c9e56cb: [Issue 922] add Batch type (#994) (Yurii Romanyshyn) #994
    • c26bb69: Updating Main Readme Visuals (#1002) (shir22) #1002
    • b513428: Simple feature contribution issue 842 (#989) (Nir Hutnik) #989
    • 4c01c1f: 0.6.0.dev0 (#1001) (Itay Gabbay) #1001
    • 9ba1cc8: fix errors in classification tutorial (#1003) (Noam Bressler) #1003
    • 42b33d0: Some condition changes (#1007) (JKL98ISR) #1007
    • daf7456: label and predictions should be tensors (#999) (Noam Bressler) #999
    • 4e19795: Add sample option to vision (#991) (matanper) #991
    • d33cc5f: Fixing RTD badge (#1008) (Itay Gabbay) #1008
    • 6fafb35: vision model error analysis (#964) (DBI) #964
    • aebd6cc: Refactor image & label & prediction properties (#1013) (matanper) #1013
    • 8ddc5fc: Heatmap per class fix (#1017) (JKL98ISR) #1017
    • 04913ac: Add map of class to list of dataset index (#1024) (matanper) #1024
    • 7e313a4: refactor _evaluate_image to improve performance (#1027) (DBI) #1027
    • ac91e11: Robustness images sampled using classes_indices (#1028) (matanper) #1028
    • 3d5e760: [Issue 985] Confusion matrix report - unseen classes bug fix (#1029) (Yurii Romanyshyn) #1029
    • 7a07749: deepchecks label to wandb runs (#1035) (JKL98ISR) #1035
    • 26ec933: Remove auto cache calc on classes_indices (#1033) (matanper) #1033
    • 78b2a49: Docs restructure (#1032) (Itay Gabbay) #1032
    • 2aa48a7: Add "other" task type (#1037) (matanper) #1037
    • 4dc894d: Update Readme.rst (#1044) (Julien Schuermans) #1044
    • 66a6514: Class performance alt metrics fix (#1036) (JKL98ISR) #1036
    • 5a5183d: Add properties user guide (#1026) (matanper) #1026
    • ddb23ad: Raise an error if model is in training mode when passed to check (#1050) (Noam Bressler) #1050
    • 3438789: Visiondata docs (#1038) (Itay Gabbay) #1038
    • df3207b: Revert "Visiondata docs (#1038)" (#1063) (matanper) #1063
    • 3b48e2c: Adding mocks to yolo and mnist (#1046) (Itay Gabbay) #1046
    • 5098293: [Issue-990] "numpy_to_html_image" improvement (#1043) (Yurii Romanyshyn) #1043
    • 14916c9: Custom vision check (#1047) (Noam Bressler) #1047
    • 5167de4: Automatic deployment of docs to github pages (#1060) (Itay Gabbay) #1060
    • 279b0a2: Fixing refs bug (Itay Gabbay)
    • d87a63d: [issue-1005] SimpleClasifficationDataset (#1045) (Yurii Romanyshyn) #1045
    • 145bb7a: Visiondata docs (#1065) (Itay Gabbay) #1065
    • 44c54bc: Fixing deprecation warning while importing deepchecks (#1068) (Itay Gabbay) #1068
    • e9e42a2: Adding version selector to the documentation (#1071) (Itay Gabbay) #1071
    • dc4e007: Add classes_to_display to property drift (#1070) (Noam Bressler) #1070
    • 1e065a3: add test to vision base (#1049) (DBI) #1049
    • 1919d1b: some_visual_improvments (#1059) (JKL98ISR) #1059
    • e9f8fcb: Add custom task tutorial (#1057) (matanper) #1057
    • a65952e: Check display validation (#1034) (JKL98ISR) #1034
    • 3981b95: change x axis name to column name instead of value in drift graphs (#1073) (DBI) #1073
    • 39ace96: Add a classes_to_display optional list parameter to Simple Feature Contribution (Nir Hutnik) #1072
    • 58c421c: docs version selector fix (#1075) (Itay Gabbay) #1075
    • Add assert in vision objects to Batch (#1078) #1078 (matanper)
    • 327cc01: Implement sphinx gallery (#1041) (Shiv Shankar Dayal) #1041
    • d1e74e5: updating docs main page for vision (#1079) (shir22) #1079
    • 8802bcd: Deleting notebooks (#1088) (Itay Gabbay) #1088
    • a559684: [Issue-667] added **kwargs to all checks init (#1082) (Yurii Romanyshyn) #1082
    • 191ae90: fix display bug (#1092) (JKL98ISR) #1092
