vccarvalho11 commited on
Commit
1496de7
·
verified ·
1 Parent(s): ac5a071

Upload distilbert-NER ONNX model

Browse files
Files changed (4) hide show
  1. README.md +57 -0
  2. config.json +48 -0
  3. model.onnx +3 -0
  4. vocab.txt +0 -0
README.md ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: onnx
3
+ tags:
4
+ - distilbert
5
+ - ner
6
+ - named-entity-recognition
7
+ - token-classification
8
+ - conll2003
9
+ - onnx
10
+ - inference4j
11
+ license: apache-2.0
12
+ pipeline_tag: token-classification
13
+ ---
14
+
15
+ # DistilBERT NER — ONNX
16
+
17
+ ONNX export of [dslim/distilbert-NER](https://huggingface.co/dslim/distilbert-NER), a DistilBERT model fine-tuned on CoNLL-2003 for Named Entity Recognition. Identifies persons, organizations, locations, and miscellaneous entities in text using IOB2 tagging.
18
+
19
+ Mirrored for use with [inference4j](https://github.com/inference4j/inference4j), an inference-only AI library for Java.
20
+
21
+ ## Original Source
22
+
23
+ - **Repository:** [dslim (ONNX by onnx-community)](https://huggingface.co/dslim/distilbert-NER)
24
+ - **License:** apache-2.0
25
+
26
+ ## Usage with inference4j
27
+
28
+ ```java
29
+ try (BertNerRecognizer ner = BertNerRecognizer.builder()
30
+ .modelId("inference4j/distilbert-NER")
31
+ .build()) {
32
+ List<NamedEntity> entities = ner.recognize("John works at Google in London.");
33
+ for (NamedEntity e : entities) {
34
+ System.out.printf("%s (%s)%n", e.text(), e.label());
35
+ }
36
+ // John (PER)
37
+ // Google (ORG)
38
+ // London (LOC)
39
+ }
40
+ ```
41
+
42
+ ## Model Details
43
+
44
+ | Property | Value |
45
+ |----------|-------|
46
+ | Architecture | DistilBERT (6 layers, 768 hidden, 66M params) |
47
+ | Task | Named Entity Recognition (IOB2 tagging) |
48
+ | Labels | O, B-PER, I-PER, B-ORG, I-ORG, B-LOC, I-LOC, B-MISC, I-MISC |
49
+ | Training data | CoNLL-2003 |
50
+ | F1 score | 92.17 |
51
+ | Max sequence length | 512 |
52
+ | Tokenizer | WordPiece (cased) |
53
+ | Original framework | PyTorch (HuggingFace Transformers) |
54
+
55
+ ## License
56
+
57
+ This model is licensed under the [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0). Original model by [dslim](https://huggingface.co/dslim/distilbert-NER), ONNX export by [onnx-community](https://huggingface.co/onnx-community).
config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "dslim/distilbert-NER",
4
+ "activation": "gelu",
5
+ "architectures": [
6
+ "DistilBertForTokenClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "dim": 768,
10
+ "dropout": 0.1,
11
+ "hidden_dim": 3072,
12
+ "id2label": {
13
+ "0": "O",
14
+ "1": "B-PER",
15
+ "2": "I-PER",
16
+ "3": "B-ORG",
17
+ "4": "I-ORG",
18
+ "5": "B-LOC",
19
+ "6": "I-LOC",
20
+ "7": "B-MISC",
21
+ "8": "I-MISC"
22
+ },
23
+ "initializer_range": 0.02,
24
+ "label2id": {
25
+ "B-LOC": 5,
26
+ "B-MISC": 7,
27
+ "B-ORG": 3,
28
+ "B-PER": 1,
29
+ "I-LOC": 6,
30
+ "I-MISC": 8,
31
+ "I-ORG": 4,
32
+ "I-PER": 2,
33
+ "O": 0
34
+ },
35
+ "max_position_embeddings": 512,
36
+ "model_type": "distilbert",
37
+ "n_heads": 12,
38
+ "n_layers": 6,
39
+ "output_past": true,
40
+ "pad_token_id": 0,
41
+ "qa_dropout": 0.1,
42
+ "seq_classif_dropout": 0.2,
43
+ "sinusoidal_pos_embds": false,
44
+ "tie_weights_": true,
45
+ "torch_dtype": "float32",
46
+ "transformers_version": "4.49.0",
47
+ "vocab_size": 28996
48
+ }
model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5f77096d0d9f425d34a2e263f8a2dfb845cdc757dc00c7a1e69e9cbb93115d5
3
+ size 260924462
vocab.txt ADDED
The diff for this file is too large to render. See raw diff