ddi-fw 0.0.227__py3-none-any.whl → 0.0.228__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/pipeline/ner_pipeline.py +5 -0
- {ddi_fw-0.0.227.dist-info → ddi_fw-0.0.228.dist-info}/METADATA +1 -1
- {ddi_fw-0.0.227.dist-info → ddi_fw-0.0.228.dist-info}/RECORD +5 -5
- {ddi_fw-0.0.227.dist-info → ddi_fw-0.0.228.dist-info}/WHEEL +0 -0
- {ddi_fw-0.0.227.dist-info → ddi_fw-0.0.228.dist-info}/top_level.txt +0 -0
ddi_fw/pipeline/ner_pipeline.py
CHANGED
|
@@ -81,6 +81,9 @@ class NerParameterSearch(BaseModel):
|
|
|
81
81
|
df = idf.to_dataframe()
|
|
82
82
|
self.min_threshold_dict = {key: np.floor(df.describe()[key]["min"]) for key in df.describe().keys()}
|
|
83
83
|
self.max_threshold_dict = {key: np.ceil(df.describe()[key]["max"]) for key in df.describe().keys()}
|
|
84
|
+
|
|
85
|
+
print("Minimum thresholds:", self.min_threshold_dict)
|
|
86
|
+
print("Maximum thresholds:", self.max_threshold_dict)
|
|
84
87
|
|
|
85
88
|
# Generate datasets and items
|
|
86
89
|
for column in self.columns:
|
|
@@ -103,6 +106,8 @@ class NerParameterSearch(BaseModel):
|
|
|
103
106
|
if column.startswith("entities"):
|
|
104
107
|
kwargs["entities_threshold"] = threshold
|
|
105
108
|
|
|
109
|
+
print(f"Loading dataset for column: {column} with threshold: {threshold}")
|
|
110
|
+
# Create a new dataset instance for each threshold
|
|
106
111
|
dataset = self.dataset_type(
|
|
107
112
|
columns=[column],
|
|
108
113
|
dataset_splitter_type=self.dataset_splitter_type,
|
|
@@ -86,7 +86,7 @@ ddi_fw/pipeline/__init__.py,sha256=tKDM_rW4vPjlYTeOkNgi9PujDzb4e9O3LK1w5wqnebw,2
|
|
|
86
86
|
ddi_fw/pipeline/multi_modal_combination_strategy.py,sha256=JSyuP71b1I1yuk0s2ecCJZTtCED85jBtkpwTUxibJvI,1706
|
|
87
87
|
ddi_fw/pipeline/multi_pipeline.py,sha256=EjJnA3Vzd-WeEvUBaA2LDOy_iQ5-2eW2VhtxvvxDPfQ,9857
|
|
88
88
|
ddi_fw/pipeline/multi_pipeline_org.py,sha256=AbErwu05-3YIPnCcXRsj-jxPJG8HG2H7cMZlGjzaYa8,9037
|
|
89
|
-
ddi_fw/pipeline/ner_pipeline.py,sha256=
|
|
89
|
+
ddi_fw/pipeline/ner_pipeline.py,sha256=AVp18Og4AozZvXwXcm94QNpUczdkawCaGAwoJoUYBp4,7246
|
|
90
90
|
ddi_fw/pipeline/pipeline.py,sha256=q1kMkW9-fOlrA4BOGUku40U_PuEYfcbtH2EvlRM4uTM,6243
|
|
91
91
|
ddi_fw/utils/__init__.py,sha256=WNxkQXk-694roG50D355TGLXstfdWVb_tUyr-PM-8rg,537
|
|
92
92
|
ddi_fw/utils/categorical_data_encoding_checker.py,sha256=T1X70Rh4atucAuqyUZmz-iFULllY9dY0NRyV9-jTjJ0,3438
|
|
@@ -101,7 +101,7 @@ ddi_fw/utils/zip_helper.py,sha256=YRZA4tKZVBJwGQM0_WK6L-y5MoqkKoC-nXuuHK6CU9I,55
|
|
|
101
101
|
ddi_fw/vectorization/__init__.py,sha256=LcJOpLVoLvHPDw9phGFlUQGeNcST_zKV-Oi1Pm5h_nE,110
|
|
102
102
|
ddi_fw/vectorization/feature_vector_generation.py,sha256=EBf-XAiwQwr68az91erEYNegfeqssBR29kVgrliIyac,4765
|
|
103
103
|
ddi_fw/vectorization/idf_helper.py,sha256=_Gd1dtDSLaw8o-o0JugzSKMt9FpeXewTh4wGEaUd4VQ,2571
|
|
104
|
-
ddi_fw-0.0.
|
|
105
|
-
ddi_fw-0.0.
|
|
106
|
-
ddi_fw-0.0.
|
|
107
|
-
ddi_fw-0.0.
|
|
104
|
+
ddi_fw-0.0.228.dist-info/METADATA,sha256=f0FUmtvqF6hIEG0HfiB2Epcy9prPHxa0i-OxIw5RNb8,2632
|
|
105
|
+
ddi_fw-0.0.228.dist-info/WHEEL,sha256=lTU6B6eIfYoiQJTZNc-fyaR6BpL6ehTzU3xGYxn2n8k,91
|
|
106
|
+
ddi_fw-0.0.228.dist-info/top_level.txt,sha256=PMwHICFZTZtcpzQNPV4UQnfNXYIeLR_Ste-Wfc1h810,7
|
|
107
|
+
ddi_fw-0.0.228.dist-info/RECORD,,
|
|
File without changes
|
|
File without changes
|