    • a68b10a: Refactor for labels & image outlier (#1091) (matanper) #1091
    • 9382cb4: update docs main page (#1090) (shir22) #1090
    • 1a29cf2: adding code owners (#1094) (Itay Gabbay) #1094
    • e95572e: add test to conditions in class performance check (#1087) (DBI) #1087
    • 2df9203: add example of limiting classes to property drift display (#1098) (Noam Bressler) #1098
    • 6054e6a: Rst to gallery (#1099) (Itay Gabbay) #1099
    • f409302: Json cond fix + condition refactor (#1100) (JKL98ISR) #1100
    • 88aa120: fix_quick_start_example (#1101) (JKL98ISR) #1101
    • b2b149f: Adding condition to CV simple model comparison (#1089) (Itay Gabbay) #1089
    • d748d30: cond_order_fix (#1106) (JKL98ISR) #1106
    • 9a43ef6: Fix Model Error Analysis visualization and log_loss computation (JKL98ISR) #1085
    • 14578ee: remove_accidental_code (#1111) (JKL98ISR) #1111
    • fdb0612: fix docstring (#1093) (DBI) #1093
    • c254ff2: fixed TOC, thumbnails, plots and search (#1110) (Shiv Shankar Dayal) #1110
    • eff50be: improve colors (#1117) (DBI) #1117
    • 739f042: Adding the adult dataset to deepchecks.datasets (#1112) (Itay Gabbay) #1112
    • 77c9ec1: updated setup.py (#1076) (Yurii Romanyshyn) #1076
    • 1e6f8b4: fix read more url (#1103) (matanper) #1103
    • 45b7735: Adding citation (#1121) (Itay Gabbay) #1121
    • 9bd9f2b: version selector is now working (#1113) (Shiv Shankar Dayal) #1113
    • 6cc2d25: Add a tutorial for using simple classification data to run a train test distribution suite(#1119) (Noam Bressler) #1119
    • 2e630e8: Image property outlier (#1102) (matanper) #1102
    • 535d0cb: make numerical type explicit + other edge case fixes (#1114) (JKL98ISR) #1114
    • 0e7b25a: fixing search (Itay Gabbay)
    • ef7ffa3: add prediction drift to tabular (#1105) (DBI) #1105
    • 83e991f: fixed broken links (#1125) (Shiv Shankar Dayal) #1125
    • 27a18f5: [Issue-721][tabular] IdentifierLeakage fix (#1104) (Yurii Romanyshyn) #1104
    • 4e5f63e: fixing search (#1126) (Itay Gabbay) #1126
    • 2178e4c: Fix bad model metrics/bbox properties (#1128) (JKL98ISR) #1128
    • 7b59f08: tabular unitests for more model types (#1131) (JKL98ISR) #1131
    • 2aea59b: Label property outlier (#1132) (matanper) #1132
    • 05c06bc: Fix outliers with none properties (#1136) (matanper) #1136
    • 08b936e: remove trust score (#1146) (Noam Bressler) #1146
    • 8ff977c: Heatmap Comparison fix to work on custom task (#1134) (matanper) #1134
    • dab4a3f: Sending version (#1137) (Itay Gabbay) #1137
    • eeb7ac5: fix_typo_context (#1149) (JKL98ISR) #1149
    • 983a064: Remove the word alternative from properties params (#1150) (matanper) #1150
    • 65b81ad: Fix problems in old torch version (#1152) (JKL98ISR) #1152
    • b3143b6: Add device to validate formatters (#1153) (matanper) #1153
    • a5a1bb4: Add new integrity suite and kwargs to vision suites (#1155) (matanper) #1155
    • d3606e8: Add README for Swindler Detection (#1158) (shir22) #1158
    • adde700: Fix incorrect exceptions (#1157) (matanper) #1157
    • 01a3462: check ordinal encoder kwargs (#1154) (JKL98ISR) #1154
    • b545c21: Docs get started (#1156) (shir22) #1156
    • 20bd3e0: Revert "Add README for Swindler Detection (#1158)" (Itay Gabbay)
    • eeb3274: adding 0.6 (Itay Gabbay)
    • c90fb78: adding 0.6 (Itay Gabbay)
    • 9c3eb17: adding 0.6 (Itay Gabbay)
    • e8c4699: adding 0.6 (Itay Gabbay)
    • 4e5f510: adding 0.6 (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.6.0-py3-none-any.whl(335.18 KB)
    deepchecks-0.6.0.tar.gz(226.08 KB)
    LICENSE(34.03 KB)
  • 0.5.0(Mar 3, 2022)

  • 0.5.0.dev2(Feb 23, 2022)

    What's Changed

    • Move responsibilities from context to dataset by @matanper in https://github.com/deepchecks/deepchecks/pull/767
    • [issue-650] functionality split by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/751
    • functionality-split additional fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/770
    • links-checks github action fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/775
    • added tests for single feature data + improved tests by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/771
    • updated makefile by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/780
    • Adding base vision package by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/773
    • Add links to suite by @matanper in https://github.com/deepchecks/deepchecks/pull/783
    • Making the notebooks gen multi-procces by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/785
    • Vision docs placeholder by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/788
    • Fixing rtd build by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/790
    • Add sample data loader & sample predictions and label caching by @matanper in https://github.com/deepchecks/deepchecks/pull/786
    • Fix pandas warning by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/792
    • [issue-781] moved checks base classes to core by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/789
    • Fixing nb generation by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/795
    • [issue-757] updated MNIST dataset api by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/798
    • Adding 2022 to license by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/802
    • Label drift fixes by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/797
    • Adding labels and detection encoder classes by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/800
    • Fix widget plotly bug by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/807
    • 📓 Updating Readme by @shir22 in https://github.com/deepchecks/deepchecks/pull/814
    • rename VisionDataset to VisionData by @noamzbr in https://github.com/deepchecks/deepchecks/pull/820
    • [DOCS] Updated main page and when should you use page by @shir22 in https://github.com/deepchecks/deepchecks/pull/821
    • add DataFormatter by @noamzbr in https://github.com/deepchecks/deepchecks/pull/812
    • Fixing label drift in vision submodule by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/823
    • rename to ImageFormatter and fix bug in VisionData init by @noamzbr in https://github.com/deepchecks/deepchecks/pull/824
    • [issue-799] coco split into train/test; updated coco dataset module api by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/805
    • change checks to init - update - compute structure by @noamzbr in https://github.com/deepchecks/deepchecks/pull/826
    • Fixing CI by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/838
    • Label drift fixes by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/808
    • Label drift refactor by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/843
    • Adding performance report nb by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/844
    • fix README.rst typo train-test by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/849
    • Added typehints where they were missing in CheckResult's methods. by @RonItay in https://github.com/deepchecks/deepchecks/pull/848
    • Formatter docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/837
    • [issue-787] added device parameter to the vision checks context by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/832
    • Added properties in image formatter by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/855
    • Vision quickstart notebook by @noamzbr in https://github.com/deepchecks/deepchecks/pull/853
    • Robustness Report by @matanper in https://github.com/deepchecks/deepchecks/pull/801
    • Image properties by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/858
    • suite fixes by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/857
    • Vision docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/860
    • add export to html to quickstart by @noamzbr in https://github.com/deepchecks/deepchecks/pull/865
    • Adding class display limit in ClassPerformance check by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/866
    • Add title to drift plot by @noamzbr in https://github.com/deepchecks/deepchecks/pull/867
    • Improved metric by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/825
    • Added imgaug-equivalent augmentations. by @nirbenz in https://github.com/deepchecks/deepchecks/pull/869
    • add wrap_run to ModelComparisonCheck by @noamzbr in https://github.com/deepchecks/deepchecks/pull/868
    • remove warnings by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/870
    • Modifying make notebook by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/873
    • Add recall to default metrics by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/872
    • Fixing deployment by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/861
    • Vision domain classifier issue 841 by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/879
    • MeanAveragePrecisionReport check by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/880
    • Formatters validation by @matanper in https://github.com/deepchecks/deepchecks/pull/875
    • Mean Average Recall Report check by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/885
    • Fix version read in init by @matanper in https://github.com/deepchecks/deepchecks/pull/891
    • Adding label_map to dataset by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/877
    • Vision Suite refactor by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/890
    • Fixed (maybe?) the tests by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/893
    • Make image drift tests consistent by @matanper in https://github.com/deepchecks/deepchecks/pull/902
    • fixing docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/898
    • Vision model comparison by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/897
    • Save mnist model for tests by @matanper in https://github.com/deepchecks/deepchecks/pull/905
    • Set seed in vision context and fix YOLO tests by @matanper in https://github.com/deepchecks/deepchecks/pull/904
    • [Issue 834] label formatters by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/882
    • Changing alternative_metrics to Dict by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/907
    • set default prediction formatters by @noamzbr in https://github.com/deepchecks/deepchecks/pull/909
    • The notebook is back by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/900
    • Robustness + VisionData: copy generator when copying dataset by @matanper in https://github.com/deepchecks/deepchecks/pull/924
    • Update robustness report notebook by @matanper in https://github.com/deepchecks/deepchecks/pull/913
    • Remove state copying and setting seed in robustness_report by @matanper in https://github.com/deepchecks/deepchecks/pull/926
    • Gpu fixes by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/912
    • Heatmap comp by @noamzbr in https://github.com/deepchecks/deepchecks/pull/901
    • add confusion matrix by @benisraeldan in https://github.com/deepchecks/deepchecks/pull/828
    • [Issue 755] Property Drift Check by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/908

    New Contributors

    • @RonItay made their first contribution in https://github.com/deepchecks/deepchecks/pull/848
    • @nirbenz made their first contribution in https://github.com/deepchecks/deepchecks/pull/869

    Full Changelog: https://github.com/deepchecks/deepchecks/compare/0.4.0...0.5.0.dev2

    Source code(tar.gz)
    Source code(zip)
  • 0.5.0.dev0(Feb 14, 2022)

    What's Changed

    • Move responsibilities from context to dataset by @matanper in https://github.com/deepchecks/deepchecks/pull/767
    • [issue-650] functionality split by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/751
    • functionality-split additional fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/770
    • links-checks github action fix by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/775
    • added tests for single feature data + improved tests by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/771
    • updated makefile by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/780
    • Adding base vision package by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/773
    • Add links to suite by @matanper in https://github.com/deepchecks/deepchecks/pull/783
    • Making the notebooks gen multi-procces by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/785
    • Vision docs placeholder by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/788
    • Fixing rtd build by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/790
    • Add sample data loader & sample predictions and label caching by @matanper in https://github.com/deepchecks/deepchecks/pull/786
    • Fix pandas warning by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/792
    • [issue-781] moved checks base classes to core by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/789
    • Fixing nb generation by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/795
    • [issue-757] updated MNIST dataset api by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/798
    • Adding 2022 to license by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/802
    • Label drift fixes by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/797
    • Adding labels and detection encoder classes by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/800
    • Fix widget plotly bug by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/807
    • 📓 Updating Readme by @shir22 in https://github.com/deepchecks/deepchecks/pull/814
    • rename VisionDataset to VisionData by @noamzbr in https://github.com/deepchecks/deepchecks/pull/820
    • [DOCS] Updated main page and when should you use page by @shir22 in https://github.com/deepchecks/deepchecks/pull/821
    • add DataFormatter by @noamzbr in https://github.com/deepchecks/deepchecks/pull/812
    • Fixing label drift in vision submodule by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/823
    • rename to ImageFormatter and fix bug in VisionData init by @noamzbr in https://github.com/deepchecks/deepchecks/pull/824
    • [issue-799] coco split into train/test; updated coco dataset module api by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/805
    • change checks to init - update - compute structure by @noamzbr in https://github.com/deepchecks/deepchecks/pull/826
    • Fixing CI by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/838
    • Label drift fixes by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/808
    • Label drift refactor by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/843
    • Adding performance report nb by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/844
    • fix README.rst typo train-test by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/849
    • Added typehints where they were missing in CheckResult's methods. by @RonItay in https://github.com/deepchecks/deepchecks/pull/848
    • Formatter docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/837
    • [issue-787] added device parameter to the vision checks context by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/832
    • Added properties in image formatter by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/855
    • Vision quickstart notebook by @noamzbr in https://github.com/deepchecks/deepchecks/pull/853
    • Robustness Report by @matanper in https://github.com/deepchecks/deepchecks/pull/801
    • Image properties by @nirhutnik in https://github.com/deepchecks/deepchecks/pull/858
    • suite fixes by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/857
    • Vision docs by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/860

