Instructions to use legal-pythia/crime-type-classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use legal-pythia/crime-type-classifier with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("legal-pythia/crime-type-classifier", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update config.json
Browse files- config.json +0 -37
config.json
CHANGED
|
@@ -1,42 +1,5 @@
|
|
| 1 |
{
|
| 2 |
"model_type": "deberta-v2",
|
| 3 |
-
"architectures": ["DebertaV2ForSequenceClassification"],
|
| 4 |
-
"problem_type": "multi_label_classification",
|
| 5 |
-
"num_labels": 14,
|
| 6 |
-
|
| 7 |
-
"id2label": {
|
| 8 |
-
"0": "robbery",
|
| 9 |
-
"1": "murder",
|
| 10 |
-
"2": "assault",
|
| 11 |
-
"3": "theft",
|
| 12 |
-
"4": "kidnapping",
|
| 13 |
-
"5": "burglary",
|
| 14 |
-
"6": "stabbing",
|
| 15 |
-
"7": "shooting",
|
| 16 |
-
"8": "carjacking",
|
| 17 |
-
"9": "fraud",
|
| 18 |
-
"10": "scam",
|
| 19 |
-
"11": "arson",
|
| 20 |
-
"12": "vandalism",
|
| 21 |
-
"13": "armed robbery"
|
| 22 |
-
},
|
| 23 |
-
|
| 24 |
-
"label2id": {
|
| 25 |
-
"robbery": 0,
|
| 26 |
-
"murder": 1,
|
| 27 |
-
"assault": 2,
|
| 28 |
-
"theft": 3,
|
| 29 |
-
"kidnapping": 4,
|
| 30 |
-
"burglary": 5,
|
| 31 |
-
"stabbing": 6,
|
| 32 |
-
"shooting": 7,
|
| 33 |
-
"carjacking": 8,
|
| 34 |
-
"fraud": 9,
|
| 35 |
-
"scam": 10,
|
| 36 |
-
"arson": 11,
|
| 37 |
-
"vandalism": 12,
|
| 38 |
-
"armed robbery": 13
|
| 39 |
-
},
|
| 40 |
"attention_probs_dropout_prob": 0.1,
|
| 41 |
"hidden_act": "gelu",
|
| 42 |
"hidden_dropout_prob": 0.1,
|
|
|
|
| 1 |
{
|
| 2 |
"model_type": "deberta-v2",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"attention_probs_dropout_prob": 0.1,
|
| 4 |
"hidden_act": "gelu",
|
| 5 |
"hidden_dropout_prob": 0.1,
|