CodeTrans model for code documentation generation php
Pretrained model on programming language php using the t5 small model architecture. It was first released in
this repository. This model is trained on tokenized php code functions: it works best with tokenized php functions.
Model description
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. It is then fine-tuned on the code documentation generation task for the php function/method.
Intended uses & limitations
The model could be used to generate the description for the php function or be fine-tuned on other php code tasks. It can be used on unparsed and untokenized php code. However, if the php code is tokenized, the performance should be better.
How to use
Here is how to use this model to generate php function documentation using Transformers SummarizationPipeline:
from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline
pipeline = SummarizationPipeline(
model=AutoModelWithLMHead.from_pretrained("SEBIS/code_trans_t5_small_code_documentation_generation_php_multitask_finetune"),
tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_small_code_documentation_generation_php_multitask_finetune", skip_special_tokens=True),
device=0
)
tokenized_code = "public static function update ( $ table ) { if ( ! is_array ( $ table ) ) { $ table = json_decode ( $ table , true ) ; } if ( ! SchemaManager :: tableExists ( $ table [ 'oldName' ] ) ) { throw SchemaException :: tableDoesNotExist ( $ table [ 'oldName' ] ) ; } $ updater = new self ( $ table ) ; $ updater -> updateTable ( ) ; }"
pipeline([tokenized_code])
Run this example in colab notebook.
Training data
The supervised training tasks datasets can be downloaded on Link
Training procedure
Multi-task Pretraining
The model was trained on a single TPU Pod V3-8 for half million steps in total, using sequence length 512 (batch size 4096).
It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture.
The optimizer used is AdaFactor with inverse square root learning rate schedule for pre-training.
Fine-tuning
This model was then fine-tuned on a single TPU Pod V2-8 for 10,000 steps in total, using sequence length 512 (batch size 256), using only the dataset only containing php code.
Test results :
| Language / Model |
Python |
Java |
Go |
Php |
Ruby |
JavaScript |
| CodeTrans-ST-Small |
17.31 |
16.65 |
16.89 |
23.05 |
9.19 |
13.7 |
| CodeTrans-ST-Base |
16.86 |
17.17 |
17.16 |
22.98 |
8.23 |
13.17 |
| CodeTrans-TF-Small |
19.93 |
19.48 |
18.88 |
25.35 |
13.15 |
17.23 |
| CodeTrans-TF-Base |
20.26 |
20.19 |
19.50 |
25.84 |
14.07 |
18.25 |
| CodeTrans-TF-Large |
20.35 |
20.06 |
19.54 |
26.18 |
14.94 |
18.98 |
| CodeTrans-MT-Small |
19.64 |
19.00 |
19.15 |
24.68 |
14.91 |
15.26 |
| CodeTrans-MT-Base |
20.39 |
21.22 |
19.43 |
26.23 |
15.26 |
16.11 |
| CodeTrans-MT-Large |
20.18 |
21.87 |
19.38 |
26.08 |
15.00 |
16.23 |
| CodeTrans-MT-TF-Small |
19.77 |
20.04 |
19.36 |
25.55 |
13.70 |
17.24 |
| CodeTrans-MT-TF-Base |
19.77 |
21.12 |
18.86 |
25.79 |
14.24 |
18.62 |
| CodeTrans-MT-TF-Large |
18.94 |
21.42 |
18.77 |
26.20 |
14.19 |
18.83 |
| State of the art |
19.06 |
17.65 |
18.07 |
25.16 |
12.16 |
14.90 |
Created by Ahmed Elnaggar | LinkedIn and Wei Ding | LinkedIn
