remove telemetry (#347)

* removing telemetry * remoes telemetry * update * update * test * update
ploomber · Sep 18, 2024 · 804da4c · 804da4c
1 parent 96456ee
commit 804da4c
Show file tree

Hide file tree

Showing 32 changed files with 9 additions and 451 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -5,23 +5,19 @@ jobs:
   unit:
     strategy:
       matrix:
-        python-version: [3.7, 3.8, 3.9, '3.10', '3.11']
+        python-version: ['3.10', '3.11']
         os: [ubuntu-latest, windows-latest, macos-latest]
     runs-on: ${{ matrix.os }}
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - name: Set up Python ${{ matrix.python-version }}
-      uses: conda-incubator/setup-miniconda@v2
+      uses: actions/setup-python@v5
       with:
-        miniconda-version: latest
-        activate-environment: conda-env
-        channels: conda-forge, defaults
         python-version: ${{ matrix.python-version }}
 
     - name: Lint
       shell: bash -l {0}
       run: |
-        eval "$(conda shell.bash hook)"
         pip install pkgmt
         pkgmt lint
 
@@ -32,7 +28,6 @@ jobs:
         PYTHON_VERSION: ${{ matrix.python-version }}
       shell: bash -l {0}
       run: |
-        eval "$(conda shell.bash hook)"
         # if we remove the --editable flag pytest throws an error, because there
         # are two copies of the pkg (src/ and site-packages/), this is a quick
         # way to fix it
@@ -48,9 +43,9 @@ jobs:
   check:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v4
       - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v2
+        uses: actions/setup-python@v5
         with:
           python-version: '3.10'
       - name: Install dependencies

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -2,6 +2,8 @@
 
 ## 0.12.2dev
 
+- [Feature] Removes telemetry
+
 ## 0.12.1 (2024-02-07)
 
 - [Feature] Adds `ploomber-extension` as dependency

diff --git a/doc/reference/community.md b/doc/reference/community.md
@@ -4,6 +4,3 @@
 
 [See here](https://docs.ploomber.io/en/latest/community/coc.html)
 
-## Telemetry
-
-[See here](https://docs.ploomber.io/en/latest/community/user-stats.html)
diff --git a/doc/reference/developer.md b/doc/reference/developer.md
@@ -403,53 +403,3 @@ See the [`roc`](https://github.com/ploomber/sklearn-evaluation/blob/8056bc31ec5e
 The `ax` object must be returned at the end of the function.
 
 See the [`roc`](https://github.com/ploomber/sklearn-evaluation/blob/8056bc31ec5e372102d0ee5ada988e380b077c4b/src/sklearn_evaluation/plot/roc.py#L45) function for an example.
-
-## Telemetry
-
-Monitoring the state of `sklearn-evaluation`
-
-Use [`SKLearnEvaluationLogger`](https://github.com/ploomber/sklearn-evaluation/blob/f32c15a43f4a9b4c2e588b3c0f71ba6dc5a71a7e/src/sklearn_evaluation/telemetry.py#L19) decorator to generate logs
-
-Example:
-
-```python
-@SKLearnEvaluationLogger.log(feature='plot')
-def confusion_matrix(
-        y_true,
-        y_pred,
-        target_names=None,
-        normalize=False,
-        cmap=None,
-        ax=None,
-        **kwargs):
-pass
-```
-
-this will generate the following log:
-
-```json
-        {
-          "metadata": {
-          "action": "confusion_matrix"
-          "feature": "plot",
-          "args": {
-                        "target_names": "None",
-                        "normalize": "False",
-                        "cmap": "None",
-                        "ax": "None"
-                    }
-          }
-        }
-```
-
-\*\* since `y_true` and `y_pred` are positional arguments without default values it won't log them
-
-### Queries
-
-1. Run queries and filter out `sklearn-evaluation` events by the event name: `sklearn-evaluation`
-2. Break these events by feature ('plot', 'report', 'SQLiteTracker', 'NotebookCollection')
-3. Break events by actions/func name (i.e: 'confusion_matrix', 'roc', etc...)
-
-### Errors
-
-Failing runnings will be named: `sklearn-evaluation-error`
diff --git a/setup.py b/setup.py
@@ -22,7 +22,7 @@
     "jupytext",
     "ploomber-engine",
     "ipykernel",
-    "pytest",
+    "pytest<8",
     # need to pin this version because pytest 4 breaks matplotlib image
     # comparison tests
     "pytest-cov",

diff --git a/src/sklearn_evaluation/__init__.py b/src/sklearn_evaluation/__init__.py
@@ -5,7 +5,6 @@
 from sklearn_evaluation.nb.NotebookCollection import NotebookCollection
 from sklearn_evaluation.nb.NotebookDatabase import NotebookDatabase
 from sklearn_evaluation.tracker import SQLiteTracker
-from sklearn_evaluation.telemetry import telemetry
 
 __all__ = [
     "ClassifierEvaluator",
@@ -14,5 +13,3 @@
     "NotebookCollection",
     "NotebookDatabase",
 ]
-
-telemetry.log_api("imported")
diff --git a/src/sklearn_evaluation/evaluator.py b/src/sklearn_evaluation/evaluator.py
@@ -5,7 +5,6 @@
 from sklearn_evaluation.report.report import Report
 from sklearn_evaluation.util import estimator_type, class_name
 from sklearn_evaluation import plot
-from sklearn_evaluation.telemetry import SKLearnEvaluationLogger
 
 
 class ClassifierEvaluator(object):
@@ -153,7 +152,6 @@ def html_serializable(self):
         """
         return EvaluatorHTMLSerializer(self)
 
-    @SKLearnEvaluationLogger.log(feature="report")
     def make_report(self, template=None):
         """
         Make HTML report

diff --git a/src/sklearn_evaluation/grid/random_forest_classifier_grid.py b/src/sklearn_evaluation/grid/random_forest_classifier_grid.py
@@ -5,7 +5,6 @@
 from sklearn_evaluation.grid.classifier_grid import AbstractClassifierGrid, GridTypes
 from sklearn.utils.validation import check_consistent_length
 import warnings
-from sklearn_evaluation.telemetry import SKLearnEvaluationLogger
 
 
 class RandomForestClassifierGrid(AbstractClassifierGrid):
@@ -49,7 +48,6 @@ class RandomForestClassifierGrid(AbstractClassifierGrid):
         }
     )
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-init")
     def __init__(self, grid, cv=3, verbose=0):
         """
         A random forest classifier grid.
@@ -96,7 +94,6 @@ def __init__(self, grid, cv=3, verbose=0):
             verbose=verbose,
         )
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-fit")
     def fit(self, X, y):
         """
         Fit estimator.
@@ -121,7 +118,6 @@ def fit(self, X, y):
         self.grid_search_cv_.fit(X, y, sample_weight=None)
         return self
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-set-test-data")
     def set_test_data(self, X_test, y_test) -> None:
         """
         Set the test data
@@ -140,7 +136,6 @@ def set_test_data(self, X_test, y_test) -> None:
         self.X_test = X_test
         self.y_test = y_test
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-confusion-matrix")
     def confusion_matrix(self):
         """
         Plots a confusion matrix based on `GridSearchCV.best_estimator_`.
@@ -160,7 +155,6 @@ def confusion_matrix(self):
         y_pred = self.grid_search_cv_.best_estimator_.predict(X_test)
         return plot.confusion_matrix(y_test, y_pred)
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-roc")
     def roc(self):
         """
         Plots an ROC based on `GridSearchCV.best_estimator_`.
@@ -185,7 +179,6 @@ def roc(self):
 
         return plot.roc(y_test, y_pred)
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-feature-importances")
     def feature_importances(self):
         """
         Plots feature importances based on `GridSearchCV.best_estimator_`.
@@ -203,7 +196,6 @@ def feature_importances(self):
         feature_importances = self.grid_search_cv_.best_estimator_.feature_importances_
         return plot.feature_importances(feature_importances)
 
-    @SKLearnEvaluationLogger.log("RandomForestClassifierGrid-grid-search-results")
     def grid_search_results(self, change="n_estimators", kind="line"):
         """
         Plots grid search results based on `GridSearchCV.best_estimator_`.

diff --git a/src/sklearn_evaluation/nb/NotebookCollection.py b/src/sklearn_evaluation/nb/NotebookCollection.py
@@ -15,7 +15,6 @@
 from sklearn_evaluation.nb.NotebookIntrospector import NotebookIntrospector
 from sklearn_evaluation.nb.sets import differences
 from sklearn_evaluation.table import Table
-from sklearn_evaluation.telemetry import SKLearnEvaluationLogger
 
 _env = Environment(loader=PackageLoader("sklearn_evaluation", "assets/nb"))
 _fm = black.FileMode(string_normalization=False, line_length=40)
@@ -41,9 +40,6 @@ class NotebookCollection(Mapping):
         as identifier (ignores extension)
     """  # noqa
 
-    @SKLearnEvaluationLogger.log(
-        feature="NotebookCollection", action="init-NotebookCollection"
-    )
     def __init__(self, paths, ids=None, scores=False):
         if ids is None:
             ids = paths

diff --git a/src/sklearn_evaluation/plot/calibration.py b/src/sklearn_evaluation/plot/calibration.py
@@ -34,7 +34,6 @@
 from sklearn_evaluation import __version__
 from sklearn_evaluation.util import isiterofiter
 from ploomber_core.exceptions import modify_exceptions
-from sklearn_evaluation.telemetry import SKLearnEvaluationLogger
 from sklearn_evaluation.plot.plot import AbstractComposedPlot, AbstractPlot
 from sklearn_evaluation.plot.style import apply_theme, get_color_palette
 
