wei commited on
Commit
246bacf
1 Parent(s): e699735

Update from weiding

Browse files
README.md ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - summarization
4
+ widget:
5
+ - text: "def add ( severity , progname , & block ) return true if io . nil? || severity < level message = format_message ( severity , progname , yield ) MUTEX . synchronize { io . write ( message ) } true end"
6
+
7
+ ---
8
+
9
+
10
+
11
+ # CodeTrans model for code documentation generation ruby
12
+ Pretrained model on programming language ruby using the t5 large model architecture. It was first released in
13
+ [this repository](https://github.com/agemagician/CodeTrans). This model is trained on tokenized ruby code functions: it works best with tokenized ruby functions.
14
+
15
+
16
+ ## Model description
17
+
18
+ This CodeTrans model is based on the `t5-large` model. It has its own SentencePiece vocabulary model. It used multi-task training on 13 supervised tasks in the software development domain and 7 unsupervised datasets. It is then fine-tuned on the code documentation generation task for the ruby function/method.
19
+
20
+ ## Intended uses & limitations
21
+
22
+ The model could be used to generate the description for the ruby function or be fine-tuned on other ruby code tasks. It can be used on unparsed and untokenized ruby code. However, if the ruby code is tokenized, the performance should be better.
23
+
24
+ ### How to use
25
+
26
+ Here is how to use this model to generate ruby function documentation using Transformers SummarizationPipeline:
27
+
28
+ ```python
29
+ from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline
30
+
31
+ pipeline = SummarizationPipeline(
32
+ model=AutoModelWithLMHead.from_pretrained("SEBIS/code_trans_t5_large_code_documentation_generation_ruby_multitask_finetune"),
33
+ tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_large_code_documentation_generation_ruby_multitask_finetune", skip_special_tokens=True),
34
+ device=0
35
+ )
36
+
37
+ tokenized_code = "def add ( severity , progname , & block ) return true if io . nil? || severity < level message = format_message ( severity , progname , yield ) MUTEX . synchronize { io . write ( message ) } true end"
38
+ pipeline([tokenized_code])
39
+ ```
40
+ Run this example in [colab notebook](https://github.com/agemagician/CodeTrans/blob/main/prediction/multitask/fine-tuning/function%20documentation%20generation/ruby/large_model.ipynb).
41
+ ## Training data
42
+
43
+ The supervised training tasks datasets can be downloaded on [Link](https://www.dropbox.com/sh/488bq2of10r4wvw/AACs5CGIQuwtsD7j_Ls_JAORa/finetuning_dataset?dl=0&subfolder_nav_tracking=1)
44
+
45
+ ## Training procedure
46
+
47
+ ### Multi-task Pretraining
48
+
49
+ The model was trained on a single TPU Pod V3-8 for half million steps in total, using sequence length 512 (batch size 4096).
50
+ It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture.
51
+ The optimizer used is AdaFactor with inverse square root learning rate schedule for pre-training.
52
+
53
+ ### Fine-tuning
54
+
55
+ This model was then fine-tuned on a single TPU Pod V2-8 for 2,000 steps in total, using sequence length 512 (batch size 256), using only the dataset only containing ruby code.
56
+
57
+
58
+ ## Evaluation results
59
+
60
+ For the code documentation tasks, different models achieves the following results on different programming languages (in BLEU score):
61
+
62
+ Test results :
63
+
64
+
65
+ | Language / Model | Python | Java | Go | Php | Ruby | JavaScript |
66
+ | -------------------- | :------------: | :------------: | :------------: | :------------: | :------------: | :------------: |
67
+ | CodeTrans-ST-Small | 17.31 | 16.65 | 16.89 | 23.05 | 9.19 | 13.7 |
68
+ | CodeTrans-ST-Base | 16.86 | 17.17 | 17.16 | 22.98 | 8.23 | 13.17 |
69
+ | CodeTrans-TF-Small | 19.93 | 19.48 | 18.88 | 25.35 | 13.15 | 17.23 |
70
+ | CodeTrans-TF-Base | 20.26 | 20.19 | 19.50 | 25.84 | 14.07 | 18.25 |
71
+ | CodeTrans-TF-Large | 20.35 | 20.06 | **19.54** | 26.18 | 14.94 | **18.98** |
72
+ | CodeTrans-MT-Small | 19.64 | 19.00 | 19.15 | 24.68 | 14.91 | 15.26 |
73
+ | CodeTrans-MT-Base | **20.39** | 21.22 | 19.43 | **26.23** | **15.26** | 16.11 |
74
+ | CodeTrans-MT-Large | 20.18 | **21.87** | 19.38 | 26.08 | 15.00 | 16.23 |
75
+ | CodeTrans-MT-TF-Small | 19.77 | 20.04 | 19.36 | 25.55 | 13.70 | 17.24 |
76
+ | CodeTrans-MT-TF-Base | 19.77 | 21.12 | 18.86 | 25.79 | 14.24 | 18.62 |
77
+ | CodeTrans-MT-TF-Large | 18.94 | 21.42 | 18.77 | 26.20 | 14.19 | 18.83 |
78
+ | State of the art | 19.06 | 17.65 | 18.07 | 25.16 | 12.16 | 14.90 |
79
+
80
+
81
+ > Created by [Ahmed Elnaggar](https://twitter.com/Elnaggar_AI) | [LinkedIn](https://www.linkedin.com/in/prof-ahmed-elnaggar/) and Wei Ding | [LinkedIn](https://www.linkedin.com/in/wei-ding-92561270/)
82
+
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5Model"
4
+ ],
5
+ "d_ff": 4096,
6
+ "d_kv": 64,
7
+ "d_model": 1024,
8
+ "decoder_start_token_id": 0,
9
+ "dropout_rate": 0.1,
10
+ "eos_token_id": 1,
11
+ "initializer_factor": 1.0,
12
+ "is_encoder_decoder": true,
13
+ "layer_norm_epsilon": 1e-06,
14
+ "model_type": "t5",
15
+ "n_positions": 512,
16
+ "num_decoder_layers": 24,
17
+ "num_heads": 16,
18
+ "num_layers": 24,
19
+ "output_past": true,
20
+ "pad_token_id": 0,
21
+ "relative_attention_num_buckets": 32,
22
+ "task_specific_params": {
23
+ "summarization": {
24
+ "max_length": 512,
25
+ "num_beams": 4,
26
+ "prefix": "function documentation generation ruby: "
27
+ }
28
+ },
29
+ "vocab_size": 32128
30
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0703b6f3872b64fa4d5e47c7924108bea4e87933477506a6b6c91043565a481
3
+ size 2950910481
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9856b76e9978cc5805f0566cedabd2fc7bdb1a3ee22d52545100c056cb09a59c
3
+ size 797030
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": false}