terumi commited on
Commit
5d3b771
1 Parent(s): ca384fc

First version of biobertpt-all model and tokenizer.

Browse files

README.md CHANGED
@@ -1 +1,24 @@
1
- "hello"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <img src="https://raw.githubusercontent.com/HAILab-PUCPR/BioBERTpt/master/logo-biobertpr1.png" alt="Logo BioBERTpt">
2
+
3
+ # BioBERTpt - Portuguese Clinical and Biomedical BERT
4
+
5
+ The [BioBERTpt - A Portuguese Neural Language Model for Clinical Named Entity Recognition](https://www.aclweb.org/anthology/2020.clinicalnlp-1.7/) paper contains clinical and biomedical BERT-based models for Portuguese Language, initialized with BERT-Multilingual-Cased & trained on clinical notes and biomedical literature.
6
+
7
+ This model card describes the BioBERTpt(all) model, a full version with clinical narratives and biomedical literature in Portuguese language.
8
+
9
+ ## How to use the model
10
+
11
+ Load the model via the transformers library:
12
+ ```
13
+ from transformers import AutoTokenizer, AutoModel
14
+ tokenizer = AutoTokenizer.from_pretrained("pucpr/biobertpt-all")
15
+ model = AutoModel.from_pretrained("pucpr/biobertpt-all")
16
+ ```
17
+
18
+ ## More Information
19
+
20
+ Refer to the original paper, [BioBERTpt - A Portuguese Neural Language Model for Clinical Named Entity Recognition](https://www.aclweb.org/anthology/2020.clinicalnlp-1.7/) for additional details and performance on Portuguese NER tasks.
21
+
22
+ ## Questions?
23
+
24
+ Post a Github issue on the [BioBERTpt repo](https://github.com/HAILab-PUCPR/BioBERTpt).
config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_num_labels": 2,
3
+ "architectures": [
4
+ "BertForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "directionality": "bidi",
9
+ "do_sample": false,
10
+ "early_stopping": false,
11
+ "eos_token_ids": null,
12
+ "finetuning_task": null,
13
+ "hidden_act": "gelu",
14
+ "hidden_dropout_prob": 0.1,
15
+ "hidden_size": 768,
16
+ "id2label": {
17
+ "0": "LABEL_0",
18
+ "1": "LABEL_1"
19
+ },
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 3072,
22
+ "is_decoder": false,
23
+ "is_encoder_decoder": false,
24
+ "label2id": {
25
+ "LABEL_0": 0,
26
+ "LABEL_1": 1
27
+ },
28
+ "layer_norm_eps": 1e-12,
29
+ "length_penalty": 1.0,
30
+ "max_length": 20,
31
+ "max_position_embeddings": 512,
32
+ "min_length": 0,
33
+ "model_type": "bert",
34
+ "no_repeat_ngram_size": 0,
35
+ "num_attention_heads": 12,
36
+ "num_beams": 1,
37
+ "num_hidden_layers": 12,
38
+ "num_return_sequences": 1,
39
+ "output_attentions": false,
40
+ "output_hidden_states": false,
41
+ "output_past": true,
42
+ "pad_token_id": null,
43
+ "pooler_fc_size": 768,
44
+ "pooler_num_attention_heads": 12,
45
+ "pooler_num_fc_layers": 3,
46
+ "pooler_size_per_head": 128,
47
+ "pooler_type": "first_token_transform",
48
+ "pruned_heads": {},
49
+ "repetition_penalty": 1.0,
50
+ "temperature": 1.0,
51
+ "top_k": 50,
52
+ "top_p": 1.0,
53
+ "torchscript": false,
54
+ "type_vocab_size": 2,
55
+ "use_bfloat16": false,
56
+ "vocab_size": 119547
57
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6351c9e5220971b536c32884ee3711bcf1f1da5bb186312a6b44f1fc2c26817
3
+ size 714311810
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:049ae81a141d362f6fcf829395db9c7b1f44ffebd5c30e8636376e1a57f59489
3
+ size 711690400
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
1
+ {"do_lower_case": false, "max_len": 512}
vocab.txt ADDED
The diff for this file is too large to render. See raw diff