wei commited on
Commit
c31379b
1 Parent(s): 8a1e750

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +78 -0
README.md CHANGED
@@ -5,3 +5,81 @@ widget:
5
  - text: '''with open ( CODE_STRING , CODE_STRING ) as in_file : buf = in_file . readlines ( ) with open ( CODE_STRING , CODE_STRING ) as out_file : for line in buf : if line == " ; Include this text " : line = line + " Include below " out_file . write ( line ) '''
6
 
7
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  - text: '''with open ( CODE_STRING , CODE_STRING ) as in_file : buf = in_file . readlines ( ) with open ( CODE_STRING , CODE_STRING ) as out_file : for line in buf : if line == " ; Include this text " : line = line + " Include below " out_file . write ( line ) '''
6
 
7
  ---
8
+
9
+
10
+
11
+ # CodeTrans model for source code summarization python
12
+ Pretrained model on programming language python 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 python code functions: it works best with tokenized python 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 source code summarization task for the python code snippets.
19
+
20
+
21
+
22
+ ## Intended uses & limitations
23
+
24
+ The model could be used to generate the description for the python function or be fine-tuned on other python code tasks. It can be used on unparsed and untokenized python code. However, if the python code is tokenized, the performance should be better.
25
+
26
+ ### How to use
27
+
28
+ Here is how to use this model to generate python function documentation using Transformers SummarizationPipeline:
29
+
30
+ ```python
31
+ from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline
32
+
33
+ pipeline = SummarizationPipeline(
34
+ model=AutoModelWithLMHead.from_pretrained("SEBIS/code_trans_t5_large_source_code_summarization_python_multitask_finetune"),
35
+ tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_large_source_code_summarization_python_multitask_finetune", skip_special_tokens=True),
36
+ device=0
37
+ )
38
+
39
+ tokenized_code = '''with open ( CODE_STRING , CODE_STRING ) as in_file : buf = in_file . readlines ( ) with open ( CODE_STRING , CODE_STRING ) as out_file : for line in buf : if line == " ; Include this text " : line = line + " Include below " out_file . write ( line ) '''
40
+ pipeline([tokenized_code])
41
+ ```
42
+ Run this example in [colab notebook](https://github.com/agemagician/CodeTrans/blob/main/prediction/multitask/fine-tuning/source%20code%20summarization/python/large_model.ipynb).
43
+ ## Training data
44
+
45
+ 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)
46
+
47
+
48
+ ## Training procedure
49
+
50
+ ### Multi-task Pretraining
51
+
52
+ The model was trained on a single TPU Pod V3-8 for 500,000 steps in total, using sequence length 512 (batch size 4096).
53
+ It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture.
54
+ The optimizer used is AdaFactor with inverse square root learning rate schedule for pre-training.
55
+
56
+ ### Fine-tuning
57
+
58
+ This model was then fine-tuned on a single TPU Pod V2-8 for 100 steps in total, using sequence length 512 (batch size 256), using only the dataset only containing python code.
59
+
60
+
61
+ ## Evaluation results
62
+
63
+ For the source code summarization tasks, different models achieves the following results on different programming languages (in BLEU score):
64
+
65
+ Test results :
66
+
67
+ | Language / Model | Python | SQL | C# |
68
+ | -------------------- | :------------: | :------------: | :------------: |
69
+ | CodeTrans-ST-Small | 8.45 | 17.55 | 19.74 |
70
+ | CodeTrans-ST-Base | 9.12 | 15.00 | 18.65 |
71
+ | CodeTrans-TF-Small | 10.06 | 17.71 | 20.40 |
72
+ | CodeTrans-TF-Base | 10.94 | 17.66 | 21.12 |
73
+ | CodeTrans-TF-Large | 12.41 | 18.40 | 21.43 |
74
+ | CodeTrans-MT-Small | 13.11 | 19.15 | 22.39 |
75
+ | CodeTrans-MT-Base | **13.37** | 19.24 | 23.20 |
76
+ | CodeTrans-MT-Large | 13.24 | 19.40 | **23.57** |
77
+ | CodeTrans-MT-TF-Small | 12.10 | 18.25 | 22.03 |
78
+ | CodeTrans-MT-TF-Base | 10.64 | 16.91 | 21.40 |
79
+ | CodeTrans-MT-TF-Large | 12.14 | **19.98** | 21.10 |
80
+ | CODE-NN | -- | 18.40 | 20.50 |
81
+
82
+
83
+ > 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/)
84
+
85
+