Jon Ander commited on
Commit
df352e1
1 Parent(s): 624b5a4

Added SciBERT-SQuAD-QuAC model

Browse files
Files changed (4) hide show
  1. README.md +26 -0
  2. config.json +54 -0
  3. pytorch_model.bin +3 -0
  4. vocab.txt +0 -0
README.md ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ ---
4
+
5
+ # SciBERT-SQuAD-QuAC
6
+
7
+ This is the [SciBERT language representation model](https://huggingface.co/allenai/scibert_scivocab_uncased) fine tuned for Question Answering. SciBERT is a pre-trained language model based on BERT that has been trained on a large corpus of scientific text. When fine tuning for Question Answering we combined [SQuAD2.0](https://www.aclweb.org/anthology/P18-2124/) and [QuAC](https://arxiv.org/abs/1808.07036) datasets.
8
+
9
+ If using this model, please cite the following paper:
10
+ ```
11
+ @inproceedings{otegi-etal-2020-automatic,
12
+ title = "Automatic Evaluation vs. User Preference in Neural Textual {Q}uestion{A}nswering over {COVID}-19 Scientific Literature",
13
+ author = "Otegi, Arantxa and
14
+ Campos, Jon Ander and
15
+ Azkune, Gorka and
16
+ Soroa, Aitor and
17
+ Agirre, Eneko",
18
+ booktitle = "Proceedings of the 1st Workshop on {NLP} for {COVID}-19 (Part 2) at {EMNLP} 2020",
19
+ month = dec,
20
+ year = "2020",
21
+ address = "Online",
22
+ publisher = "Association for Computational Linguistics",
23
+ url = "https://www.aclweb.org/anthology/2020.nlpcovid19-2.15",
24
+ doi = "10.18653/v1/2020.nlpcovid19-2.15",
25
+ }
26
+ ```
config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_num_labels": 2,
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "decoder_start_token_id": null,
9
+ "do_sample": false,
10
+ "early_stopping": false,
11
+ "eos_token_id": 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": 0,
43
+ "prefix": null,
44
+ "pruned_heads": {},
45
+ "repetition_penalty": 1.0,
46
+ "task_specific_params": null,
47
+ "temperature": 1.0,
48
+ "top_k": 50,
49
+ "top_p": 1.0,
50
+ "torchscript": false,
51
+ "type_vocab_size": 2,
52
+ "use_bfloat16": false,
53
+ "vocab_size": 31090
54
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:865781d9dad87cc2822b17e845109d10c428bf20e1d19938269d0f0d0e972004
3
+ size 439727846
vocab.txt ADDED
The diff for this file is too large to render. See raw diff