dream-faster/krisi issues and pull requests

#255 - fix(Library): disable benchmarkling for kappa & s-score

Pull Request - State: closed - Opened by almostintuitive 9 months ago

#254 - feature(Benchmarking): Optional property to disable benchmarking on a metric.

Pull Request - State: closed - Opened by szemyd 9 months ago - 1 comment

#253 - fix(Reporting): Benchmarking didn't call model predict.

Pull Request - State: closed - Opened by szemyd 11 months ago

#252 - feature(Benchmarking): Added `zscore` function.

Pull Request - State: closed - Opened by szemyd 11 months ago

#251 - feature(Benchmarking): Benchmarking is not a Group anymore.

Pull Request - State: closed - Opened by szemyd 11 months ago

#250 - feature(Metric): Metric is returned after evaluating.

Pull Request - State: closed - Opened by szemyd 11 months ago

#249 - feature(Metric): Deepcopy metric with `call`.

Pull Request - State: closed - Opened by szemyd 12 months ago - 1 comment

#248 - fix(Metric): Passing in sample_weights.

Pull Request - State: closed - Opened by szemyd 12 months ago

#247 - fix(Benchmarking): Name of chunked model should be also simply NS.

Pull Request - State: closed - Opened by szemyd 12 months ago

#246 - fix(Benchmarking): Random model now receives `X` and random chunking fixed.

Pull Request - State: closed - Opened by szemyd 12 months ago

#245 - feature(ScoreCard): Union logic on comparison scores.

Pull Request - State: closed - Opened by szemyd 12 months ago

#244 - feature(Library): added back brier score, removed calibration from imbalanced_metrics

Pull Request - State: closed - Opened by almostintuitive 12 months ago

#243 - fix(ScoreCard): Fail no unknown key.

Pull Request - State: closed - Opened by szemyd 12 months ago

#242 - feature(ScoreCard): Index by comparison result

Pull Request - State: closed - Opened by szemyd almost 1 year ago

#241 - feature(Scorecard): Scorecard now has the option to get comparison results.

Pull Request - State: closed - Opened by szemyd almost 1 year ago

#240 - refactor(Benchmarking): rename `RandomClassifierChunked` to `ShufflePredictionsByChunk`

Issue - State: open - Opened by almostintuitive about 1 year ago

#239 - fix(ScoreCard): `get_ds()` should not return `Group` type.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#238 - fix(Union): Only apply function if the result is an int or a float.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#237 - chore(Renaming): result to value

Pull Request - State: closed - Opened by szemyd about 1 year ago

#236 - feature(ScoreCard): Benchmark metrics now retrievable with `get_ds`.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#235 - fix(Presets): adjusted balanced metric presets

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#234 - feature(Metrics): updated default metrics, threshold to start using imbalanced metrics

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#233 - feature(Reporting & Types): ScoreCard arithmetic and `sample_type` can be also a string.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#232 - fix(Printing): Handling debug and repr printing properly.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#231 - fix(Printing): Took out annoying printing in debug mode.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#230 - feature(ScoreCard): ScoreCard `pd.Series` function should return metric keys as indexes.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#229 - fix(Metrics): don't calculate s-score on a rolling basis by default

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#228 - Fix: don't print to `stdout` in `repr()`, return the string instead

Issue - State: closed - Opened by almostintuitive about 1 year ago

#227 - feature(Metrics): don't let roc_auc fail if there's no instance of the other class

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#226 - feature(Types): Updated types to be ParsableEnums.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#225 - chore(Utils): Smaller synthetic dataset to speed up testing.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#224 - feature(Utils): Combinatorial shuffling.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#223 - feature(Util): Handle datetime index in shuffling in chunks.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#222 - fix(Utils): shuffle_df_in_chunks should only shuffle indices, not the whole dataframe

Pull Request - State: closed - Opened by almostintuitive about 1 year ago - 2 comments

#221 - feature(Evaluation): New SampleTypes - Validation Sample.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#220 - feature(Modeling): Random model shuffling now accpets float as chunk_size.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#219 - fix(Metrics): disable rolling metrics for roc_auc

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#218 - feature(Data): Shuffling in chunks updated to work with `copy_on_write = True`.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#217 - chore(Exporting): Export all metrics to be accessed from `krisi.library`.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#216 - fix(Evaluation): Benchmark metrics were switched around.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#215 - chore(Analysis): Matplotlib and Statsmodel dependency optional now.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#214 - fix(Evaluate): Random chunking now supports length that are not wholly divisible by chunk_size.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#213 - feature(Evaluate): Random model has a shuffled chunk version.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#212 - feature(Metrics): added avg_precision metric

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#211 - feature(Metrics): pred_y_imbalance_ratio is now the absolute difference between the ratio of y and pred's imbalance

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#210 - fix(Metric): Imbalance warning

Pull Request - State: closed - Opened by szemyd about 1 year ago

#209 - fix(ScoreCard): ScoreCard indexing should return a copy not the actual object.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#208 - feature(Analysis): Random model updated to have a smoothing factor.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#207 - feature(Library): Added imbalance metrics.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#206 - fix(Calibration): propagate pos_label throughout

Pull Request - State: closed - Opened by almostintuitive about 1 year ago

#205 - feature(Utils): Export `ParsableEnum`.

