wei commited on
Commit
e8190a4
1 Parent(s): 2b8e76c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +72 -0
README.md CHANGED
@@ -5,3 +5,75 @@ widget:
5
  - text: "protected String renderUri ( URI uri ) { return uri . toASCIIString ( ) ; }"
6
 
7
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  - text: "protected String renderUri ( URI uri ) { return uri . toASCIIString ( ) ; }"
6
 
7
  ---
8
+
9
+
10
+ # CodeTrans model for code documentation generation java
11
+ Pretrained model on programming language java using the t5 small model architecture. It was first released in
12
+ [this repository](https://github.com/agemagician/CodeTrans). This model is trained on tokenized java code functions: it works best with tokenized java functions.
13
+
14
+
15
+ ## Model description
16
+
17
+ This CodeTrans model is based on the `t5-small` 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.
18
+
19
+ ## Intended uses & limitations
20
+
21
+ The model could be used to generate the description for the java function or be fine-tuned on other java code tasks. It can be used on unparsed and untokenized java code. However, if the java code is tokenized, the performance should be better.
22
+
23
+ ### How to use
24
+
25
+ Here is how to use this model to generate java function documentation using Transformers SummarizationPipeline:
26
+
27
+ ```python
28
+ from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline
29
+
30
+ pipeline = SummarizationPipeline(
31
+ model=AutoModelWithLMHead.from_pretrained("SEBIS/code_trans_t5_small_code_comment_generation_java_multitask"),
32
+ tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_small_code_comment_generation_java_multitask", skip_special_tokens=True),
33
+ device=0
34
+ )
35
+
36
+ tokenized_code = "protected String renderUri ( URI uri ) { return uri . toASCIIString ( ) ; }"
37
+ pipeline([tokenized_code])
38
+ ```
39
+ Run this example in [colab notebook](https://github.com/agemagician/CodeTrans/blob/main/prediction/multitask/pre-training/code%20comment%20generation/small_model.ipynb).
40
+ ## Training data
41
+
42
+ 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)
43
+
44
+
45
+ ## Training procedure
46
+
47
+ ### Multi-task Pretraining
48
+
49
+ The model was trained on a single TPU Pod V3-8 for 360,000 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
+
54
+ ## Evaluation results
55
+
56
+ For the code documentation tasks, different models achieves the following results on different programming languages (in BLEU score):
57
+
58
+ Test results :
59
+
60
+ | Language / Model | Java |
61
+ | -------------------- | :------------: |
62
+ | CodeTrans-ST-Small | 37.98 |
63
+ | CodeTrans-ST-Base | 38.07 |
64
+ | CodeTrans-TF-Small | 38.56 |
65
+ | CodeTrans-TF-Base | 39.06 |
66
+ | CodeTrans-TF-Large | **39.50** |
67
+ | CodeTrans-MT-Small | 20.15 |
68
+ | CodeTrans-MT-Base | 27.44 |
69
+ | CodeTrans-MT-Large | 34.69 |
70
+ | CodeTrans-MT-TF-Small | 38.37 |
71
+ | CodeTrans-MT-TF-Base | 38.90 |
72
+ | CodeTrans-MT-TF-Large | 39.25 |
73
+ | State of the art | 38.17 |
74
+
75
+
76
+
77
+ > 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/)
78
+
79
+