@@ -147,7 +146,6 @@ class CalibrationCurve(AbstractPlot):
     """
 
     @modify_exceptions
-    @SKLearnEvaluationLogger.log(feature="plot", action="calibration-curve-init")
     def __init__(
         self,
         mean_predicted_value,
@@ -304,7 +302,6 @@ def _get_data(self):
 
 class CalibrationCurveAdd(AbstractComposedPlot):
     @modify_exceptions
-    @SKLearnEvaluationLogger.log(feature="plot", action="calibration-curve-add-init")
     def __init__(
         self,
         mean_predicted_value_list,
@@ -360,7 +357,6 @@ def plot(self, ax=None):
 
 
 @modify_exceptions
-@SKLearnEvaluationLogger.log(feature="plot")
 def calibration_curve(
     y_true, probabilities, clf_names=None, n_bins=10, cmap="nipy_spectral", ax=None
 ):

diff --git a/src/sklearn_evaluation/plot/classification.py b/src/sklearn_evaluation/plot/classification.py
@@ -10,7 +10,6 @@
 from sklearn.metrics import confusion_matrix as sk_confusion_matrix
 
 from sklearn_evaluation import __version__
-from sklearn_evaluation.telemetry import SKLearnEvaluationLogger
 from sklearn_evaluation.plot.matplotlib import bar
 from sklearn_evaluation.metrics import precision_at
 from sklearn_evaluation import compute
@@ -93,7 +92,6 @@ class ConfusionMatrix(AbstractPlot):
         Added ``cmap`` argument
     """
 
-    @SKLearnEvaluationLogger.log(feature="plot", action="confusion-matrix-init")
     @modify_exceptions
     def __init__(self, cm, *, target_names=None, normalize=False, cmap=None):
         self.cm = cm
@@ -112,14 +110,12 @@ def plot(self, ax=None):
 
         return self
 
-    @SKLearnEvaluationLogger.log(feature="plot", action="confusion-matrix-sub")
     def __sub__(self, other):
         cm = self.cm - other.cm
         obj = ConfusionMatrixSub(cm, self.target_names)
         obj.plot()
         return obj
 
-    @SKLearnEvaluationLogger.log(feature="plot", action="confusion-matrix-add")
     def __add__(self, other):
         obj = ConfusionMatrixAdd(self.cm, other.cm, self.target_names)
         obj.plot()
@@ -178,7 +174,6 @@ def _confusion_matrix(y_true, y_pred, normalize):
     return cm
 
 
-@SKLearnEvaluationLogger.log(feature="plot")
 @modify_exceptions
 def confusion_matrix(
     y_true, y_pred, target_names=None, normalize=False, cmap=None, ax=None
@@ -314,7 +309,6 @@ def _plot_cm(cm, cmap, ax, target_names, normalize):
 
 
 # http://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html
-@SKLearnEvaluationLogger.log(feature="plot")
 @modify_exceptions
 def feature_importances(
     data, top_n=None, feature_names=None, orientation="horizontal", ax=None
@@ -371,7 +365,6 @@ def feature_importances(
     return ax
 
 
-@SKLearnEvaluationLogger.log(feature="plot")
 @modify_exceptions
 def precision_at_proportions(y_true, y_score, ax=None):
     """

diff --git a/src/sklearn_evaluation/plot/classification_report.py b/src/sklearn_evaluation/plot/classification_report.py
@@ -4,7 +4,6 @@
 import numpy as np
 import matplotlib.pyplot as plt
 from sklearn.metrics import classification_report as sk_classification_report
-from sklearn_evaluation.telemetry import SKLearnEvaluationLogger
 
 from sklearn_evaluation.plot.classification import _add_values_to_matrix
 from sklearn_evaluation.plot.plot import AbstractPlot, AbstractComposedPlot
@@ -85,7 +84,6 @@ class ClassificationReport(AbstractPlot):
 
     """
 
-    @SKLearnEvaluationLogger.log(feature="plot", action="classification-report-init")
     def __init__(
         self,
         matrix,
@@ -109,13 +107,11 @@ def plot(self, ax=None):
 
         return self
 
-    @SKLearnEvaluationLogger.log(feature="plot", action="classification-report-sub")
     def __sub__(self, other):
         return ClassificationReportSub(
             self.matrix, other.matrix, self.keys, target_names=self.target_names
         ).plot()
 
-    @SKLearnEvaluationLogger.log(feature="plot", action="classification-report-add")
     def __add__(self, other):
         return ClassificationReportAdd(
             self.matrix, other.matrix, keys=self.keys, target_names=self.target_names
Original file line number	Diff line number	Diff line change
Expand Up		@@ -4,6 +4,3 @@

		[See here](https://docs.ploomber.io/en/latest/community/coc.html)

		## Telemetry

		[See here](https://docs.ploomber.io/en/latest/community/user-stats.html)