Zhengbao Jiang commited on
Commit
0bf7f3b
1 Parent(s): 0de63cc

init commit

Browse files
Files changed (7) hide show
  1. README.md +52 -0
  2. config.json +38 -0
  3. merges.txt +0 -0
  4. pytorch_model.bin +3 -0
  5. special_tokens_map.json +1 -0
  6. tokenizer_config.json +67 -0
  7. vocab.json +0 -0
README.md ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - tapex
5
+ - table-question-answering
6
+ datasets:
7
+ - wikitablequestions
8
+ ---
9
+
10
+ # OmniTab
11
+
12
+ OmniTab is a table-based QA model proposed in [OmniTab: Pretraining with Natural and Synthetic Data for Few-shot Table-based Question Answering](https://arxiv.org/pdf/2207.03637.pdf). The original Github repository is [https://github.com/jzbjyb/OmniTab](https://github.com/jzbjyb/OmniTab).
13
+
14
+ ## Description
15
+
16
+ `neulab/omnitab-large-128shot` (based on BART architecture) is initialized with `microsoft/tapex-large` and continuously pretrained on natural and synthetic data (SQL2NL model trained in the 128-shot setting).
17
+
18
+ ## Usage
19
+
20
+ ```python
21
+ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
22
+ import pandas as pd
23
+
24
+ tokenizer = AutoTokenizer.from_pretrained("neulab/omnitab-large-128shot")
25
+ model = AutoModelForSeq2SeqLM.from_pretrained("neulab/omnitab-large-128shot")
26
+
27
+ data = {
28
+ "year": [1896, 1900, 1904, 2004, 2008, 2012],
29
+ "city": ["athens", "paris", "st. louis", "athens", "beijing", "london"]
30
+ }
31
+ table = pd.DataFrame.from_dict(data)
32
+
33
+ query = "In which year did beijing host the Olympic Games?"
34
+ encoding = tokenizer(table=table, query=query, return_tensors="pt")
35
+
36
+ outputs = model.generate(**encoding)
37
+
38
+ print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
39
+ # [' 2008']
40
+ ```
41
+
42
+ ## Reference
43
+
44
+ ```bibtex
45
+ @inproceedings{jiang-etal-2022-omnitab,
46
+ title = "{O}mni{T}ab: Pretraining with Natural and Synthetic Data for Few-shot Table-based Question Answering",
47
+ author = "Jiang, Zhengbao and Mao, Yi and He, Pengcheng and Neubig, Graham and Chen, Weizhu",
48
+ booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
49
+ month = jul,
50
+ year = "2022",
51
+ }
52
+ ```
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "neulab/omnitab-large-128shot",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "BartForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "classifier_dropout": 0.0,
11
+ "d_model": 1024,
12
+ "decoder_attention_heads": 16,
13
+ "decoder_ffn_dim": 4096,
14
+ "decoder_layerdrop": 0.0,
15
+ "decoder_layers": 12,
16
+ "decoder_start_token_id": 2,
17
+ "dropout": 0.1,
18
+ "encoder_attention_heads": 16,
19
+ "encoder_ffn_dim": 4096,
20
+ "encoder_layerdrop": 0.0,
21
+ "encoder_layers": 12,
22
+ "eos_token_id": 2,
23
+ "forced_bos_token_id": 0,
24
+ "forced_eos_token_id": 2,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "max_length": 1024,
28
+ "max_position_embeddings": 1024,
29
+ "model_type": "bart",
30
+ "num_hidden_layers": 12,
31
+ "pad_token_id": 1,
32
+ "scale_embedding": false,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.17.0.dev0",
35
+ "use_cache": true,
36
+ "vocab_size": 50265,
37
+ "no_repeat_ngram_size": 3
38
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50dfb30432e7b1e1b7245e0e14c3bccd31ef5bc12387ab433d7ea5ca23fee025
3
+ size 812847041
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}
tokenizer_config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_lower_case": false,
3
+ "errors": "replace",
4
+ "bos_token": {
5
+ "content": "<s>",
6
+ "single_word": false,
7
+ "lstrip": false,
8
+ "rstrip": false,
9
+ "normalized": true,
10
+ "__type": "AddedToken"
11
+ },
12
+ "eos_token": {
13
+ "content": "</s>",
14
+ "single_word": false,
15
+ "lstrip": false,
16
+ "rstrip": false,
17
+ "normalized": true,
18
+ "__type": "AddedToken"
19
+ },
20
+ "unk_token": {
21
+ "content": "<unk>",
22
+ "single_word": false,
23
+ "lstrip": false,
24
+ "rstrip": false,
25
+ "normalized": true,
26
+ "__type": "AddedToken"
27
+ },
28
+ "sep_token": {
29
+ "content": "</s>",
30
+ "single_word": false,
31
+ "lstrip": false,
32
+ "rstrip": false,
33
+ "normalized": true,
34
+ "__type": "AddedToken"
35
+ },
36
+ "cls_token": {
37
+ "content": "<s>",
38
+ "single_word": false,
39
+ "lstrip": false,
40
+ "rstrip": false,
41
+ "normalized": true,
42
+ "__type": "AddedToken"
43
+ },
44
+ "pad_token": {
45
+ "content": "<pad>",
46
+ "single_word": false,
47
+ "lstrip": false,
48
+ "rstrip": false,
49
+ "normalized": true,
50
+ "__type": "AddedToken"
51
+ },
52
+ "mask_token": {
53
+ "content": "<mask>",
54
+ "single_word": false,
55
+ "lstrip": true,
56
+ "rstrip": false,
57
+ "normalized": true,
58
+ "__type": "AddedToken"
59
+ },
60
+ "add_prefix_space": false,
61
+ "max_cell_length": 15,
62
+ "model_max_length": 1024,
63
+ "special_tokens_map_file": null,
64
+ "name_or_path": "neulab/omnitab-large-128shot",
65
+ "use_fast": true,
66
+ "tokenizer_class": "TapexTokenizer"
67
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff