[FEATURE] Implement R-Precision with PySpark #2087 #2219

demoncoder-crypto · 2025-04-06T17:21:22Z

Description

This pull request implements the R-Precision evaluation metric for PySpark within the SparkRankingEvaluation class.

This change is required to provide a standard ranking evaluation metric for recommendation models evaluated using PySpark, addressing feature request #2087 and achieving parity with potential implementations for other frameworks (related to #2086). It allows users to measure the fraction of relevant items within the top R recommendations, where R is the total number of relevant items for a user.

Related Issues

Addresses issue [FEATURE] Implement R-Precision with PySpark #2087: [FEATURE] Implement R-Precision with PySpark [FEATURE] Implement R-Precision with PySpark #2087

References

R-Precision is a standard metric for evaluating ranked retrieval results.

Checklist:

I have followed the contribution guidelines and code style for this project.
I have added tests covering my contributions.
I have updated the documentation accordingly. (Note: Docstrings were added, but no higher-level documentation files were modified).
I have signed the commits, e.g. git commit -s -m "your commit message". (Note: The commit was not signed).
This PR is being made to staging branch AND NOT TO main branch. (Assuming staging is the target development branch for this repository).

miguelgfierro

Thanks for the contribution! Really nice. I have some comments.

Also, could you please send the PR to staging instead of main? https://github.com/recommenders-team/recommenders/blob/main/CONTRIBUTING.md#steps-to-contributing

miguelgfierro · 2025-04-08T16:59:15Z

recommenders/evaluation/spark_evaluation.py

        self.col_rating = col_rating
        self.col_prediction = col_prediction
        self.threshold = threshold
+        self.rating_pred_raw = rating_pred # Store raw predictions before processing


self.rating_pred_raw is only used in the R-Precision. Either use directly self.rating_pred or do the data treatment internally in the function

recommenders/evaluation/spark_evaluation.py

miguelgfierro · 2025-04-08T17:29:32Z

@anargyri @SimonYansenZhao @loomlike can you please review?

demoncoder-crypto · 2025-04-20T15:48:53Z

Ok I am on it

demoncoder-crypto · 2025-04-23T21:32:41Z

I have made the changes as requested by you @miguelgfierro, I am currently testing jupyter notebook will provide an update on it very soon.

miguelgfierro · 2025-04-24T15:49:30Z

@demoncoder-crypto if you want to add the notebook, I think it would be interesting to extend https://github.com/recommenders-team/recommenders/blob/main/examples/03_evaluate/evaluation.ipynb here we have a large number of metrics. I think it would be valuable to explain what is R-precision and how is different to the other metrics.

demoncoder-crypto · 2025-04-24T16:46:10Z

Do you want me to implement r-precision in the evaluation.ipynb you mentioned or do it separate. @miguelgfierro. I will just complete the other issue implementation then look at the jupyter notebook you mentioned. Thanks for the support

miguelgfierro · 2025-04-24T17:05:04Z

I think it would make more sense in evaluation.ipynb, because all the metrics are there. I think it would be better for users to understand the differences

demoncoder-crypto · 2025-04-24T18:31:14Z

For sure I will do that

review-notebook-app · 2025-05-02T14:52:22Z

Check out this pull request on

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

demoncoder-crypto · 2025-05-02T14:54:24Z

@miguelgfierro I have implemented the changes as needed, do let me know if they work, I am now pushing the second issue I am working on and I will send the new PR to staging as requested. Thanks

miguelgfierro

I changed the branch from staging to main, there are some errors in the computation

miguelgfierro · 2025-05-03T08:03:08Z

examples/03_evaluate/evaluation.ipynb

+                "#### 2.2.8 R-Precision\n",
+                "\n",
+                "R-Precision evaluates the fraction of relevant items among the top R recommended items, where R is the total number of *truly* relevant items for a specific user. It's equivalent to Recall@R.\n",
+                "\n",
+                "**Difference from Precision@k:** Precision@k measures relevance within a fixed top *k* items, regardless of the total number of relevant items (R). R-Precision adapts the evaluation depth (*R*) based on the user's specific ground truth, making it potentially more user-centric when the number of relevant items varies significantly across users.\n",


