yilunzhao commited on
Commit
ea4fd88
1 Parent(s): fd46991

Upload folder using huggingface_hub

Browse files
README.md CHANGED
@@ -1,3 +1,60 @@
1
  ---
2
- license: mit
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language: en
3
+ tags:
4
+ - table-question-answering
5
+ datasets:
6
+ - wikisql
7
  ---
8
+
9
+ # ReasTAP
10
+
11
+ ReasTAP is a table reasoning model proposed in the EMNLP 2022 paper [ReasTAP: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examples](https://arxiv.org/pdf/2210.12374.pdf). The original Github repository is [https://github.com/Yale-LILY/ReasTAP](https://github.com/Yale-LILY/ReasTAP).
12
+
13
+ ## Description
14
+
15
+ `Yale-LILY/reastap-large-finetuned-wikisql` is initialized with `Yale-LILY/reastap-large` and finetuned on [WikiSQL](https://huggingface.co/datasets/wikisql).
16
+
17
+ ## Usage
18
+
19
+ ```python
20
+ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
21
+ import pandas as pd
22
+
23
+ tokenizer = AutoTokenizer.from_pretrained("Yale-LILY/reastap-large-finetuned-wikisql")
24
+ model = AutoModelForSeq2SeqLM.from_pretrained("Yale-LILY/reastap-large-finetuned-wikisql")
25
+
26
+ data = {
27
+ "year": [1896, 1900, 1904, 2004, 2008, 2012],
28
+ "city": ["athens", "paris", "st. louis", "athens", "beijing", "london"]
29
+ }
30
+ table = pd.DataFrame.from_dict(data)
31
+
32
+ query = "In which year did beijing host the Olympic Games?"
33
+ encoding = tokenizer(table=table, query=query, return_tensors="pt")
34
+
35
+ outputs = model.generate(**encoding)
36
+
37
+ print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
38
+ # [' 2008']
39
+ ```
40
+
41
+ ## Reference
42
+
43
+ ```bibtex
44
+ @inproceedings{zhao-etal-2022-reastap,
45
+ title = "{R}eas{TAP}: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examples",
46
+ author = "Zhao, Yilun and
47
+ Nan, Linyong and
48
+ Qi, Zhenting and
49
+ Zhang, Rui and
50
+ Radev, Dragomir",
51
+ booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
52
+ month = dec,
53
+ year = "2022",
54
+ address = "Abu Dhabi, United Arab Emirates",
55
+ publisher = "Association for Computational Linguistics",
56
+ url = "https://aclanthology.org/2022.emnlp-main.615",
57
+ pages = "9006--9018",
58
+ abstract = "Reasoning over tabular data requires both table structure understanding and a broad set of table reasoning skills. Current models with table-specific architectures and pre-training methods perform well on understanding table structures, but they still struggle with tasks that require various table reasoning skills. In this work, we develop ReasTAP to show that high-level table reasoning skills can be injected into models during pre-training without a complex table-specific architecture design. We define 7 table reasoning skills, such as numerical operation, temporal comparison, and conjunction. Each reasoning skill is associated with one example generator, which synthesizes questions over semi-structured tables according to the sampled templates. We model the table pre-training task as a sequence generation task and pre-train ReasTAP to generate precise answers of the synthetic examples. ReasTAP is evaluated on four benchmarks covering three downstream tasks including 1) WikiSQL-Weak and WikiTQ for Table Question Answering, 2) TabFact for Table Fact Verification, and 3) LogicNLG for Faithful Table-to-Text Generation. Experimental results demonstrate that ReasTAP achieves new state-of-the-art results on all of them and delivers a significant improvement under low-resource setting. Our code is publicly available at https://github.com/Yale-LILY/ReasTAP.",
59
+ }
60
+ ```
config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Yale-LILY/reastap-large-finetuned-wikisql",
3
+ "activation_dropout": 0.1,
4
+ "activation_function": "gelu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "BartForConditionalGeneration"
9
+ ],
10
+ "attention_dropout": 0.1,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.1,
13
+ "classifier_dropout": 0.0,
14
+ "d_model": 1024,
15
+ "decoder_attention_heads": 16,
16
+ "decoder_ffn_dim": 4096,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 12,
19
+ "decoder_start_token_id": 2,
20
+ "dropout": 0.1,
21
+ "encoder_attention_heads": 16,
22
+ "encoder_ffn_dim": 4096,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 12,
25
+ "eos_token_id": 2,
26
+ "forced_bos_token_id": 0,
27
+ "forced_eos_token_id": 2,
28
+ "gradient_checkpointing": false,
29
+ "id2label": {
30
+ "0": "LABEL_0",
31
+ "1": "LABEL_1",
32
+ "2": "LABEL_2"
33
+ },
34
+ "init_std": 0.02,
35
+ "is_encoder_decoder": true,
36
+ "label2id": {
37
+ "LABEL_0": 0,
38
+ "LABEL_1": 1,
39
+ "LABEL_2": 2
40
+ },
41
+ "max_length": 1024,
42
+ "max_position_embeddings": 1024,
43
+ "model_type": "bart",
44
+ "no_repeat_ngram_size": 3,
45
+ "normalize_before": false,
46
+ "num_beams": 4,
47
+ "num_hidden_layers": 12,
48
+ "pad_token_id": 1,
49
+ "scale_embedding": false,
50
+ "task_specific_params": {
51
+ "summarization": {
52
+ "length_penalty": 1.0,
53
+ "max_length": 128,
54
+ "min_length": 12,
55
+ "num_beams": 4
56
+ },
57
+ "summarization_cnn": {
58
+ "length_penalty": 2.0,
59
+ "max_length": 142,
60
+ "min_length": 56,
61
+ "num_beams": 4
62
+ },
63
+ "summarization_xsum": {
64
+ "length_penalty": 1.0,
65
+ "max_length": 62,
66
+ "min_length": 11,
67
+ "num_beams": 6
68
+ }
69
+ },
70
+ "torch_dtype": "float32",
71
+ "transformers_version": "4.17.0",
72
+ "use_cache": true,
73
+ "vocab_size": 50265
74
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "eos_token_id": 2,
5
+ "forced_bos_token_id": 0,
6
+ "forced_eos_token_id": 2,
7
+ "max_length": 1024,
8
+ "no_repeat_ngram_size": 3,
9
+ "num_beams": 4,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.17.0"
12
+ }
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 sha256:46cefa8bd8146bdf8e0c00e14956e00ad1e7e0d8661d11cea8122e2e632557a7
3
+ size 1625541389
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "bos_token": {
4
+ "__type": "AddedToken",
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "clean_up_tokenization_spaces": true,
12
+ "cls_token": {
13
+ "__type": "AddedToken",
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "do_lower_case": true,
21
+ "eos_token": {
22
+ "__type": "AddedToken",
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": true,
26
+ "rstrip": false,
27
+ "single_word": false
28
+ },
29
+ "errors": "replace",
30
+ "mask_token": {
31
+ "__type": "AddedToken",
32
+ "content": "<mask>",
33
+ "lstrip": true,
34
+ "normalized": true,
35
+ "rstrip": false,
36
+ "single_word": false
37
+ },
38
+ "max_cell_length": 15,
39
+ "model_max_length": 1024,
40
+ "pad_token": {
41
+ "__type": "AddedToken",
42
+ "content": "<pad>",
43
+ "lstrip": false,
44
+ "normalized": true,
45
+ "rstrip": false,
46
+ "single_word": false
47
+ },
48
+ "sep_token": {
49
+ "__type": "AddedToken",
50
+ "content": "</s>",
51
+ "lstrip": false,
52
+ "normalized": true,
53
+ "rstrip": false,
54
+ "single_word": false
55
+ },
56
+ "tokenizer_class": "TapexTokenizer",
57
+ "unk_token": {
58
+ "__type": "AddedToken",
59
+ "content": "<unk>",
60
+ "lstrip": false,
61
+ "normalized": true,
62
+ "rstrip": false,
63
+ "single_word": false
64
+ },
65
+ "use_fast": true
66
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff