ddi-fw 0.0.153__py3-none-any.whl → 0.0.154__py3-none-any.whl
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.
- ddi_fw/datasets/dataset_splitter.py +3 -3
- ddi_fw/pipeline/pipeline.py +5 -0
- {ddi_fw-0.0.153.dist-info → ddi_fw-0.0.154.dist-info}/METADATA +1 -1
- {ddi_fw-0.0.153.dist-info → ddi_fw-0.0.154.dist-info}/RECORD +6 -6
- {ddi_fw-0.0.153.dist-info → ddi_fw-0.0.154.dist-info}/WHEEL +0 -0
- {ddi_fw-0.0.153.dist-info → ddi_fw-0.0.154.dist-info}/top_level.txt +0 -0
@@ -14,11 +14,11 @@ class DatasetSplitter(BaseModel):
|
|
14
14
|
class Config:
|
15
15
|
arbitrary_types_allowed = True
|
16
16
|
|
17
|
-
def split(self, X: pd.DataFrame, y: pd.Series)-> Tuple[
|
18
|
-
|
17
|
+
def split(self, X: pd.DataFrame, y: pd.Series) -> Tuple[
|
18
|
+
pd.DataFrame, pd.DataFrame, pd.Series, pd.Series, pd.Index, pd.Index, List[np.ndarray], List[np.ndarray]]:
|
19
19
|
print(
|
20
20
|
f"Splitting dataset into {self.fold_size} folds with shuffle={self.shuffle}...")
|
21
|
-
#TODO check it
|
21
|
+
# TODO check it
|
22
22
|
if len(y.shape) == 1:
|
23
23
|
y = pd.Series(np.expand_dims(y.to_numpy(), axis=1).flatten())
|
24
24
|
stacked = np.vstack(tuple(y.to_numpy()))
|
ddi_fw/pipeline/pipeline.py
CHANGED
@@ -1,4 +1,5 @@
|
|
1
1
|
from typing import Any, Dict, List, Optional, Type, Union
|
2
|
+
from ddi_fw.datasets.dataset_splitter import DatasetSplitter
|
2
3
|
import numpy as np
|
3
4
|
import pandas as pd
|
4
5
|
import chromadb
|
@@ -23,6 +24,7 @@ class Pipeline(BaseModel):
|
|
23
24
|
artifact_location: Optional[str] = None
|
24
25
|
tracking_uri: Optional[str] = None
|
25
26
|
dataset_type: Type[BaseDataset]
|
27
|
+
dataset_splitter_type: Type[DatasetSplitter] = DatasetSplitter
|
26
28
|
columns: Optional[List[str]] = None
|
27
29
|
embedding_dict: Optional[Dict[str, Any]] = None
|
28
30
|
column_embedding_configs: Optional[Dict] = None
|
@@ -181,6 +183,8 @@ class Pipeline(BaseModel):
|
|
181
183
|
# self.ner_df = CTakesNER(df=None).load(
|
182
184
|
# filename=self.ner_data_file) if self.ner_data_file else None
|
183
185
|
|
186
|
+
dataset_splitter = self.dataset_splitter_type()
|
187
|
+
|
184
188
|
if issubclass(self.dataset_type, TextDatasetMixin):
|
185
189
|
key, value = next(iter(embedding_dict.items()))
|
186
190
|
embedding_size = value[next(iter(value))][0].shape[0]
|
@@ -190,6 +194,7 @@ class Pipeline(BaseModel):
|
|
190
194
|
embedding_dict=embedding_dict,
|
191
195
|
embedding_size=embedding_size,
|
192
196
|
embeddings_pooling_strategy=pooling_strategy,