    New Contributors

    • @RonItay made their first contribution in https://github.com/deepchecks/deepchecks/pull/848

    Full Changelog: https://github.com/deepchecks/deepchecks/compare/0.4.0...0.5.0.dev0

    Source code(tar.gz)
    Source code(zip)
  • 0.4.2(Feb 4, 2022)

  • 0.4.1(Jan 26, 2022)

  • 0.4.0(Jan 26, 2022)

    Deepchecks Version 0.4.0

    What's Changed

    • Update index.rst by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/699
    • Adding notebooks by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/702
    • Build model if loading causes errors by @benisraeldan in https://github.com/deepchecks/deepchecks/pull/703
    • merge kaggle widgets hotfix to main by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/709
    • update suite exceptions and links by @matanper in https://github.com/deepchecks/deepchecks/pull/706
    • Colab prettier suite view by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/717
    • Check display read more by @matanper in https://github.com/deepchecks/deepchecks/pull/715
    • [Issue-571] make docs errors/warnings fixes by @yromanyshyn in https://github.com/deepchecks/deepchecks/pull/725
    • make conda build trigger by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/726
    • Add check context by @matanper in https://github.com/deepchecks/deepchecks/pull/700
    • merge html resources fix from 0.3.2 (#728) by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/733
    • update readme plot by @JKL98ISR in https://github.com/deepchecks/deepchecks/pull/720
    • Added validations to numpydoc docstrings ✅ by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/738
    • Update docstrings in default_suites.py by @matanper in https://github.com/deepchecks/deepchecks/pull/718
    • explicitly set default timeout for fi calculation by @noamzbr in https://github.com/deepchecks/deepchecks/pull/739
    • Rename scorers related parameters by @matanper in https://github.com/deepchecks/deepchecks/pull/734
    • Removed generated html files generated from notebook by @matanper in https://github.com/deepchecks/deepchecks/pull/741
    • Warn user when trying to show result in a non-interactive shell by @matanper in https://github.com/deepchecks/deepchecks/pull/745
    • add check display tests by @benisraeldan in https://github.com/deepchecks/deepchecks/pull/735
    • Add more warning for the user when calculating FI by @matanper in https://github.com/deepchecks/deepchecks/pull/749
    • Add FI permutation options to UnusedFeatures by @matanper in https://github.com/deepchecks/deepchecks/pull/740
    • label amibuity notebook improvemnts by @benisraeldan in https://github.com/deepchecks/deepchecks/pull/760
    • add conda recipe for ipython by @benisraeldan in https://github.com/deepchecks/deepchecks/pull/764
    • 0.4.0 version bump 🚀 by @ItayGabbay in https://github.com/deepchecks/deepchecks/pull/766

    Full Changelog: https://github.com/deepchecks/deepchecks/compare/0.3.0...0.4.0

    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.4.0-py3-none-any.whl(208.74 KB)
    deepchecks-0.4.0.tar.gz(142.55 KB)
    LICENSE(34.03 KB)
  • 0.3.1(Jan 20, 2022)

  • 0.3.0(Jan 19, 2022)

    Deepchecks Version 0.3.0

    Supported Environments

    • Support running from non-jupyter environments - Suite outputs can be exported to html, and Check or Suite outputs can be exported to JSON for serialization. HTML export: #638, JSON export: #665
    • Added support also for non-sklearn models. Supporting any model object that implements the predict method (and also predict_proba for classification), see supported models #594
    • Fixed dependency warning when installing on google colab #577
    • Fixed links from inside suite and check output to open in a new tab

    Display and Output Improvements

    • Output can be exported to html #638
    • Suite output is now in a widget, enabling to browse quickly between checks, with and without conditions or display #610
    • Distribution graphs smoothened, graphs looks better also when feature has outlier values #664
    • Improved error handling. Checks that didn’t run due to irrelevant input (e.g. no index column) are no longer displayed as errors. #640

    Docs

    Improved documentation for many check examples:

    Added documentation for export types:

    New Contributors

    • @idow09 made their first contribution 🥳
    • @BioGeek made their first contribution 🥳
    • @shaypal5 made their first contribution 🥳