I get an error when trying to compute the notebook:

--------------------------------------------------------------------------- AnalysisException Traceback (most recent call last) Cell In[19], [line 6](vscode-notebook-cell:?execution_count=19&line=6) [1](vscode-notebook-cell:?execution_count=19&line=1) # Note: The spark_rank_eval object was initialized with k=3. [2](vscode-notebook-cell:?execution_count=19&line=2) # R-Precision intrinsically uses R (number of relevant items for the user) as the cutoff. [3](vscode-notebook-cell:?execution_count=19&line=3) # The 'k' parameter passed during initialization doesn't directly affect R-Precision calculation itself, [4](vscode-notebook-cell:?execution_count=19&line=4) # but it might affect how the rating_pred dataframe is pre-processed if relevancy_method relies on k. [5](vscode-notebook-cell:?execution_count=19&line=5) # For a direct comparison with other metrics at a fixed k, ensure the underlying data processing is consistent. ----> [6](vscode-notebook-cell:?execution_count=19&line=6) print(f"The R-Precision is {spark_rank_eval.r_precision()}") File ~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:386, in SparkRankingEvaluation.r_precision(self) [384](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:384) # Rank predictions per user [385](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:385) window_spec = Window.partitionBy(self.col_user).orderBy(F.col(self.col_prediction).desc()) --> [386](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:386) ranked_preds = self.rating_pred.select( [387](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:387) self.col_user, self.col_item, self.col_prediction [388](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:388) ).withColumn("rank", F.row_number().over(window_spec)) [390](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:390) # Join ranked predictions with R [391](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:391) preds_with_r = ranked_preds.join( [392](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:392) ground_truth_with_R.select(self.col_user, "R"), on=self.col_user [393](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/MS/review/demoncoder-recommenders/recommenders/evaluation/spark_evaluation.py:393) ) File ~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3227, in DataFrame.select(self, *cols) [3182](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3182) def select(self, *cols: "ColumnOrName") -> "DataFrame": # type: ignore[misc] [3183](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3183) """Projects a set of expressions and returns a new :class:`DataFrame`. [3184](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3184) [3185](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3185) .. versionadded:: 1.3.0 (...) [3225](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3225) +-----+---+ [3226](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3226) """ -> [3227](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3227) jdf = self._jdf.select(self._jcols(*cols)) [3228](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3228) return DataFrame(jdf, self.sparkSession) File ~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1322, in JavaMember.__call__(self, *args) [1316](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1316) command = proto.CALL_COMMAND_NAME +\ [1317](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1317) self.command_header +\ [1318](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1318) args_command +\ [1319](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1319) proto.END_COMMAND_PART [1321](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1321) answer = self.gateway_client.send_command(command) -> [1322](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1322) return_value = get_return_value( [1323](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1323) answer, self.gateway_client, self.target_id, self.name) [1325](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1325) for temp_arg in temp_args: [1326](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1326) if hasattr(temp_arg, "_detach"): File ~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:185, in capture_sql_exception.<locals>.deco(*a, **kw) [181](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:181) converted = convert_exception(e.java_exception) [182](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:182) if not isinstance(converted, UnknownException): [183](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:183) # Hide where the exception came from that shows a non-Pythonic [184](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:184) # JVM exception message. --> [185](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:185) raise converted from None [186](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:186) else: [187](https://vscode-remote+wsl-002bubuntu-002d20-002e04.vscode-resource.vscode-cdn.net/home/miguel/MS/review/demoncoder-recommenders/examples/03_evaluate/~/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:187) raise AnalysisException: [UNRESOLVED_COLUMN.WITH_SUGGESTION] A column or function parameter with name `MovieId` cannot be resolved. Did you mean one of the following? [`UserId`, `prediction`].; 'Project [UserId#6L, 'MovieId, 'Rating] +- Aggregate [UserId#6L], [UserId#6L, collect_list(MovieId#7L, 0, 0) AS prediction#237] +- Filter (rank#227 <= 3) +- Project [UserId#6L, MovieId#7L, Rating#8L, rank#227] +- Project [UserId#6L, MovieId#7L, Rating#8L, rank#227, rank#227] +- Window [row_number() windowspecdefinition(UserId#6L, Rating#8L DESC NULLS LAST, specifiedwindowframe(RowFrame, unboundedpreceding$(), currentrow$())) AS rank#227], [UserId#6L], [Rating#8L DESC NULLS LAST] +- Project [UserId#6L, MovieId#7L, Rating#8L] +- LogicalRDD [UserId#6L, MovieId#7L, Rating#8L], false

miguelgfierro · 2025-05-03T08:04:05Z

tests/unit/recommenders/evaluation/test_spark_evaluation.py

    assert evaluator.serendipity() == pytest.approx(0.4028, TOL)
+
+
+@pytest.mark.spark


same with the tests:

$ pytest tests/unit/recommenders/evaluation/test_spark_evaluation.py ======================================================================== test session starts ========================================================================= platform linux -- Python 3.11.11, pytest-8.2.2, pluggy-1.5.0 rootdir: /home/miguel/MS/review/demoncoder-recommenders configfile: pyproject.toml plugins: anyio-4.4.0, cov-5.0.0, typeguard-4.3.0, hypothesis-6.104.2, mock-3.14.0 collected 30 items tests/unit/recommenders/evaluation/test_spark_evaluation.py .............................F [100%] ============================================================================== FAILURES ============================================================================== _______________________________________________________________________ test_spark_r_precision _______________________________________________________________________ spark_data = (DataFrame[userID: bigint, itemID: bigint, rating: bigint], DataFrame[userID: bigint, itemID: bigint, prediction: bigint, rating: bigint]) @pytest.mark.spark def test_spark_r_precision(spark_data): df_true, df_pred = spark_data # Test perfect prediction (R-Precision should be 1.0) evaluator_perfect = SparkRankingEvaluation(df_true, df_true, col_prediction="rating") > assert evaluator_perfect.r_precision() == pytest.approx(1.0, TOL) tests/unit/recommenders/evaluation/test_spark_evaluation.py:526: _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ recommenders/evaluation/spark_evaluation.py:386: in r_precision ranked_preds = self.rating_pred.select( ../../../anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/dataframe.py:3227: in select jdf = self._jdf.select(self._jcols(*cols)) ../../../anaconda/envs/recommenders311/lib/python3.11/site-packages/py4j/java_gateway.py:1322: in __call__ return_value = get_return_value( _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ a = ('xro5784', <py4j.clientserver.JavaClient object at 0x7fb0e286f110>, 'o5697', 'select'), kw = {}, converted = AnalysisException() def deco(*a: Any, **kw: Any) -> Any: try: return f(*a, **kw) except Py4JJavaError as e: converted = convert_exception(e.java_exception) if not isinstance(converted, UnknownException): # Hide where the exception came from that shows a non-Pythonic # JVM exception message. > raise converted from None E pyspark.errors.exceptions.captured.AnalysisException: [UNRESOLVED_COLUMN.WITH_SUGGESTION] A column or function parameter with name `itemID` cannot be resolved. Did you mean one of the following? [`userID`, `prediction`].; E 'Project [userID#4984L, 'itemID, 'rating] E +- Aggregate [userID#4984L], [userID#4984L, collect_list(itemID#4985L, 0, 0) AS prediction#5009] E +- Filter (rank#4999 <= 10) E +- Project [userID#4984L, itemID#4985L, rating#4986L, rank#4999] E +- Project [userID#4984L, itemID#4985L, rating#4986L, rank#4999, rank#4999] E +- Window [row_number() windowspecdefinition(userID#4984L, rating#4986L DESC NULLS LAST, specifiedwindowframe(RowFrame, unboundedpreceding$(), currentrow$())) AS rank#4999], [userID#4984L], [rating#4986L DESC NULLS LAST] E +- Project [userID#4984L, itemID#4985L, rating#4986L] E +- LogicalRDD [userID#4984L, itemID#4985L, rating#4986L], false ../../../anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/errors/exceptions/captured.py:185: AnalysisException ------------------------------------------------------------------------ Captured stderr call ------------------------------------------------------------------------ ========================================================================== warnings summary ========================================================================== tests/unit/recommenders/evaluation/test_spark_evaluation.py: 164 warnings /home/miguel/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/pandas/utils.py:37: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead. if LooseVersion(pandas.__version__) < LooseVersion(minimum_pandas_version): tests/unit/recommenders/evaluation/test_spark_evaluation.py: 181 warnings /home/miguel/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/pandas/conversion.py:485: DeprecationWarning: is_datetime64tz_dtype is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.DatetimeTZDtype)` instead. if should_localize and is_datetime64tz_dtype(s.dtype) and s.dt.tz is not None: tests/unit/recommenders/evaluation/test_spark_evaluation.py: 15 warnings /home/miguel/anaconda/envs/recommenders311/lib/python3.11/site-packages/pyspark/sql/context.py:158: FutureWarning: Deprecated in 3.0.0. Use SparkSession.builder.getOrCreate() instead. warnings.warn( -- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html ====================================================================== short test summary info ======================================================================= FAILED tests/unit/recommenders/evaluation/test_spark_evaluation.py::test_spark_r_precision - pyspark.errors.exceptions.captured.AnalysisException: [UNRESOLVED_COLUMN.WITH_SUGGESTION] A column or function parameter with name `itemID` cannot be resolved. D... ======================================================= 1 failed, 29 passed, 360 warnings in 247.48s (0:04:07) =======================================================

