deep-river 0.2.8__tar.gz → 0.3.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (40) hide show
  1. {deep_river-0.2.8 → deep_river-0.3.0}/PKG-INFO +32 -11
  2. {deep_river-0.2.8 → deep_river-0.3.0}/README.md +27 -6
  3. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/__version__.py +1 -1
  4. deep_river-0.3.0/deep_river/anomaly/__init__.py +23 -0
  5. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/anomaly/ae.py +181 -23
  6. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/anomaly/probability_weighted_ae.py +96 -6
  7. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/anomaly/rolling_ae.py +115 -5
  8. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/anomaly/scaler.py +1 -2
  9. deep_river-0.3.0/deep_river/base.py +700 -0
  10. deep_river-0.3.0/deep_river/classification/__init__.py +23 -0
  11. deep_river-0.3.0/deep_river/classification/classifier.py +598 -0
  12. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/classification/rolling_classifier.py +200 -12
  13. deep_river-0.3.0/deep_river/classification/zoo.py +277 -0
  14. deep_river-0.3.0/deep_river/regression/__init__.py +21 -0
  15. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/regression/multioutput.py +117 -8
  16. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/regression/regressor.py +118 -6
  17. deep_river-0.3.0/deep_river/regression/rolling_regressor.py +436 -0
  18. deep_river-0.3.0/deep_river/regression/zoo.py +234 -0
  19. deep_river-0.3.0/deep_river/utils/estimator_checks.py +70 -0
  20. deep_river-0.3.0/deep_river/utils/inspect.py +7 -0
  21. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/tensor_conversion.py +47 -19
  22. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/test_estimators.py +7 -3
  23. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/test_tensor_conversion.py +14 -14
  24. {deep_river-0.2.8 → deep_river-0.3.0}/pyproject.toml +4 -4
  25. deep_river-0.2.8/deep_river/anomaly/__init__.py +0 -17
  26. deep_river-0.2.8/deep_river/base.py +0 -312
  27. deep_river-0.2.8/deep_river/classification/__init__.py +0 -13
  28. deep_river-0.2.8/deep_river/classification/classifier.py +0 -344
  29. deep_river-0.2.8/deep_river/classification/zoo.py +0 -264
  30. deep_river-0.2.8/deep_river/regression/__init__.py +0 -17
  31. deep_river-0.2.8/deep_river/regression/rolling_regressor.py +0 -222
  32. deep_river-0.2.8/deep_river/regression/zoo.py +0 -167
  33. deep_river-0.2.8/deep_river/utils/estimator_checks.py +0 -21
  34. {deep_river-0.2.8 → deep_river-0.3.0}/LICENSE +0 -0
  35. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/__init__.py +0 -0
  36. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/__init__.py +0 -0
  37. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/hooks.py +0 -0
  38. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/layer_adaptation.py +0 -0
  39. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/params.py +0 -0
  40. {deep_river-0.2.8 → deep_river-0.3.0}/deep_river/utils/test_layer_adaptation.py +0 -0
@@ -1,8 +1,7 @@
1
- Metadata-Version: 2.1
1
+ Metadata-Version: 2.3
2
2
  Name: deep_river
3
- Version: 0.2.8
3
+ Version: 0.3.0
4
4
  Summary: Online Deep Learning for river
5
- Home-page: https://github.com/online-ml/deep-river
6
5
  License: BSD-3-Clause
7
6
  Keywords: online-learning,deep-learning,data-streams
8
7
  Author: Cedric Kulbach
@@ -18,15 +17,16 @@ Classifier: Programming Language :: Python :: Implementation :: CPython
18
17
  Classifier: Programming Language :: Python :: Implementation :: PyPy
19
18
  Requires-Dist: mkdocstrings[python] (>=0.19.0)
20
19
  Requires-Dist: numpy (>=1.26.4,<1.27.0)
21
- Requires-Dist: ordered-set (>=4.1.0,<4.2.0)
22
20
  Requires-Dist: pandas (>=2.2.2,<2.3.0)
23
21
  Requires-Dist: pytkdocs[numpy-style] (>=0.5.0)
24
22
  Requires-Dist: river
25
23
  Requires-Dist: scikit-learn (>=1.5.0,<1.6.0)
26
- Requires-Dist: torch (>=2.2.2,<2.3.0)
24
+ Requires-Dist: sortedcontainers (>=2.4.0,<3.0.0)
25
+ Requires-Dist: torch (==2.2.2)
27
26
  Requires-Dist: torchviz (>=0.0.2,<0.1.0)
28
27
  Requires-Dist: tqdm (>=4.66.5,<4.67.0)
29
28
  Project-URL: Documentation, https://github.com/online-ml/deep-river
29
+ Project-URL: Homepage, https://github.com/online-ml/deep-river
30
30
  Project-URL: Repository, https://github.com/online-ml/deep-river
31
31
  Description-Content-Type: text/markdown
32
32
 
@@ -42,6 +42,7 @@ Description-Content-Type: text/markdown
42
42
  <img alt="PyPI - Downloads" src="https://img.shields.io/pypi/dm/deep-river">
43
43
  <img alt="GitHub" src="https://img.shields.io/github/license/online-ml/deep-river">
44
44
  <a href="https://joss.theoj.org/papers/6a76784f55e8b041d71a7fa776eb386a"><img src="https://joss.theoj.org/papers/6a76784f55e8b041d71a7fa776eb386a/status.svg"></a>
45
+ <a href="https://doi.org/10.5281/zenodo.14601980"><img src="https://zenodo.org/badge/DOI/10.5281/zenodo.14601980.svg" alt="DOI"></a>
45
46
  </p>
46
47
  <p align="center">
47
48
  deep-river is a Python library for online deep learning.
@@ -102,7 +103,7 @@ For further examples check out the <a href="https://online-ml.github.io/deep-riv
102
103
 
103
104
  >>> model_pipeline = compose.Pipeline(
104
105
  ... preprocessing.StandardScaler(),
105
- ... classification.Classifier(module=MyModule, loss_fn='binary_cross_entropy', optimizer_fn='adam')
106
+ ... classification.ClassifierInitialized(module=MyModule(10), loss_fn='binary_cross_entropy', optimizer_fn='adam')
106
107
  ... )
107
108
 
108
109
  >>> dataset = datasets.Phishing()
@@ -124,7 +125,7 @@ Accuracy: 0.7264
124
125
  >>> from river import stream
125
126
  >>> from sklearn import datasets
126
127
  >>> from torch import nn
127
- >>> from deep_river.regression.multioutput import MultiTargetRegressor
128
+ >>> from deep_river.regression.multioutput import MultiTargetRegressorInitialized
128
129
 
129
130
  >>> class MyModule(nn.Module):
130
131
  ... def __init__(self, n_features):
@@ -142,8 +143,8 @@ Accuracy: 0.7264
142
143
  ... )
143
144
  >>> model = compose.Pipeline(
144
145
  ... preprocessing.StandardScaler(),
145
- ... MultiTargetRegressor(
146
- ... module=MyModule,
146
+ ... MultiTargetRegressorInitialized(
147
+ ... module=MyModule(10),
147
148
  ... loss_fn='mse',
148
149
  ... lr=0.3,
149
150
  ... optimizer_fn='sgd',
@@ -158,7 +159,7 @@ MicroAverage(MAE): 34.31
158
159
  ### Anomaly Detection
159
160
 
160
161
  ```python
161
- >>> from deep_river.anomaly import Autoencoder
162
+ >>> from deep_river.anomaly import AutoencoderInitialized
162
163
  >>> from river import metrics
163
164
  >>> from river.datasets import CreditCard
164
165
  >>> from torch import nn
@@ -183,7 +184,7 @@ MicroAverage(MAE): 34.31
183
184
  ... X = self.linear2(X)
184
185
  ... return self.sigmoid(X)
185
186
 
186
- >>> ae = Autoencoder(module=MyAutoEncoder, lr=0.005)
187
+ >>> ae = AutoencoderInitialized(module=MyAutoEncoder(10), lr=0.005)
187
188
  >>> scaler = MinMaxScaler()
188
189
  >>> model = Pipeline(scaler, ae)
189
190
 
@@ -197,6 +198,26 @@ Rolling ROCAUC: 0.8901
197
198
 
198
199
  ```
199
200
 
201
+ ## 💬 Citation
202
+
203
+ To acknowledge the use of the `DeepRiver` library in your research, please refer to our [paper](https://joss.theoj.org/papers/10.21105/joss.07226) published on Journal of Open Source Software (JOSS):
204
+
205
+ ```bibtex
206
+ @article{Kulbach2025,
207
+ doi = {10.21105/joss.07226},
208
+ url = {https://doi.org/10.21105/joss.07226},
209
+ year = {2025},
210
+ publisher = {The Open Journal},
211
+ volume = {10},
212
+ number = {105},
213
+ pages = {7226},
214
+ author = {Cedric Kulbach and Lucas Cazzonelli and Hoang-Anh Ngo and Max Halford and Saulo Martiello Mastelini},
215
+ title = {DeepRiver: A Deep Learning Library for Data Streams},
216
+ journal = {Journal of Open Source Software}
217
+ }
218
+ ```
219
+
220
+
200
221
  ## 🏫 Affiliations
201
222
 
202
223
  <p align="center">
@@ -10,6 +10,7 @@
10
10
  <img alt="PyPI - Downloads" src="https://img.shields.io/pypi/dm/deep-river">
11
11
  <img alt="GitHub" src="https://img.shields.io/github/license/online-ml/deep-river">
12
12
  <a href="https://joss.theoj.org/papers/6a76784f55e8b041d71a7fa776eb386a"><img src="https://joss.theoj.org/papers/6a76784f55e8b041d71a7fa776eb386a/status.svg"></a>
13
+ <a href="https://doi.org/10.5281/zenodo.14601980"><img src="https://zenodo.org/badge/DOI/10.5281/zenodo.14601980.svg" alt="DOI"></a>
13
14
  </p>
14
15
  <p align="center">
15
16
  deep-river is a Python library for online deep learning.
@@ -70,7 +71,7 @@ For further examples check out the <a href="https://online-ml.github.io/deep-riv
70
71
 
71
72
  >>> model_pipeline = compose.Pipeline(
72
73
  ... preprocessing.StandardScaler(),
73
- ... classification.Classifier(module=MyModule, loss_fn='binary_cross_entropy', optimizer_fn='adam')
74
+ ... classification.ClassifierInitialized(module=MyModule(10), loss_fn='binary_cross_entropy', optimizer_fn='adam')
74
75
  ... )
75
76
 
76
77
  >>> dataset = datasets.Phishing()
@@ -92,7 +93,7 @@ Accuracy: 0.7264
92
93
  >>> from river import stream
93
94
  >>> from sklearn import datasets
94
95
  >>> from torch import nn
95
- >>> from deep_river.regression.multioutput import MultiTargetRegressor
96
+ >>> from deep_river.regression.multioutput import MultiTargetRegressorInitialized
96
97
 
97
98
  >>> class MyModule(nn.Module):
98
99
  ... def __init__(self, n_features):
@@ -110,8 +111,8 @@ Accuracy: 0.7264
110
111
  ... )
111
112
  >>> model = compose.Pipeline(
112
113
  ... preprocessing.StandardScaler(),
113
- ... MultiTargetRegressor(
114
- ... module=MyModule,
114
+ ... MultiTargetRegressorInitialized(
115
+ ... module=MyModule(10),
115
116
  ... loss_fn='mse',
116
117
  ... lr=0.3,
117
118
  ... optimizer_fn='sgd',
@@ -126,7 +127,7 @@ MicroAverage(MAE): 34.31
126
127
  ### Anomaly Detection
127
128
 
128
129
  ```python
129
- >>> from deep_river.anomaly import Autoencoder
130
+ >>> from deep_river.anomaly import AutoencoderInitialized
130
131
  >>> from river import metrics
131
132
  >>> from river.datasets import CreditCard
132
133
  >>> from torch import nn
@@ -151,7 +152,7 @@ MicroAverage(MAE): 34.31
151
152
  ... X = self.linear2(X)
152
153
  ... return self.sigmoid(X)
153
154
 
154
- >>> ae = Autoencoder(module=MyAutoEncoder, lr=0.005)
155
+ >>> ae = AutoencoderInitialized(module=MyAutoEncoder(10), lr=0.005)
155
156
  >>> scaler = MinMaxScaler()
156
157
  >>> model = Pipeline(scaler, ae)
157
158
 
@@ -165,6 +166,26 @@ Rolling ROCAUC: 0.8901
165
166
 
166
167
  ```
167
168
 
169
+ ## 💬 Citation
170
+
171
+ To acknowledge the use of the `DeepRiver` library in your research, please refer to our [paper](https://joss.theoj.org/papers/10.21105/joss.07226) published on Journal of Open Source Software (JOSS):
172
+
173
+ ```bibtex
174
+ @article{Kulbach2025,
175
+ doi = {10.21105/joss.07226},
176
+ url = {https://doi.org/10.21105/joss.07226},
177
+ year = {2025},
178
+ publisher = {The Open Journal},
179
+ volume = {10},
180
+ number = {105},
181
+ pages = {7226},
182
+ author = {Cedric Kulbach and Lucas Cazzonelli and Hoang-Anh Ngo and Max Halford and Saulo Martiello Mastelini},
183
+ title = {DeepRiver: A Deep Learning Library for Data Streams},
184
+ journal = {Journal of Open Source Software}
185
+ }
186
+ ```
187
+
188
+
168
189
  ## 🏫 Affiliations
169
190
 
170
191
  <p align="center">
@@ -1,3 +1,3 @@
1
- VERSION = (0, 2, 8)
1
+ VERSION = (0, 3, 0)
2
2
 
3
3
  __version__ = ".".join(map(str, VERSION)) # noqa: F401
@@ -0,0 +1,23 @@
1
+ from .ae import Autoencoder, AutoencoderInitialized
2
+ from .probability_weighted_ae import (
3
+ ProbabilityWeightedAutoencoder,
4
+ ProbabilityWeightedAutoencoderInitialized,
5
+ )
6
+ from .rolling_ae import RollingAutoencoder, RollingAutoencoderInitialized
7
+ from .scaler import AnomalyMeanScaler, AnomalyMinMaxScaler, AnomalyStandardScaler
8
+
9
+ """
10
+ This module contains the anomaly detection algorithms for the
11
+ deep_river package.
12
+ """
13
+ __all__ = [
14
+ "Autoencoder",
15
+ "RollingAutoencoder",
16
+ "ProbabilityWeightedAutoencoder",
17
+ "AnomalyStandardScaler",
18
+ "AnomalyMeanScaler",
19
+ "AnomalyMinMaxScaler",
20
+ "AutoencoderInitialized",
21
+ "ProbabilityWeightedAutoencoderInitialized",
22
+ "RollingAutoencoderInitialized",
23
+ ]
@@ -1,3 +1,4 @@
1
+ import warnings
1
2
  from typing import Any, Callable, Dict, Type, Union
2
3
 
3
4
  import numpy as np
@@ -6,7 +7,7 @@ import torch
6
7
  from river.anomaly.base import AnomalyDetector
7
8
  from torch import nn
8
9
 
9
- from deep_river.base import DeepEstimator
10
+ from deep_river.base import DeepEstimator, DeepEstimatorInitialized
10
11
  from deep_river.utils import dict2tensor
11
12
  from deep_river.utils.layer_adaptation import expand_layer
12
13
  from deep_river.utils.tensor_conversion import df2tensor
@@ -15,6 +16,8 @@ from deep_river.utils.tensor_conversion import df2tensor
15
16
  class _TestAutoencoder(torch.nn.Module):
16
17
  def __init__(self, n_features, latent_dim=3):
17
18
  super().__init__()
19
+ self.n_features = n_features
20
+ self.latent_dim = latent_dim
18
21
  self.linear1 = nn.Linear(n_features, latent_dim)
19
22
  self.nonlin = torch.nn.LeakyReLU()
20
23
  self.linear2 = nn.Linear(latent_dim, n_features)
@@ -108,6 +111,12 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
108
111
  seed: int = 42,
109
112
  **kwargs,
110
113
  ):
114
+ warnings.warn(
115
+ "This is deprecated and will be removed in future releases. "
116
+ "Please instead use the AutoencoderInitialized class and "
117
+ "initialize the module beforehand"
118
+ )
119
+
111
120
  super().__init__(
112
121
  module=module,
113
122
  loss_fn=loss_fn,
@@ -147,15 +156,10 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
147
156
  Most estimators pass the full test suite. However, in some cases,
148
157
  some estimators might not
149
158
  be able to pass certain checks.
150
-
151
- Returns
152
- -------
153
- set
154
- Set of checks to skip during unit testing.
155
159
  """
156
160
  return set()
157
161
 
158
- def learn_one(self, x: dict, y: Any = None, **kwargs) -> "Autoencoder":
162
+ def learn_one(self, x: dict, y: Any = None, **kwargs) -> None:
159
163
  """
160
164
  Performs one step of training with a single example.
161
165
 