    Maintainers

    @matanper , @ItayGabbay , @yromanyshyn , @JKL98ISR , @benisraeldan , @deepchecks-bot , @shir22 , @nirhutnik , @noamzbr

    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.3.0-py3-none-any.whl(204.63 KB)
    deepchecks-0.3.0.tar.gz(139.59 KB)
    LICENSE(34.03 KB)
  • 0.2.1(Jan 7, 2022)

    Commits

    • d8b1d43: New version release (github-actions[bot])
    • 1ad95b1: Update README.rst (#513) (shir22) #513
    • 69100ea: Add examples folder and routing from readme (#514) (shir22) #514
    • d96117b: changing gh labels (#515) (Itay Gabbay) #515
    • aee01a6: Update README.rst (#516) (shir22) #516
    • 233fbce: Update performance.rst (#519) (shir22) #519
    • 4c7a4c1: Update display_suite.py (#518) (shir22) #518
    • 214213a: Documentation and Readme (#521) (shir22) #521
    • c919dd5: Merge branch 'main' of https://github.com/deepchecks/MLChecks into 0.2.x (Itay Gabbay)
    • 48fd68c: docs fixes (Itay Gabbay)
    • 0cfa6f3: Fix PerfectModel when data have nulls in label (#526) (matanper) #526
    • 2051d65: - Adding iris to the datasets section (#530) (Itay Gabbay) #530
    • dfa08fa: Update the Dataset Object Documentation (#500) (DBI)
    • 9a0e202: version bump (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.2.1-py3-none-any.whl(191.64 KB)
    deepchecks-0.2.1.tar.gz(129.63 KB)
    LICENSE(34.03 KB)
  • 0.2.0(Jan 5, 2022)

    Commits

    • 0df2f8e: Performance report degradation fix (#444) (JKL98ISR) #444
    • 2a243b4: Change binary scorers to also return per class (#470) (matanper) #470
    • fb94d10: N top show how many (#469) (JKL98ISR) #469
    • 84ee22f: remove_zero_warn (#476) (JKL98ISR) #476
    • b52add9: no_links_collab (#478) (JKL98ISR) #478
    • 5ec8829: [Issue-439, addition] SimpleModelComparison: added 'Number of samples' column to the result df (#473) (Yurii Romanyshyn) #473
    • f368333: pipeline model can also no be of type BaseEstimator (#468) (DBI) #468
    • b3e73d2: added favicons to the docs (#481) (Yurii Romanyshyn) #481
    • 33b4820: [issue-409] added datetime formating function to utils (#464) (Yurii Romanyshyn) #464
    • 31018fa: nothing to show on when pps is 0 on all fields (#480) (DBI) #480
    • ca7866b: Small typos fixes (#487) (nirhutnik) #487
    • 18436ed: Moving examples into docs (#495) (Itay Gabbay) #495
    • 9b3b0b8: [Automatic] Examples Update (#496) (deepchecks-bot) #496
    • 556c01a: Docs and readme fixes (#497) (Itay Gabbay) #497
    • 9d29f0e: [issue-431] modified suite output text (#499) (Yurii Romanyshyn) #499
    • f834aee: use_undeprecated_tqdm (#498) (JKL98ISR) #498
    • 421d332: fix errors related to string labels (#501) (JKL98ISR) #501
    • a54f6fe: Change condition of simple model check to use gain (#488) (matanper) #488
    • a33f540: Remove default from metrics (#507) (JKL98ISR) #507
    • 682c4cf: add dataset script and notebook for phishing url example (#485) (Noam Bressler) #485
    • 1112b97: Fix string labels in performance report / simple model comparison (#504) (JKL98ISR) #504
    • de8742f: dataset mutable properties fix (#509) (Yurii Romanyshyn) #509
    • a192c42: [Automatic] Examples Update (#510) (deepchecks-bot) #510
    • 44d69b9: Update README.rst (#511) (shir22) #511
    • 8076354: Limit permutation time (#512) (Noam Bressler) #512
    • f9e8484: 0.2.0 version bump (Itay Gabbay)
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.2.0-py3-none-any.whl(189.64 KB)
    deepchecks-0.2.0.tar.gz(128.65 KB)
    LICENSE(34.03 KB)
  • 0.1.3(Jan 4, 2022)