|
197
|
+
dataset_splitter = dataset_splitter,
|
193
198
|
**kwargs)
|
194
199
|
else:
|
195
200
|
dataset = self.dataset_type(**kwargs)
|
@@ -1,6 +1,6 @@
|
|
1
1
|
ddi_fw/datasets/__init__.py,sha256=yDsRQD_9Ijpm_Rl2wSDwdutG5Q_wca_UBPEvm7nBx04,444
|
2
2
|
ddi_fw/datasets/core.py,sha256=vRMpUsIHIbOKi-6TuUPNu1Ve3ny3cS9hdsydQxFCNvE,8078
|
3
|
-
ddi_fw/datasets/dataset_splitter.py,sha256=
|
3
|
+
ddi_fw/datasets/dataset_splitter.py,sha256=8H8uZTAf8N9LUZeSeHOMawtJFJhnDgUUqFcnl7dquBQ,1672
|
4
4
|
ddi_fw/datasets/db_utils.py,sha256=OTsa3d-Iic7z3HmzSQK9UigedRbHDxYChJk0s4GfLnw,6191
|
5
5
|
ddi_fw/datasets/embedding_generator.py,sha256=jiDKwLaPMaQkloxQkuCrhl-A-2OdvocmkSzjWtUnk4g,2255
|
6
6
|
ddi_fw/datasets/feature_vector_generation.py,sha256=gvjpEzkgVV8dp4V8NMMv59u0v-1tNAJ7v83R-keWGoA,4748
|
@@ -88,7 +88,7 @@ ddi_fw/pipeline/__init__.py,sha256=tKDM_rW4vPjlYTeOkNgi9PujDzb4e9O3LK1w5wqnebw,2
|
|
88
88
|
ddi_fw/pipeline/multi_modal_combination_strategy.py,sha256=JSyuP71b1I1yuk0s2ecCJZTtCED85jBtkpwTUxibJvI,1706
|
89
89
|
ddi_fw/pipeline/multi_pipeline.py,sha256=D_BZ3ciHbVGuuB7m7cEmVQHESruh1gqhA-vxCMfNKj0,5407
|
90
90
|
ddi_fw/pipeline/ner_pipeline.py,sha256=q1aKjb54Ra1HzZ7dARvBw6lB37je9R-POEf2h6QT_nU,6018
|
91
|
-
ddi_fw/pipeline/pipeline.py,sha256=
|
91
|
+
ddi_fw/pipeline/pipeline.py,sha256=70lYsluAnTWDLTlf6rbecffw3Bl34L1_6ALfLUoSvtY,11324
|
92
92
|
ddi_fw/utils/__init__.py,sha256=77563ikqAtdzjjgRlLp5OAsJBbpLA1Cao8iecGaVUXQ,354
|
93
93
|
ddi_fw/utils/enums.py,sha256=19eJ3fX5eRK_xPvkYcukmug144jXPH4X9zQqtsFBj5A,671
|
94
94
|
ddi_fw/utils/json_helper.py,sha256=BVU6wmJgdXPxyqLPu3Ck_9Es5RrP1PDanKvE-OSj1D4,571
|
@@ -97,7 +97,7 @@ ddi_fw/utils/package_helper.py,sha256=erl8_onmhK-41zQoaED2qyDUV9GQxmT9sdoyRp9_q5
|
|
97
97
|
ddi_fw/utils/py7zr_helper.py,sha256=gOqaFIyJvTjUM-btO2x9AQ69jZOS8PoKN0wetYIckJw,4747
|
98
98
|
ddi_fw/utils/utils.py,sha256=szwnxMTDRrZoeNRyDuf3aCbtzriwtaRk4mHSH3asLdA,4301
|
99
99
|
ddi_fw/utils/zip_helper.py,sha256=YRZA4tKZVBJwGQM0_WK6L-y5MoqkKoC-nXuuHK6CU9I,5567
|
100
|
-
ddi_fw-0.0.
|
101
|
-
ddi_fw-0.0.
|
102
|
-
ddi_fw-0.0.
|
103
|
-
ddi_fw-0.0.
|
100
|
+
ddi_fw-0.0.154.dist-info/METADATA,sha256=u9w1h_7YiMy64J3Ul7uEkz_YQAePYycw9AL12gU7FNI,2082
|
101
|
+
ddi_fw-0.0.154.dist-info/WHEEL,sha256=DK49LOLCYiurdXXOXwGJm6U4DkHkg4lcxjhqwRa0CP4,91
|
102
|
+
ddi_fw-0.0.154.dist-info/top_level.txt,sha256=PMwHICFZTZtcpzQNPV4UQnfNXYIeLR_Ste-Wfc1h810,7
|
103
|
+
ddi_fw-0.0.154.dist-info/RECORD,,
|
File without changes
|
File without changes
|