Compare commits

...

10 Commits

Author SHA1 Message Date
Niels Rogge b60f7b51d9 encoder-encoder -> encoder-decoder (#1)
- encoder-encoder -> encoder-decoder (efb1f2a83adf338d09a84cacb045aeac50454c24)


Co-authored-by: scott schmidt <scottschmidt@users.noreply.huggingface.co>
2023-03-10 12:32:02 +00:00
Joao Gante 85ba1baa39 Adding generation config file(s) 2023-01-24 16:57:20 +00:00
Qian Liu 968109c940 Update config.json 2022-05-17 08:25:49 +00:00
Qian Liu ec98f1dbc9 Delete tokenizer.json 2022-05-17 07:31:49 +00:00
Qian Liu d9f3aaa913 Upload tokenizer_config.json 2022-05-17 07:31:35 +00:00
Niels Rogge 96afae46dc Update README.md 2022-05-05 07:01:31 +00:00
Qian Liu e88355ec1a Upload config.json 2022-03-10 05:03:35 +00:00
Qian Liu 09df58cdd5 Upload config.json
Keep aligned with bart-base on forced_bos_token_id
2022-03-10 05:00:12 +00:00
Qian Liu df3358a924 Update README.md 2022-02-28 04:33:51 +00:00
Qian Liu 74976ce54e Upload config.json 2022-02-24 12:54:56 +00:00
5 changed files with 115 additions and 137 deletions

127
README.md
View File

@ -1,64 +1,65 @@
--- ---
language: en language: en
tags: tags:
- tapex - tapex
license: apache-2.0 - table-question-answering
--- license: mit
---
# TAPEX (base-sized model)
# TAPEX (base-sized model)
TAPEX was proposed in [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou. The original repo can be found [here](https://github.com/microsoft/Table-Pretraining).
TAPEX was proposed in [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou. The original repo can be found [here](https://github.com/microsoft/Table-Pretraining).
## Model description
## Model description
TAPEX (**Ta**ble **P**re-training via **Ex**ecution) is a conceptually simple and empirically powerful pre-training approach to empower existing models with *table reasoning* skills. TAPEX realizes table pre-training by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries.
TAPEX (**Ta**ble **P**re-training via **Ex**ecution) is a conceptually simple and empirically powerful pre-training approach to empower existing models with *table reasoning* skills. TAPEX realizes table pre-training by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries.
TAPEX is based on the BART architecture, the transformer encoder-encoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder.
TAPEX is based on the BART architecture, the transformer encoder-decoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder.
## Intended Uses
## Intended Uses
You can use the raw model for simulating neural SQL execution, i.e., employ TAPEX to execute a SQL query on a given table. However, the model is mostly meant to be fine-tuned on a supervised dataset. Currently TAPEX can be fine-tuned to tackle table question answering tasks and table fact verification tasks. See the [model hub](https://huggingface.co/models?search=tapex) to look for fine-tuned versions on a task that interests you.
You can use the raw model for simulating neural SQL execution, i.e., employ TAPEX to execute a SQL query on a given table. However, the model is mostly meant to be fine-tuned on a supervised dataset. Currently TAPEX can be fine-tuned to tackle table question answering tasks and table fact verification tasks. See the [model hub](https://huggingface.co/models?search=tapex) to look for fine-tuned versions on a task that interests you.
### How to Use
### How to Use
Here is how to use this model in transformers:
Here is how to use this model in transformers:
```python
from transformers import TapexTokenizer, BartForConditionalGeneration ```python
import pandas as pd from transformers import TapexTokenizer, BartForConditionalGeneration
import pandas as pd
tokenizer = TapexTokenizer.from_pretrained("microsoft/tapex-base")
model = BartForConditionalGeneration.from_pretrained("microsoft/tapex-base") tokenizer = TapexTokenizer.from_pretrained("microsoft/tapex-base")
model = BartForConditionalGeneration.from_pretrained("microsoft/tapex-base")
data = {
"year": [1896, 1900, 1904, 2004, 2008, 2012], data = {
"city": ["athens", "paris", "st. louis", "athens", "beijing", "london"] "year": [1896, 1900, 1904, 2004, 2008, 2012],
} "city": ["athens", "paris", "st. louis", "athens", "beijing", "london"]
table = pd.DataFrame.from_dict(data) }
table = pd.DataFrame.from_dict(data)
# tapex accepts uncased input since it is pre-trained on the uncased corpus
query = "select year where city = beijing" # tapex accepts uncased input since it is pre-trained on the uncased corpus
encoding = tokenizer(table=table, query=query, return_tensors="pt") query = "select year where city = beijing"
encoding = tokenizer(table=table, query=query, return_tensors="pt")
outputs = model.generate(**encoding)
outputs = model.generate(**encoding)
print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
# ['2008'] print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
``` # ['2008']
```
### How to Fine-tuning
### How to Fine-tuning
Please find the fine-tuning script [here](https://github.com/SivilTaram/transformers/tree/add_tapex_bis/examples/research_projects/tapex).
Please find the fine-tuning script [here](https://github.com/SivilTaram/transformers/tree/add_tapex_bis/examples/research_projects/tapex).
### BibTeX entry and citation info
### BibTeX entry and citation info
```bibtex
@inproceedings{ ```bibtex
liu2022tapex, @inproceedings{
title={{TAPEX}: Table Pre-training via Learning a Neural {SQL} Executor}, liu2022tapex,
author={Qian Liu and Bei Chen and Jiaqi Guo and Morteza Ziyadi and Zeqi Lin and Weizhu Chen and Jian-Guang Lou}, title={{TAPEX}: Table Pre-training via Learning a Neural {SQL} Executor},
booktitle={International Conference on Learning Representations}, author={Qian Liu and Bei Chen and Jiaqi Guo and Morteza Ziyadi and Zeqi Lin and Weizhu Chen and Jian-Guang Lou},
year={2022}, booktitle={International Conference on Learning Representations},
url={https://openreview.net/forum?id=O50443AsCP} year={2022},
} url={https://openreview.net/forum?id=O50443AsCP}
}
``` ```

View File

@ -1,73 +1,39 @@
{ {
"_name_or_path": "bart-base", "_name_or_path": "microsoft/tapex-base",
"activation_dropout": 0.0, "activation_dropout": 0.0,
"activation_function": "gelu", "activation_function": "gelu",
"add_bias_logits": false, "architectures": [
"add_final_layer_norm": false, "BartForConditionalGeneration"
"architectures": [ ],
"BartModel" "attention_dropout": 0.1,
], "bos_token_id": 0,
"attention_dropout": 0.1, "classifier_dropout": 0.0,
"bos_token_id": 0, "d_model": 768,
"classifier_dropout": 0.0, "decoder_attention_heads": 12,
"d_model": 768, "decoder_ffn_dim": 3072,
"decoder_attention_heads": 12, "decoder_layerdrop": 0.0,
"decoder_ffn_dim": 3072, "decoder_layers": 6,
"decoder_layerdrop": 0.0, "decoder_start_token_id": 2,
"decoder_layers": 6, "dropout": 0.1,
"decoder_start_token_id": 2, "encoder_attention_heads": 12,
"dropout": 0.1, "encoder_ffn_dim": 3072,
"early_stopping": true, "encoder_layerdrop": 0.0,
"encoder_attention_heads": 12, "encoder_layers": 6,
"encoder_ffn_dim": 3072, "eos_token_id": 2,
"encoder_layerdrop": 0.0, "forced_bos_token_id": 0,
"encoder_layers": 6, "forced_eos_token_id": 2,
"eos_token_id": 2, "init_std": 0.02,
"forced_eos_token_id": 2, "is_encoder_decoder": true,
"gradient_checkpointing": false, "max_position_embeddings": 1024,
"id2label": { "model_type": "bart",
"0": "LABEL_0", "normalize_before": false,
"1": "LABEL_1", "normalize_embedding": true,
"2": "LABEL_2" "num_beams": 4,
}, "num_hidden_layers": 6,
"init_std": 0.02, "pad_token_id": 1,
"is_encoder_decoder": true, "scale_embedding": false,
"label2id": { "torch_dtype": "float32",
"LABEL_0": 0, "transformers_version": "4.17.0.dev0",
"LABEL_1": 1, "use_cache": true,
"LABEL_2": 2 "vocab_size": 50265
}, }
"max_position_embeddings": 1024,
"model_type": "bart",
"no_repeat_ngram_size": 3,
"normalize_before": false,
"normalize_embedding": true,
"num_beams": 4,
"num_hidden_layers": 6,
"pad_token_id": 1,
"scale_embedding": false,
"task_specific_params": {
"summarization": {
"length_penalty": 1.0,
"max_length": 128,
"min_length": 12,
"num_beams": 4
},
"summarization_cnn": {
"length_penalty": 2.0,
"max_length": 142,
"min_length": 56,
"num_beams": 4
},
"summarization_xsum": {
"length_penalty": 1.0,
"max_length": 62,
"min_length": 11,
"num_beams": 6
}
},
"torch_dtype": "float32",
"transformers_version": "4.12.0.dev0",
"use_cache": true,
"vocab_size": 50265
}

11
generation_config.json Normal file
View File

@ -0,0 +1,11 @@
{
"_from_model_config": true,
"bos_token_id": 0,
"decoder_start_token_id": 2,
"eos_token_id": 2,
"forced_bos_token_id": 0,
"forced_eos_token_id": 2,
"num_beams": 4,
"pad_token_id": 1,
"transformers_version": "4.27.0.dev0"
}

File diff suppressed because one or more lines are too long

1
tokenizer_config.json Normal file
View File

@ -0,0 +1 @@
{"do_lower_case": true, "errors": "replace", "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "max_cell_length": 15, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "microsoft/tapex-base", "tokenizer_class": "TapexTokenizer"}