vaex-ml 0.18.3__py3-none-any.whl → 0.19.0__py3-none-any.whl

vaex/ml/_version.py

@@ -1,2 +1,2 @@
-__version__ = '0.18.3'
-__version_tuple__ = (0, 18, 3)
+__version__ = '0.19.0'
+__version_tuple__ = (0, 19, 0)

vaex/ml/catboost.py

@@ -10,50 +10,49 @@ from . import generate
 import numpy as np
 import catboost
 
-
 @vaex.serialize.register
 @generate.register
 class CatBoostModel(state.HasState):
     '''The CatBoost algorithm.
 
-    This class provides an interface to the CatBoost aloritham.
-    CatBoost is a fast, scalable, high performance Gradient Boosting on
-    Decision Trees library, used for ranking, classification, regression and
-    other machine learning tasks. For more information please visit
-    https://github.com/catboost/catboost
-
-    Example:
-
-    >>> import vaex
-    >>> import vaex.ml.catboost
-    >>> df = vaex.datasets.iris()
-    >>> features = ['sepal_width', 'petal_length', 'sepal_length', 'petal_width']
-    >>> df_train, df_test = df.ml.train_test_split()
-    >>> params = {
-        'leaf_estimation_method': 'Gradient',
-        'learning_rate': 0.1,
-        'max_depth': 3,
-        'bootstrap_type': 'Bernoulli',
-        'objective': 'MultiClass',
-        'eval_metric': 'MultiClass',
-        'subsample': 0.8,
-        'random_state': 42,
-        'verbose': 0}
-    >>> booster = vaex.ml.catboost.CatBoostModel(features=features, target='class_', num_boost_round=100, params=params)
-    >>> booster.fit(df_train)
-    >>> df_train = booster.transform(df_train)
-    >>> df_train.head(3)
-    #    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction
-    0             5.4            3               4.5            1.5         1  [0.00615039 0.98024259 0.01360702]
-    1             4.8            3.4             1.6            0.2         0  [0.99034267 0.00526382 0.0043935 ]
-    2             6.9            3.1             4.9            1.5         1  [0.00688241 0.95190908 0.04120851]
-    >>> df_test = booster.transform(df_test)
-    >>> df_test.head(3)
-    #    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction
-    0             5.9            3               4.2            1.5         1  [0.00464228 0.98883351 0.00652421]
-    1             6.1            3               4.6            1.4         1  [0.00350424 0.9882139  0.00828186]
-    2             6.6            2.9             4.6            1.3         1  [0.00325705 0.98891631 0.00782664]
-    '''
+This class provides an interface to the CatBoost algorithm.
+CatBoost is a fast, scalable, high performance Gradient Boosting on
+Decision Trees library, used for ranking, classification, regression and
+other machine learning tasks. For more information please visit
+https://github.com/catboost/catboost
+
+Example:
+
+>>> import vaex
+>>> import vaex.ml.catboost
+>>> df = vaex.datasets.iris()
+>>> features = ['sepal_width', 'petal_length', 'sepal_length', 'petal_width']
+>>> df_train, df_test = df.ml.train_test_split()
+>>> params = {
+    'leaf_estimation_method': 'Gradient',
+    'learning_rate': 0.1,
+    'max_depth': 3,
+    'bootstrap_type': 'Bernoulli',
+    'objective': 'MultiClass',
+    'eval_metric': 'MultiClass',
+    'subsample': 0.8,
+    'random_state': 42,
+    'verbose': 0}
+>>> booster = vaex.ml.catboost.CatBoostModel(features=features, target='class_', num_boost_round=100, params=params)
+>>> booster.fit(df_train)
+>>> df_train = booster.transform(df_train)
+>>> df_train.head(3)
+#    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction
+0             5.4            3               4.5            1.5         1  [0.00615039 0.98024259 0.01360702]
+1             4.8            3.4             1.6            0.2         0  [0.99034267 0.00526382 0.0043935 ]
+2             6.9            3.1             4.9            1.5         1  [0.00688241 0.95190908 0.04120851]
+>>> df_test = booster.transform(df_test)
+>>> df_test.head(3)
+#    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction
+0             5.9            3               4.2            1.5         1  [0.00464228 0.98883351 0.00652421]
+1             6.1            3               4.6            1.4         1  [0.00350424 0.9882139  0.00828186]
+2             6.6            2.9             4.6            1.3         1  [0.00325705 0.98891631 0.00782664]
+'''
     snake_name = "catboost_model"
     features = traitlets.List(traitlets.Unicode(), help='List of features to use when fitting the CatBoostModel.')
     target = traitlets.Unicode(allow_none=False, help='The name of the target column.')

vaex/ml/lightgbm.py

@@ -86,7 +86,7 @@ class LightGBMModel(state.HasState):
         copy.add_virtual_column(self.prediction_name, expression, unique=False)
         return copy
 
-    def fit(self, df, valid_sets=None, valid_names=None, early_stopping_rounds=None, evals_result=None, verbose_eval=None, **kwargs):
+    def fit(self, df, valid_sets=None, valid_names=None, early_stopping_rounds=None, evals_result=None, verbose_eval=False, **kwargs):
         """Fit the LightGBMModel to the DataFrame.
 
         The model will train until the validation score stops improving.
@@ -112,14 +112,19 @@ class LightGBMModel(state.HasState):
         else:
             valid_sets = ()
 
+        callbacks = [
+            lightgbm.callback.record_evaluation(eval_result=evals_result) if evals_result is not None else None,
+            lightgbm.callback.early_stopping(stopping_rounds=early_stopping_rounds) if early_stopping_rounds else None,
+            lightgbm.callback.log_evaluation() if verbose_eval else None
+        ]
+        callbacks = [callback for callback in callbacks if callback is not None]
+
         self.booster = lightgbm.train(params=self.params,
                                       train_set=dtrain,
                                       num_boost_round=self.num_boost_round,
                                       valid_sets=valid_sets,
                                       valid_names=valid_names,
-                                      early_stopping_rounds=early_stopping_rounds,
-                                      evals_result=evals_result,
-                                      verbose_eval=verbose_eval,
+                                      callbacks=callbacks,
                                       **kwargs)
 
     def predict(self, df, **kwargs):

vaex/ml/spec.json

@@ -940,7 +940,7 @@
     },
     {
         "classname": "CatBoostModel",
-        "doc": "The CatBoost algorithm.\n\n    This class provides an interface to the CatBoost aloritham.\n    CatBoost is a fast, scalable, high performance Gradient Boosting on\n    Decision Trees library, used for ranking, classification, regression and\n    other machine learning tasks. For more information please visit\n    https://github.com/catboost/catboost\n\n    Example:\n\n    >>> import vaex\n    >>> import vaex.ml.catboost\n    >>> df = vaex.datasets.iris()\n    >>> features = ['sepal_width', 'petal_length', 'sepal_length', 'petal_width']\n    >>> df_train, df_test = df.ml.train_test_split()\n    >>> params = {\n        'leaf_estimation_method': 'Gradient',\n        'learning_rate': 0.1,\n        'max_depth': 3,\n        'bootstrap_type': 'Bernoulli',\n        'objective': 'MultiClass',\n        'eval_metric': 'MultiClass',\n        'subsample': 0.8,\n        'random_state': 42,\n        'verbose': 0}\n    >>> booster = vaex.ml.catboost.CatBoostModel(features=features, target='class_', num_boost_round=100, params=params)\n    >>> booster.fit(df_train)\n    >>> df_train = booster.transform(df_train)\n    >>> df_train.head(3)\n    #    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction\n    0             5.4            3               4.5            1.5         1  [0.00615039 0.98024259 0.01360702]\n    1             4.8            3.4             1.6            0.2         0  [0.99034267 0.00526382 0.0043935 ]\n    2             6.9            3.1             4.9            1.5         1  [0.00688241 0.95190908 0.04120851]\n    >>> df_test = booster.transform(df_test)\n    >>> df_test.head(3)\n    #    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction\n    0             5.9            3               4.2            1.5         1  [0.00464228 0.98883351 0.00652421]\n    1             6.1            3               4.6            1.4         1  [0.00350424 0.9882139  0.00828186]\n    2             6.6            2.9             4.6            1.3         1  [0.00325705 0.98891631 0.00782664]\n    ",
+        "doc": "The CatBoost algorithm.\n\nThis class provides an interface to the CatBoost algorithm.\nCatBoost is a fast, scalable, high performance Gradient Boosting on\nDecision Trees library, used for ranking, classification, regression and\nother machine learning tasks. For more information please visit\nhttps://github.com/catboost/catboost\n\nExample:\n\n>>> import vaex\n>>> import vaex.ml.catboost\n>>> df = vaex.datasets.iris()\n>>> features = ['sepal_width', 'petal_length', 'sepal_length', 'petal_width']\n>>> df_train, df_test = df.ml.train_test_split()\n>>> params = {\n    'leaf_estimation_method': 'Gradient',\n    'learning_rate': 0.1,\n    'max_depth': 3,\n    'bootstrap_type': 'Bernoulli',\n    'objective': 'MultiClass',\n    'eval_metric': 'MultiClass',\n    'subsample': 0.8,\n    'random_state': 42,\n    'verbose': 0}\n>>> booster = vaex.ml.catboost.CatBoostModel(features=features, target='class_', num_boost_round=100, params=params)\n>>> booster.fit(df_train)\n>>> df_train = booster.transform(df_train)\n>>> df_train.head(3)\n#    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction\n0             5.4            3               4.5            1.5         1  [0.00615039 0.98024259 0.01360702]\n1             4.8            3.4             1.6            0.2         0  [0.99034267 0.00526382 0.0043935 ]\n2             6.9            3.1             4.9            1.5         1  [0.00688241 0.95190908 0.04120851]\n>>> df_test = booster.transform(df_test)\n>>> df_test.head(3)\n#    sepal_length    sepal_width    petal_length    petal_width    class_  catboost_prediction\n0             5.9            3               4.2            1.5         1  [0.00464228 0.98883351 0.00652421]\n1             6.1            3               4.6            1.4         1  [0.00350424 0.9882139  0.00828186]\n2             6.6            2.9             4.6            1.3         1  [0.00325705 0.98891631 0.00782664]\n",
         "module": "vaex.ml.catboost",
         "snake_name": "catboost_model",
         "traits": [

vaex/ml/tensorflow.py

@@ -1,6 +1,6 @@
 import base64
 import tempfile
-import shutil
+import os.path
 
 import numpy as np
 
@@ -95,7 +95,10 @@ class DataFrameAccessorTensorflow():
                         chunk_shape = len(chunk[0].shape) + 1
                         transpose_order = np.arange(1, chunk_shape).tolist() + [0]
                         X = np.array(chunk[:-n_target_cols]).transpose(transpose_order)
-                        y = np.array(chunk[-n_target_cols:], copy=False).T
+                        if np.lib.NumpyVersion(np.__version__) >= '2.0.0':
+                            y = np.array(chunk[-n_target_cols:]).T
+                        else:
+                            y = np.array(chunk[-n_target_cols:], copy=False).T
                         yield (X, y)
 
                 else:
@@ -177,13 +180,12 @@ class KerasModel(vaex.ml.state.HasState):
 
     def state_get(self):
         state = super(KerasModel, self).state_get()
-
         with tempfile.TemporaryDirectory() as directory:
-            self.model.save(directory)
-            zip_path = tempfile.mktemp(".zip")
-            shutil.make_archive(zip_path[:-4], 'zip', directory)
-            with open(zip_path, 'rb') as f:
+            filepath = os.path.join(directory, 'model.keras')
+            self.model.save(filepath)
+            with open(filepath, 'rb') as f:
                 data = f.read()
+            # store model as raw zip bytes base64 encoded
             state['model'] = base64.encodebytes(data).decode('ascii')
             return state
 
@@ -194,8 +196,7 @@ class KerasModel(vaex.ml.state.HasState):
 
         data = base64.decodebytes(model_data.encode('ascii'))
         with tempfile.TemporaryDirectory() as directory:
-            zip_path = tempfile.mktemp('.zip')
-            with open(zip_path, 'wb') as f:
+            filepath = os.path.join(directory, 'model.keras')
+            with open(filepath, 'wb') as f:
                 f.write(data)
-            shutil.unpack_archive(zip_path, directory)
-            self.model = K.models.load_model(directory, custom_objects=self.custom_objects)
+            self.model = K.models.load_model(filepath, custom_objects=self.custom_objects)

vaex/ml/transformations.py

@@ -423,7 +423,7 @@ class OneHotEncoder(Transformer):
         :rtype: DataFrame
         '''
         copy = df.copy()
-        downcast_uint8 = np.can_cast(self.one, np.uint8) and np.can_cast(self.zero, np.uint8)
+        downcast_uint8 = np.can_cast(np.min_scalar_type(self.one), np.uint8) and np.can_cast(np.min_scalar_type(self.zero), np.uint8)
         dtype = 'uint8' if downcast_uint8 else None
         # for each feature, add a virtual column for each unique entry
         for i, feature in enumerate(self.features):
@@ -432,7 +432,7 @@ class OneHotEncoder(Transformer):
                 column_name = self.prefix + feature + '_' + str_value
                 if value is None:
                     copy[column_name] = copy.func.where(copy[feature].ismissing(), self.one, self.zero, dtype=dtype)
-                elif isinstance(value, np.float) and np.isnan(value):
+                elif isinstance(value, float) and np.isnan(value):
                     copy[column_name] = copy.func.where(copy[feature].isnan(), self.one, self.zero, dtype=dtype)
                 else:
                     copy[column_name] = copy.func.where(copy[feature] == value, self.one, self.zero, dtype=dtype)

vaex_ml-0.19.0.dist-info/METADATA

@@ -0,0 +1,34 @@
+Metadata-Version: 2.4
+Name: vaex-ml
+Version: 0.19.0
+Summary: Machine learning support for vaex
+Home-page: https://www.github.com/vaexio/vaex
+Author: Jovan Veljanoski
+Author-email: jovan.veljanoski@gmail.com
+License: MIT
+Description-Content-Type: text/markdown
+License-File: LICENSE.txt
+Requires-Dist: vaex-core~=4.8
+Requires-Dist: numba
+Requires-Dist: traitlets
+Requires-Dist: jinja2
+Requires-Dist: annoy
+Requires-Dist: scikit-learn
+Requires-Dist: xgboost
+Requires-Dist: lightgbm~=4.0
+Requires-Dist: catboost>=1.2.8
+Provides-Extra: all
+Requires-Dist: tensorflow~=2.18; extra == "all"
+Requires-Dist: tensorflow<2.20.0; sys_platform == "darwin" and extra == "all"
+Dynamic: author
+Dynamic: author-email
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: license
+Dynamic: license-file
+Dynamic: provides-extra
+Dynamic: requires-dist
+Dynamic: summary
+
+Machine learning support for vaex

{vaex_ml-0.18.3.dist-info → vaex_ml-0.19.0.dist-info}/RECORD

@@ -1,27 +1,27 @@
 vaex/ml/__init__.py,sha256=cg8TmMc1KEJnG9YknEN8VpJ4ikEMcpqnrvxnwMR9NT4,4263
-vaex/ml/_version.py,sha256=9RC3H0aZm8vW4qzbQ2Z2CVmQm3ClG5wATjqUqK1MMJ0,54
-vaex/ml/catboost.py,sha256=x81_1YXov_JfvgG2d-EArlfxPQLx8sAfl_b8I0TT1LI,10312
+vaex/ml/_version.py,sha256=xB41s6qdt06iHS6CO_taADXuaLN_zxOmjAJox-9eQAk,54
+vaex/ml/catboost.py,sha256=gFdDfzCKKUqXTf383PNhINOtI3HdWU_0S33tbSWRZa8,10167
 vaex/ml/cluster.py,sha256=SXnEB1tZZFstC6lqqPI-Y9UjUTaAUYEKJCMsi59Jp-M,10295
 vaex/ml/generate.py,sha256=ZH3ekTUmmq74adiJLT29aA8rsICQyGw4pCddkGnF2nQ,2762
-vaex/ml/lightgbm.py,sha256=4Vdj3A5gsq7BOrRvrQqHwac0HeNkg6dR9qlAHxWyjlQ,7641
+vaex/ml/lightgbm.py,sha256=vcANvNpK0TYNPOFXbMsCg04Uhl4_AWO0rZQZvzCAKQc,7908
 vaex/ml/linear_model.py,sha256=tx55oPrSfbvH2VwOJ_j5pUPyzsDTOgBlEjfVBZOaf2E,4055
 vaex/ml/metrics.py,sha256=gFpxpE9eNxuF-09Q9X__Tz5VZ_v51Rr0Zod9VtCDxTg,20416
 vaex/ml/pipeline.py,sha256=61w3SINePdyZHbyk1q0T1lNQ3W4wMV56sjCratBEAwU,1001
 vaex/ml/sklearn.py,sha256=AKhkzSs1XLi6VPPCQ77xkpOX8EobgQzBgiPUrbuhddo,11671
-vaex/ml/spec.json,sha256=gPajtlUh4pVk1Zcnq-J9wRywZq0H1P3pJYdQd5O177M,71014
+vaex/ml/spec.json,sha256=1-kdehd6auQHVeBoKG2TCne14MPoIitvKrfgFpBh0NE,70870
 vaex/ml/spec.py,sha256=24jkgxIGzzNhsxmir1bIltZQ8ASw-lBReeXF1TyYh_o,1350
 vaex/ml/state.py,sha256=0g59xmYuJeRwBLWa0YiC9QqAdx2aaL6EG9oZoQ1E7HU,1570
-vaex/ml/tensorflow.py,sha256=tOQARvEWOy6cLoWy27HvtTMUzhhqmDjLrWSP-xr4B7U,9004
-vaex/ml/transformations.py,sha256=2dkWOcfSIeG_pOhVNomKhwgX6brevxYEHvgA_tgv_1c,51292