Pull Request - State: closed - Opened by szemyd about 1 year ago

#204 - feature(Library): added distinction between balanced vs imbalanced metrics

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#203 - feature(Utils): Exposing cross-package functions.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#202 - fix(Group): Group is now handled separately

Pull Request - State: closed - Opened by szemyd over 1 year ago

#201 - fix(Dev): Raise error when `raise_exception` passed in.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#200 - fix(Library): put back rmse to default metrics

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#199 - refactor(Types): move Task to the top

Issue - State: closed - Opened by almostintuitive over 1 year ago

#198 - fix(Data): Syntethetic data generation.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#197 - feature(Test): Don't show plots when we are testing.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#196 - fix(Tests): don't use matplotlib if not installed

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#195 - feature(Evaluation): Benchmarking metrics to a random model.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#194 - feature(Checks): check if sample_weight has the same length as target

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#193 - feature(State): Added global state that allows us to throw errors in try except when in `RunType.test`. Fixed S-Curve

Pull Request - State: closed - Opened by szemyd over 1 year ago

#192 - fix(Evaluation): set default rolling_args to a less error-prone window and step size

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#191 - feature(Library): added weighted metrics to binary presets

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#190 - feature(Report): Calibration plot in console & took out `rmse` from default metrics.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#189 - chore(Library): Take out `rmse` from `default_regression_metrics`.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#188 - feature(Library): Added Bennet's et. al `S-Score` classification metric.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#187 - fix(Library): use ovr instead of ovo for roc_auc

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#186 - fix(Evaluation): Rolling evaluation shouldn't assume the name of the arguments.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#185 - fix(Report): Callibration plot.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#184 - fix(Evaluation): Sample weight rolling window fix.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#183 - fix(Printing): Scorecard get's printed multiple times to console.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#182 - feature(Evaluation): Added support for `sample_weights`.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#181 - feature(Metrics): added ncdg, changed default metrics slightly

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#180 - feature(Group): Metric consistency over time.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#179 - fix(Utils): correctly recognize dataset types

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#178 - feature(Scorecard): missing type annotation for getitem

Issue - State: closed - Opened by almostintuitive over 1 year ago

#177 - fix(Printing): scorecard is always printed in full

Issue - State: closed - Opened by almostintuitive over 1 year ago

#176 - fix(Metrics): only used weighted and macro roc_auc score for multi-label

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#175 - fix(Metrics): wrap roc_auc score so it works when doing binary classification

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#174 - feature(Metrics): added binary vs multiclass metrics, get default metrics automatically

Pull Request - State: closed - Opened by almostintuitive over 1 year ago

#173 - feature(Scoring): add ability to pass in and propagate sample_weights

Issue - State: closed - Opened by almostintuitive over 1 year ago

#172 - feature(Classification): detect if its a binary or multi-class classification and set score parameters accordingly

Issue - State: closed - Opened by almostintuitive over 1 year ago

#171 - fix(Reports): maybe `direct_save` should be called `images`?

Issue - State: open - Opened by almostintuitive over 1 year ago

#170 - fix(Metrics): calibration curve could/should be calculating the ratio of positives based on the sum of all instaces

Issue - State: closed - Opened by almostintuitive over 1 year ago

#169 - fix(Scoring): some classification tasks are not recognized as such

Issue - State: closed - Opened by almostintuitive over 1 year ago

#168 - fix(Evaluation): Brier score default metric fixed when columns not set correctly. Updated in and out of sample scoring function. Added test.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#167 - feature(Docs): Add references to new metrics and multiclass classification in `README`.

Issue - State: open - Opened by szemyd over 1 year ago

#166 - fix(ScoreCard): `ipython` should only be imported if notebook is available.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#165 - chore(Package): Bump `krisi` major and references to it in other libraries.

Issue - State: closed - Opened by szemyd over 1 year ago

#164 - feature(Report): Calibration plot and many fixes.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#163 - feature(Test): Pytest should run examples directly.

Issue - State: closed - Opened by szemyd over 1 year ago

#162 - feature(Scoring): Handle probabilities and multiclass classification.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#161 - fix(ScoreCard): We should clean the directory first in tests and misspelling.

Pull Request - State: closed - Opened by szemyd over 1 year ago

#160 - feature(ScoreCard): Saving now handles overwriting based on model_name

Pull Request - State: closed - Opened by szemyd over 1 year ago

#159 - fix(ScoreCard): Save path shouldn't be simply `scorecard`, it should default to the scorecard name.

Issue - State: closed - Opened by szemyd over 1 year ago

#158 - fix(Report): No module named `ipython`

Issue - State: closed - Opened by szemyd over 1 year ago

#157 - feature(Report): Resample data before printing graph to console so that it is always readable.

Issue - State: open - Opened by szemyd over 1 year ago

#156 - feature(Docs): Updated walkthrough

Pull Request - State: closed - Opened by szemyd over 1 year ago

GitHub / dream-faster/krisi issues and pull requests