    Commits

    • d332548: Fix x axis to show categorical correct (#413) (matanper) #413
    • Fix legend position (#414) #414 (matanper)
    • 76f256e: [Automatic] Examples Update (#427) (deepchecks-bot) #427
    • fbe6da3: Fix code example of scorer note docstring (#426) (matanper) #426
    • 44b539a: Update README.md (#442) (shir22) #442
    • aa87c08: Roc plotly (#415) (JKL98ISR) #415
    • 4d22f3a: Changing tag in GA (#450) (Itay Gabbay) #450
    • b0c3576: Fix trying to calculate features importance on none model (#440) (matanper) #440
    • 6c61816: Add to suite output the checks that haven't run because of missing parameters (#436) (matanper) #436
    • ed5fdf1: catch condition errors (#449) (JKL98ISR) #449
    • 6e93077: Issue template (#448) (Itay Gabbay) #448
    • 00d6726: Update config.yml (Itay Gabbay)
    • cc315fd: test (#445) (nirhutnik) #445
    • 07f4624: move to plotly (#455) (DBI) #455
    • b8fc116: fix - validate_fitting failing when data contains only one label (Noam Bressler) #454
    • f3dd910: Changing README files from markdown to RST (#459) (Itay Gabbay) #459
    • 63af15f: Fix overlap of titles in drift plots (#462) (DBI) #462
    • c917845: re-adding widget progressbar (#460) (JKL98ISR) #460
    • b95dc48: remove label_name from dataset constructor (#456) (matanper) #456
    • 50a49e3: Updating long description (#463) (Itay Gabbay) #463
    • 09e6fba: [issue-439] added 'Number of samples' column to the result dataframe; (#461) (Yurii Romanyshyn) #461
    • 375ae16: Create pull_request_template.md (#465) (Itay Gabbay) #465
    • 6080e21: Updating long description (#471) (Itay Gabbay) #471
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.1.3-py3-none-any.whl(182.61 KB)
    deepchecks-0.1.3.tar.gz(124.58 KB)
    LICENSE(34.03 KB)
  • 0.1.1(Jan 2, 2022)

  • 0.1.0(Jan 2, 2022)

    Commits

    • d32acfd: Mlc 456 unused feature bug (#379) (nirhutnik) #379
    • 7c2d7a6: Label + prediction distribution (#380) (nirhutnik) #380
    • b92efaf: Mlc 185 drift for label and pred (#388) (nirhutnik) #388
    • e531c07: Make trust score plot cut off upper and lower 5% of data for display (#391) (Noam Bressler) #391
    • b994b21: Fixing conda packaging (#389) (Itay Gabbay) #389
    • 839f63a: Mlc 185 drift for label and pred (#396) (nirhutnik) #396
    • 7fc29b5: Simple model classes fix (#395) (matanper) #395
    • eb1300d: Fix confusion matrix heatmap (#398) (matanper) #398
    • ceee746: Add explanation about scorers to simple model check (#387) (matanper) #387
    • c3995e7: assume task type from model or label (#397) (DBI) #397
    • 700598c: Mlc 292 model error analysis (#378) (Noam Bressler) #378
    • e537e01: Temporary fix for groupby on category (#399) (matanper) #399
    • a18fc89: Merge performance report, Class Performance Imbalance and Train test Overfit into one check (#364) (JKL98ISR) #364
    • 5b8e1bc: drop nan before gausian_kde (#402) (DBI) #402
    • eb7e6bb: only one performance report (#406) (Noam Bressler) #406
    • df7ad4e: Fix model info to work with pipeline (#405) (matanper) #405
    • 652a1f4: Updated readme, user guide and how-to-guides (#340) (shir22) #340
    • f6b70e0: v0 (#410) (JKL98ISR) #410
    • 9279b06: [Automatic] Examples Update (#411) (deepchecks-bot) #411
    Source code(tar.gz)
    Source code(zip)
    deepchecks-0.1.0-py3-none-any.whl(181.52 KB)
    deepchecks-0.1.0.tar.gz(122.71 KB)
    LICENSE(34.03 KB)
Owner
null
machine learning model deployment project of Iris classification model in a minimal UI using flask web framework and deployed it in Azure cloud using Azure app service

This is a machine learning model deployment project of Iris classification model in a minimal UI using flask web framework and deployed it in Azure cloud using Azure app service. We initially made this project as a requirement for an internship at Indian Servers. We are now making it open to contribution.

Krishna Priyatham Potluri 73 Dec 1, 2022
Data science, Data manipulation and Machine learning package.

duality Data science, Data manipulation and Machine learning package. Use permitted according to the terms of use and conditions set by the attached l

David Kundih 3 Oct 19, 2022
Python package for machine learning for healthcare using a OMOP common data model

This library was developed in order to facilitate rapid prototyping in Python of predictive machine-learning models using longitudinal medical data from an OMOP CDM-standard database.

Sontag Lab 75 Jan 3, 2023
A collection of interactive machine-learning experiments: 🏋️models training + 🎨models demo

?? Interactive Machine Learning experiments: ??️models training + ??models demo

Oleksii Trekhleb 1.4k Jan 6, 2023
Apache Liminal is an end-to-end platform for data engineers & scientists, allowing them to build, train and deploy machine learning models in a robust and agile way

Apache Liminals goal is to operationalise the machine learning process, allowing data scientists to quickly transition from a successful experiment to an automated pipeline of model training, validation, deployment and inference in production. Liminal provides a Domain Specific Language to build ML workflows on top of Apache Airflow.

The Apache Software Foundation 121 Dec 28, 2022
High performance, easy-to-use, and scalable machine learning (ML) package, including linear model (LR), factorization machines (FM), and field-aware factorization machines (FFM) for Python and CLI interface.

What is xLearn? xLearn is a high performance, easy-to-use, and scalable machine learning package that contains linear model (LR), factorization machin

Chao Ma 3k Jan 8, 2023
easyNeuron is a simple way to create powerful machine learning models, analyze data and research cutting-edge AI.

easyNeuron is a simple way to create powerful machine learning models, analyze data and research cutting-edge AI.

Neuron AI 5 Jun 18, 2022
A Python Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming.

Master status: Development status: Package information: TPOT stands for Tree-based Pipeline Optimization Tool. Consider TPOT your Data Science Assista

Epistasis Lab at UPenn 8.9k Jan 9, 2023
Python Extreme Learning Machine (ELM) is a machine learning technique used for classification/regression tasks.

Python Extreme Learning Machine (ELM) Python Extreme Learning Machine (ELM) is a machine learning technique used for classification/regression tasks.

Augusto Almeida 84 Nov 25, 2022
MIT-Machine Learning with Python–From Linear Models to Deep Learning

MIT-Machine Learning with Python–From Linear Models to Deep Learning | One of the 5 courses in MIT MicroMasters in Statistics & Data Science Welcome t

null 2 Aug 23, 2022
ELI5 is a Python package which helps to debug machine learning classifiers and explain their predictions

A library for debugging/inspecting machine learning classifiers and explaining their predictions

null 154 Dec 17, 2022
Vowpal Wabbit is a machine learning system which pushes the frontier of machine learning with techniques

Vowpal Wabbit is a machine learning system which pushes the frontier of machine learning with techniques such as online, hashing, allreduce, reductions, learning2search, active, and interactive learning.

Vowpal Wabbit 8.1k Dec 30, 2022
CD) in machine learning projectsImplementing continuous integration & delivery (CI/CD) in machine learning projects

CML with cloud compute This repository contains a sample project using CML with Terraform (via the cml-runner function) to launch an AWS EC2 instance

Iterative 19 Oct 3, 2022
Uber Open Source 1.6k Dec 31, 2022
Upgini : data search library for your machine learning pipelines

Automated data search library for your machine learning pipelines → find & deliver relevant external data & features to boost ML accuracy :chart_with_upwards_trend:

Upgini 175 Jan 8, 2023
Python package for stacking (machine learning technique)

vecstack Python package for stacking (stacked generalization) featuring lightweight functional API and fully compatible scikit-learn API Convenient wa

Igor Ivanov 671 Dec 25, 2022
A Python Package to Tackle the Curse of Imbalanced Datasets in Machine Learning

imbalanced-learn imbalanced-learn is a python package offering a number of re-sampling techniques commonly used in datasets showing strong between-cla

null 6.2k Jan 1, 2023
Data Version Control or DVC is an open-source tool for data science and machine learning projects

Continuous Machine Learning project integration with DVC Data Version Control or DVC is an open-source tool for data science and machine learning proj

Azaria Gebremichael 2 Jul 29, 2021