Upload POLARYS DistilBERT activity classifier
Browse files- README.md +33 -0
- config.json +46 -0
- label_map.json +20 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
base_model: distilbert-base-uncased
|
| 6 |
+
tags:
|
| 7 |
+
- polarys
|
| 8 |
+
- text-classification
|
| 9 |
+
- patrol-notes
|
| 10 |
+
pipeline_tag: text-classification
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# POLARYS BERT Activity Classifier
|
| 14 |
+
|
| 15 |
+
DistilBERT fine-tuned to classify Ermita PS-5 patrol activity notes into categories such as:
|
| 16 |
+
|
| 17 |
+
- Presence
|
| 18 |
+
- Assistance
|
| 19 |
+
- Security Coverage
|
| 20 |
+
- Oplan / Special Ops
|
| 21 |
+
- Community Engagement
|
| 22 |
+
- Clearing Operations
|
| 23 |
+
- Custodial / Jail
|
| 24 |
+
- Other Operations
|
| 25 |
+
|
| 26 |
+
## Intended use
|
| 27 |
+
|
| 28 |
+
Analyze officer-submitted descriptive notes for activity categorization with confidence scores.
|
| 29 |
+
Decision-support only — does not approve or modify reports.
|
| 30 |
+
|
| 31 |
+
## Training data
|
| 32 |
+
|
| 33 |
+
Cleaned `POLARYS_DATASETS.csv` WHAT/REMARKS text mapped to normalized activity labels.
|
config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DistilBertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"dtype": "float32",
|
| 11 |
+
"eos_token_id": null,
|
| 12 |
+
"hidden_dim": 3072,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "Assistance",
|
| 15 |
+
"1": "Clearing Operations",
|
| 16 |
+
"2": "Community Engagement",
|
| 17 |
+
"3": "Custodial / Jail",
|
| 18 |
+
"4": "Oplan / Special Ops",
|
| 19 |
+
"5": "Presence",
|
| 20 |
+
"6": "Security Coverage"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"label2id": {
|
| 24 |
+
"Assistance": 0,
|
| 25 |
+
"Clearing Operations": 1,
|
| 26 |
+
"Community Engagement": 2,
|
| 27 |
+
"Custodial / Jail": 3,
|
| 28 |
+
"Oplan / Special Ops": 4,
|
| 29 |
+
"Presence": 5,
|
| 30 |
+
"Security Coverage": 6
|
| 31 |
+
},
|
| 32 |
+
"max_position_embeddings": 512,
|
| 33 |
+
"model_type": "distilbert",
|
| 34 |
+
"n_heads": 12,
|
| 35 |
+
"n_layers": 6,
|
| 36 |
+
"pad_token_id": 0,
|
| 37 |
+
"problem_type": "single_label_classification",
|
| 38 |
+
"qa_dropout": 0.1,
|
| 39 |
+
"seq_classif_dropout": 0.2,
|
| 40 |
+
"sinusoidal_pos_embds": false,
|
| 41 |
+
"tie_weights_": true,
|
| 42 |
+
"tie_word_embeddings": true,
|
| 43 |
+
"transformers_version": "5.14.1",
|
| 44 |
+
"use_cache": false,
|
| 45 |
+
"vocab_size": 30522
|
| 46 |
+
}
|
label_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"label2id": {
|
| 3 |
+
"Assistance": 0,
|
| 4 |
+
"Clearing Operations": 1,
|
| 5 |
+
"Community Engagement": 2,
|
| 6 |
+
"Custodial / Jail": 3,
|
| 7 |
+
"Oplan / Special Ops": 4,
|
| 8 |
+
"Presence": 5,
|
| 9 |
+
"Security Coverage": 6
|
| 10 |
+
},
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Assistance",
|
| 13 |
+
"1": "Clearing Operations",
|
| 14 |
+
"2": "Community Engagement",
|
| 15 |
+
"3": "Custodial / Jail",
|
| 16 |
+
"4": "Oplan / Special Ops",
|
| 17 |
+
"5": "Presence",
|
| 18 |
+
"6": "Security Coverage"
|
| 19 |
+
}
|
| 20 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c1daad9e3d94cf7beaec28c20b2bc8d138f5f7a5496d18d10c9353e5edf6a00
|
| 3 |
+
size 267847948
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": true,
|
| 5 |
+
"is_local": false,
|
| 6 |
+
"local_files_only": false,
|
| 7 |
+
"mask_token": "[MASK]",
|
| 8 |
+
"model_max_length": 512,
|
| 9 |
+
"pad_token": "[PAD]",
|
| 10 |
+
"sep_token": "[SEP]",
|
| 11 |
+
"strip_accents": null,
|
| 12 |
+
"tokenize_chinese_chars": true,
|
| 13 |
+
"tokenizer_class": "BertTokenizer",
|
| 14 |
+
"unk_token": "[UNK]"
|
| 15 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8913e914164fa24a7a56b21374a2e681fecfc5058cc931d8aa9a0a22f851f3f9
|
| 3 |
+
size 4792
|