+vaex/ml/tensorflow.py,sha256=4xzBgxwUUSgK59OEoezu7DhvLAi9kkt_BjwwmGU9C9A,9146
+vaex/ml/transformations.py,sha256=xJ_ZM9SArDCXnv1NX2Y_RRPfzUHsE2NPvQ_aDCjCxGA,51329
 vaex/ml/ui.py,sha256=Zg9PWISoUTwRBIaM-4CuezVzIdpSegxNik0DYQJZVEQ,1902
 vaex/ml/xgboost.py,sha256=XxgzKBRiUTDbkaCxe4Ck5seLG1rlk7u7lLIa_uvjojU,7031
 vaex/ml/incubator/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vaex/ml/incubator/annoy.py,sha256=ZhxT7_tqeZztOmmv29aphRsITbka2ZTgq5bY0YufnhI,3336
 vaex/ml/incubator/pygbm.py,sha256=QQsbYXZ6HaYl5K1i80S3lHtdVMbo2pWKdwMEYg1EqPg,5033
 vaex/ml/incubator/river.py,sha256=1L6_vfRZY1FubyxXbeF_YKtEbq3heafpG2E3SbL8AAY,6106
-vaex_ml-0.18.3.dist-info/LICENSE.txt,sha256=Ub-0nSwxfezlH092uDbwZqynm9i5wq353cMYNPy5OqY,1087
-vaex_ml-0.18.3.dist-info/METADATA,sha256=RtbGD5eY6dC4xFjQdS5LTXoR4JlktMHyJWTcNIdL8yg,485
-vaex_ml-0.18.3.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
-vaex_ml-0.18.3.dist-info/entry_points.txt,sha256=kWLzYdLs3kYqmlZLcAg2M-GVMEftg1VzvTWx92u7dtk,177
-vaex_ml-0.18.3.dist-info/top_level.txt,sha256=BTcHbKEpfXqVlITrjFEtm95UDejFyYXslGl9YUqTIpc,5
-vaex_ml-0.18.3.dist-info/RECORD,,
+vaex_ml-0.19.0.dist-info/licenses/LICENSE.txt,sha256=Ub-0nSwxfezlH092uDbwZqynm9i5wq353cMYNPy5OqY,1087
+vaex_ml-0.19.0.dist-info/METADATA,sha256=ry6S6Y3nMmKeX37dJQTiwgSTEuz9t_2WoUU3ckc-cfo,912
+vaex_ml-0.19.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+vaex_ml-0.19.0.dist-info/entry_points.txt,sha256=cJ8A2uhjPndjLHCMesv0AcKsKK8oCbZclKR3c8yn4Lo,176
+vaex_ml-0.19.0.dist-info/top_level.txt,sha256=BTcHbKEpfXqVlITrjFEtm95UDejFyYXslGl9YUqTIpc,5
+vaex_ml-0.19.0.dist-info/RECORD,,

{vaex_ml-0.18.3.dist-info → vaex_ml-0.19.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.40.0)
+Generator: setuptools (80.10.2)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

{vaex_ml-0.18.3.dist-info → vaex_ml-0.19.0.dist-info}/entry_points.txt

@@ -2,4 +2,3 @@
 ml = vaex.ml:DataFrameAccessorML
 ml.metrics = vaex.ml.metrics:DataFrameAccessorMetrics
 ml.tensorflow = vaex.ml.tensorflow:DataFrameAccessorTensorflow
-

vaex_ml-0.18.3.dist-info/METADATA

@@ -1,20 +0,0 @@
-Metadata-Version: 2.1
-Name: vaex-ml
-Version: 0.18.3
-Summary: Machine learning support for vaex
-Home-page: https://www.github.com/vaexio/vaex
-Author: Jovan Veljanoski
-Author-email: jovan.veljanoski@gmail.com
-License: MIT
-Platform: UNKNOWN
-Requires-Dist: vaex-core (<5,>=4.8.0)
-Requires-Dist: numba
-Requires-Dist: traitlets
-Requires-Dist: jinja2
-Provides-Extra: all
-Requires-Dist: tensorflow (>=2.1.0) ; extra == 'all'
-Requires-Dist: tensorflow-io (>=0.12.0) ; extra == 'all'
-
-UNKNOWN
-
-