@@ -165,19 +169,12 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
165
169
  Input example.
166
170
 
167
171
  **kwargs
168
-
169
- Returns
170
- -------
171
- Autoencoder
172
- The model itself.
173
172
  """
174
173
  if not self.module_initialized:
175
174
  self._update_observed_features(x)
176
175
  self.initialize_module(x=x, **self.kwargs)
177
176
  self._adapt_input_dim(x)
178
- return self._learn(
179
- dict2tensor(x, features=self.observed_features, device=self.device)
180
- )
177
+ self._learn(dict2tensor(x, features=self.observed_features, device=self.device))
181
178
 
182
179
  def _learn(self, x: torch.Tensor) -> "Autoencoder":
183
180
  self.module.train()
@@ -219,7 +216,7 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
219
216
  loss = self.loss_func(x_pred, x_t).item()
220
217
  return loss
221
218
 
222
- def learn_many(self, X: pd.DataFrame) -> "Autoencoder":
219
+ def learn_many(self, X: pd.DataFrame) -> None:
223
220
  """
224
221
  Performs one step of training with a batch of examples.
225
222
 
@@ -227,12 +224,6 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
227
224
  ----------
228
225
  X
229
226
  Input batch of examples.
230
-
231
- Returns
232
- -------
233
- Autoencoder
234
- The model itself.
235
-
236
227
  """
237
228
  if not self.module_initialized:
238
229
 
@@ -241,7 +232,7 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
241
232
 
242
233
  self._adapt_input_dim(X)
243
234
  X_t = df2tensor(X, features=self.observed_features, device=self.device)
244
- return self._learn(X_t)
235
+ self._learn(X_t)
245
236
 
246
237
  def score_many(self, X: pd.DataFrame) -> np.ndarray:
247
238
  """
@@ -292,3 +283,170 @@ class Autoencoder(DeepEstimator, AnomalyDetector):
292
283
  len(self.observed_features),
293
284
  output=True,
294
285
  )
286
+
287
+
288
+ class AutoencoderInitialized(DeepEstimatorInitialized, AnomalyDetector):
289
+ """
290
+ Represents an initialized autoencoder for anomaly detection and feature learning.
291
+
292
+ This class is built upon the DeepEstimatorInitialized and AnomalyDetector
293
+ base classes. It provides methods for performing unsupervised learning
294
+ through an autoencoder mechanism. The primary objective of the class is
295
+ to train the autoencoder on input data and compute anomaly scores based
296
+ on the reconstruction error. It supports learning on individual examples
297
+ or entire batches of data.
298
+
299
+ Attributes
300
+ ----------
301
+ is_feature_incremental : bool
302
+ Indicates whether the model is designed to increment features dynamically.
303
+ module : torch.nn.Module
304
+ The PyTorch model representing the autoencoder architecture.
305
+ loss_fn : Union[str, Callable]
306
+ Specifies the loss function to compute the reconstruction error.
307
+ optimizer_fn : Union[str, Callable]
308
+ Specifies the optimizer to be used for training the autoencoder.
309
+ lr : float
310
+ The learning rate for optimization.
311
+ device : str
312
+ The device on which the model is loaded and trained (e.g., "cpu",
313
+ "cuda").
314
+ seed : int
315
+ Random seed for ensuring reproducibility.
316
+ """
317
+
318
+ def __init__(
319
+ self,
320
+ module: torch.nn.Module,
321
+ loss_fn: Union[str, Callable] = "mse",
322
+ optimizer_fn: Union[str, Callable] = "sgd",
323
+ lr: float = 1e-3,
324
+ is_feature_incremental: bool = False,
325
+ device: str = "cpu",
326
+ seed: int = 42,
327
+ **kwargs,
328
+ ):
329
+ super().__init__(
330
+ module=module,
331
+ loss_fn=loss_fn,
332
+ optimizer_fn=optimizer_fn,
333
+ lr=lr,
334
+ is_feature_incremental=is_feature_incremental,
335
+ device=device,
336
+ seed=seed,
337
+ **kwargs,
338
+ )
339
+ self.is_feature_incremental = is_feature_incremental
340
+
341
+ @classmethod
342
+ def _unit_test_params(cls):
343
+ """
344
+ Returns a dictionary of parameters to be used for unit testing
345
+ the respective class.
346
+
347
+ Yields
348
+ -------
349
+ dict
350
+ Dictionary of parameters to be used for unit testing
351
+ the respective class.
352
+ """
353
+
354
+ yield {
355
+ "module": _TestAutoencoder(30, 3),
356
+ "loss_fn": "mse",
357
+ "optimizer_fn": "sgd",
358
+ "is_feature_incremental": False,
359
+ }
360
+
361
+ @classmethod
362
+ def _unit_test_skips(self) -> set:
363
+ """
364
+ Indicates which checks to skip during unit testing.
365
+ Most estimators pass the full test suite. However, in some cases,
366
+ some estimators might not
367
+ be able to pass certain checks.
368
+ """
369
+ return set()
370
+
371
+ def learn_one(self, x: dict, y: Any = None, **kwargs) -> None:
372
+ """
373
+ Performs one step of training with a single example.
374
+
375
+ Parameters
376
+ ----------
377
+ x
378
+ Input example.
379
+
380
+ **kwargs
381
+ """
382
+ self._update_observed_features(x)
383
+ self._learn(self._dict2tensor(x))
384
+
385
+ def score_one(self, x: dict) -> float:
386
+ """
387
+ Returns an anomaly score for the provided example in the form of
388
+ the autoencoder's reconstruction error.
389
+
390
+ Parameters
391
+ ----------
392
+ x
393
+ Input example.
394
+
395
+ Returns
396
+ -------
397
+ float
398
+ Anomaly score for the given example. Larger values indicate
399
+ more anomalous examples.
400
+
401
+ """
402
+
403
+ self._update_observed_features(x)
404
+ x_t = self._dict2tensor(x)
405
+ self.module.eval()
406
+ with torch.inference_mode():
407
+ x_pred = self.module(x_t)
408
+ loss = self.loss_func(x_pred, x_t).item()
409
+ return loss
410
+
411
+ def learn_many(self, X: pd.DataFrame) -> None:
412
+ """
413
+ Performs one step of training with a batch of examples.
414
+
415
+ Parameters
416
+ ----------
417
+ X
418
+ Input batch of examples.
419
+ """
420
+
421
+ self._update_observed_features(X)
422
+ X_t = self._df2tensor(X)
423
+ self._learn(X_t)
424
+
425
+ def score_many(self, X: pd.DataFrame) -> np.ndarray:
426
+ """
427
+ Returns an anomaly score for the provided batch of examples in
428
+ the form of the autoencoder's reconstruction error.
429
+
430
+ Parameters
431
+ ----------
432
+ x
433
+ Input batch of examples.
434
+
435
+ Returns
436
+ -------
437
+ float
438
+ Anomaly scores for the given batch of examples. Larger values
439
+ indicate more anomalous examples.
440
+ """
441
+ self._update_observed_features(X)
442
+ x_t = self._df2tensor(X)
443
+
444
+ self.module.eval()
445
+ with torch.inference_mode():
446
+ x_pred = self.module(x_t)
447
+ loss = torch.mean(
448
+ self.loss_func(x_pred, x_t, reduction="none"),
449
+ dim=list(range(1, x_t.dim())),
450
+ )
451
+ score = loss.cpu().detach().numpy()
452
+ return score
@@ -1,4 +1,5 @@
1
1
  import math
2
+ import warnings
2
3
  from typing import Any, Callable, Type, Union
3
4
 
4
5
  import pandas as pd
@@ -117,9 +118,7 @@ class ProbabilityWeightedAutoencoder(ae.Autoencoder):
117
118
  self.rolling_mean = utils.Rolling(stats.Mean(), window_size=window_size)
118
119
  self.rolling_var = utils.Rolling(stats.Var(), window_size=window_size)
119
120
 
120
- def learn_one(
121
- self, x: dict, y: Any = None, **kwargs
122
- ) -> "ProbabilityWeightedAutoencoder":
121
+ def learn_one(self, x: dict, y: Any = None, **kwargs) -> None:
123
122
  """
124
123
  Performs one step of training with a single example,
125
124
  scaling the employed learning rate based on the outlier
@@ -147,7 +146,6 @@ class ProbabilityWeightedAutoencoder(ae.Autoencoder):
147
146
  x_pred = self.module(x_t)
148
147
  loss = self.loss_func(x_pred, x_t)
149
148
  self._apply_loss(loss)
150
- return self
151
149
 
152
150
  def _apply_loss(self, loss):
153
151
  losses_numpy = loss.detach().numpy()
@@ -169,7 +167,7 @@ class ProbabilityWeightedAutoencoder(ae.Autoencoder):
169
167
  loss.backward()
170
168
  self.optimizer.step()
171
169
 
172
- def learn_many(self, X: pd.DataFrame) -> "ProbabilityWeightedAutoencoder":
170
+ def learn_many(self, X: pd.DataFrame) -> None:
173
171
  if not self.module_initialized:
174
172
  self._update_observed_features(X)
175
173
  self.initialize_module(x=X, **self.kwargs)
@@ -183,4 +181,96 @@ class ProbabilityWeightedAutoencoder(ae.Autoencoder):
183
181
  dim=list(range(1, X_t.dim())),
184
182
  )
185
183
  self._apply_loss(loss)
186
- return self
184
+
185
+
186
+ class ProbabilityWeightedAutoencoderInitialized(ae.AutoencoderInitialized):
187
+ """ """
188
+
189
+ def __init__(
190
+ self,
191
+ module: torch.nn.Module,
192
+ loss_fn: Union[str, Callable] = "mse",
193
+ optimizer_fn: Union[str, Callable] = "sgd",
194
+ lr: float = 1e-3,
195
+ device: str = "cpu",
196
+ seed: int = 42,
197
+ skip_threshold: float = 0.9,
198
+ window_size=250,
199
+ **kwargs,
200
+ ):
201
+ warnings.warn(
202
+ "This is deprecated and will be removed in future releases. "
203
+ "Please instead use the AutoencoderInitialized class and "
204
+ "initialize the module beforehand"
205
+ )
206
+ super().__init__(
207
+ module=module,
208
+ loss_fn=loss_fn,
209
+ optimizer_fn=optimizer_fn,
210
+ lr=lr,
211
+ device=device,
212
+ seed=seed,
213
+ **kwargs,
214
+ )
215
+ self.window_size = window_size
216
+ self.skip_threshold = skip_threshold
217
+ self.rolling_mean = utils.Rolling(stats.Mean(), window_size=window_size)
218
+ self.rolling_var = utils.Rolling(stats.Var(), window_size=window_size)
219
+
220
+ def learn_one(self, x: dict, y: Any = None, **kwargs) -> None:
221
+ """
222
+ Performs one step of training with a single example,
223
+ scaling the employed learning rate based on the outlier
224
+ probability estimate of the input example.
225
+
226
+ Parameters
227
+ ----------
228
+ **kwargs
229
+ x
230
+ Input example.
231
+
232
+ Returns
233
+ -------
234
+ ProbabilityWeightedAutoencoder
235
+ The autoencoder itself.
236
+ """
237
+
238
+ self._update_observed_features(x)
239
+ x_t = self._dict2tensor(x)
240
+
241
+ self.module.train()
242
+ x_pred = self.module(x_t)
243
+ loss = self.loss_func(x_pred, x_t)
244
+ self._apply_loss(loss)
245
+
246
+ def _apply_loss(self, loss):
247
+ losses_numpy = loss.detach().numpy()
248
+ mean = self.rolling_mean.get()
249
+ var = self.rolling_var.get() if self.rolling_var.get() > 0 else 1
250
+ if losses_numpy.ndim == 0:
251
+ self.rolling_mean.update(losses_numpy)
252
+ self.rolling_var.update(losses_numpy)
253
+ else:
254
+ for loss_numpy in range(len(losses_numpy)):
255
+ self.rolling_mean.update(loss_numpy)
256
+ self.rolling_var.update(loss_numpy)
257
+
258
+ loss_scaled = (losses_numpy - mean) / math.sqrt(var)
259
+ prob = ndtr(loss_scaled)
260
+ loss = torch.tensor((self.skip_threshold - prob) / self.skip_threshold) * loss
261
+
262
+ self.optimizer.zero_grad()
263
+ loss.backward()
264
+ self.optimizer.step()
265
+
266
+ def learn_many(self, X: pd.DataFrame) -> None:
267
+ self._update_observed_features(X)
268
+ X_t = self._df2tensor(X)
269
+
270
+ self.module.train()
271
+ x_pred = self.module(X)
272
+ loss = torch.mean(
273
+ self.loss_func(x_pred, X_t, reduction="none"),
274
+ dim=list(range(1, X_t.dim())),
275
+ )
276
+ self._apply_loss(loss)