Update README.md
Browse files
README.md
CHANGED
@@ -5,3 +5,78 @@ widget:
|
|
5 |
- text: "parse the uses licence node of this package , if any , and returns the license definition if theres"
|
6 |
|
7 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
- text: "parse the uses licence node of this package , if any , and returns the license definition if theres"
|
6 |
|
7 |
---
|
8 |
+
|
9 |
+
|
10 |
+
# CodeTrans model for api recommendation generation
|
11 |
+
Pretrained model for api recommendation generation using the t5 base model architecture. It was first released in
|
12 |
+
[this repository](https://github.com/agemagician/CodeTrans).
|
13 |
+
|
14 |
+
|
15 |
+
## Model description
|
16 |
+
|
17 |
+
This CodeTrans model is based on the `t5-base` model. It has its own SentencePiece vocabulary model. It used transfer-learning pre-training on 7 unsupervised datasets in the software development domain. It is then fine-tuned on the api recommendation generation task for the java apis.
|
18 |
+
|
19 |
+
## Intended uses & limitations
|
20 |
+
|
21 |
+
The model could be used to generate api usage for the java programming tasks.
|
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_base_api_generation_transfer_learning_finetune"),
|
32 |
+
tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_base_api_generation_transfer_learning_finetune", skip_special_tokens=True),
|
33 |
+
device=0
|
34 |
+
)
|
35 |
+
|
36 |
+
tokenized_code = "parse the uses licence node of this package , if any , and returns the license definition if theres"
|
37 |
+
pipeline([tokenized_code])
|
38 |
+
```
|
39 |
+
Run this example in [colab notebook](https://github.com/agemagician/CodeTrans/blob/main/prediction/transfer%20learning%20fine-tuning/api%20generation/base_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 |
+
### Transfer-learning Pretraining
|
48 |
+
|
49 |
+
The model was trained on a single TPU Pod V3-8 for 1,400,000 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 |
+
### Fine-tuning
|
54 |
+
|
55 |
+
This model was then fine-tuned on a single TPU Pod V3-8 for 340,000 steps in total, using sequence length 512 (batch size 256), using only the dataset only containing api recommendation generation data.
|
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 |
+
| Language / Model | Java |
|
65 |
+
| -------------------- | :------------: |
|
66 |
+
| CodeTrans-ST-Small | 68.71 |
|
67 |
+
| CodeTrans-ST-Base | 70.45 |
|
68 |
+
| CodeTrans-TF-Small | 68.90 |
|
69 |
+
| CodeTrans-TF-Base | 72.11 |
|
70 |
+
| CodeTrans-TF-Large | 73.26 |
|
71 |
+
| CodeTrans-MT-Small | 58.43 |
|
72 |
+
| CodeTrans-MT-Base | 67.97 |
|
73 |
+
| CodeTrans-MT-Large | 72.29 |
|
74 |
+
| CodeTrans-MT-TF-Small | 69.29 |
|
75 |
+
| CodeTrans-MT-TF-Base | 72.89 |
|
76 |
+
| CodeTrans-MT-TF-Large | **73.39** |
|
77 |
+
| State of the art | 54.42 |
|
78 |
+
|
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 |
+
|