miguelgfierro · 2025-05-03T08:05:58Z

examples/03_evaluate/evaluation.ipynb

                    "output_type": "stream",
                    "text": [
-                        "\r",
+                        "\r\n",


when you submit the fixed version, could you please run the whole notebook so people can see the results?

Please use python 3.11 which is the latest one we support.

feat: Implement R-Precision metric for Spark

a4a00fa

demoncoder-crypto requested review from SimonYansenZhao, anargyri, gramhagen, loomlike, miguelgfierro and wav8k as code owners April 6, 2025 17:21

demoncoder-crypto mentioned this pull request Apr 6, 2025

[FEATURE] Implement R-Precision with PySpark #2087

Open

miguelgfierro reviewed Apr 8, 2025

View reviewed changes

miguelgfierro mentioned this pull request Apr 8, 2025

Feature/pytorch embedding ranker #2220

Closed

Fix: Addressed code review feedback for R-Precision implementation

bfaccc5

Docs: Add R-Precision explanation and example to evaluation notebook

01ee124

demoncoder-crypto requested a review from miguelgfierro May 2, 2025 14:54

miguelgfierro changed the base branch from main to staging May 3, 2025 07:48

miguelgfierro reviewed May 3, 2025

View reviewed changes

		assert evaluator.serendipity() == pytest.approx(0.4028, TOL)


		@pytest.mark.spark

[FEATURE] Implement R-Precision with PySpark #2087 #2219

Are you sure you want to change the base?

[FEATURE] Implement R-Precision with PySpark #2087 #2219

Uh oh!

Conversation

demoncoder-crypto commented Apr 6, 2025

Description

Related Issues

References

Checklist:

Uh oh!

miguelgfierro left a comment

Choose a reason for hiding this comment

Uh oh!

miguelgfierro Apr 8, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

miguelgfierro commented Apr 8, 2025

Uh oh!

demoncoder-crypto commented Apr 20, 2025

Uh oh!

demoncoder-crypto commented Apr 23, 2025

Uh oh!

miguelgfierro commented Apr 24, 2025

Uh oh!

demoncoder-crypto commented Apr 24, 2025

Uh oh!

miguelgfierro commented Apr 24, 2025

Uh oh!

demoncoder-crypto commented Apr 24, 2025

Uh oh!

review-notebook-app bot commented May 2, 2025

Uh oh!

demoncoder-crypto commented May 2, 2025

Uh oh!

miguelgfierro left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

miguelgfierro May 3, 2025

Choose a reason for hiding this comment

Uh oh!

miguelgfierro May 3, 2025

Choose a reason for hiding this comment

Uh oh!

miguelgfierro May 3, 2025

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

miguelgfierro left a comment •

edited

Loading