Hi, I want to run CodeT5-base on code generation task. I run the command: python run_exp.py --model_tag codet5_base --task concode --sub_task none

There is an error: 'tuple' object has no attribute 'loss'.

I try to change outputs = model(input_ids=source_ids, attention_mask=source_mask, labels=target_ids, decoder_attention_mask=target_mask) to outputs, _ = model(input_ids=source_ids, attention_mask=source_mask, labels=target_ids, decoder_attention_mask=target_mask)

There is an error: too many values to unpack (expected 2)

What should I do?

Dear authors,

I noticed in the paper you mentioned that you pre-train the T5 model with identifier-aware denoising for 100 epochs and further pre-train with bimodal generation for 50 epochs. I was wondering the released model only includes the first 100 epochs or the whole 150 epochs?

Thanks in advance for your clarification

Is it possible to make code summarization for raw Java code?

I can't find the example of inference for code summarization. Could you please provide an example? E.g., I expect the following code:

from transformers import RobertaTokenizer,  WHICH_MODELTO_USE

tokenizer = RobertaTokenizer.from_pretrained('Salesforce/codet5-base')
model = WHICH_MODELTO_USE.from_pretrained('Salesforce/codet5-base')

java_code = 'int i = 0; ++i;  int b = runSomeFunction(i); extract(b);'
code_summarization = model.predict(java_code)
print(code_summarization)

The expected result is the following: 'Extracts and returns max value'

Is it possible to make such the prediction? The problem is I can't understand how you are translating from code to the vector which will be used to predict the summarization without pretraining procedures.

Could you please provide an example?

I have a couple of questions:

a) How can I use CodeT5 to extract embedding for JavaScript and Python code? b) Can I feed incomplete code JavaScript and Python snippet to extract embedding? Or the code snippet needs to be complete? c) Have anyone used CodeT5 to perform code to code search?

Hi,

I am hoping to reproduce the results on code clone detection task. This might seem like a silly question but the fined-tuned checkpoints released doesn't include the RobertaClassificationHead parameters, right? I am able to load only the T5ForConditionalGeneration model using the provided checkpoints for the task.

So, how do I go about loading the entire CloneModel?

The readme mentioned that this is used for Code autocompletion in VSCode, I wonder how to use CodeT5 without fine-tuning as a language model to complete code given some previous context in code?

Can I use it for Code generation? For example if I give a query, "Add two numbers", and it should generate the code for that. And if Yes, can you please suggest how can I prepare the dataset for this task or can I use the dataset which you mentioned.

Thank you

Can you kindly elaborate on how we can use the fine-tuned checkpoints for the prediction of new data in concode task? Say this is my prediction data: {"code": "public integer sum(Integer arg0,Integer arg1) {return result;}", "nl": "Add two integers. concode_field_sep int sum concode_field_sep int result"} If I understand correctly then concode is supposed to complete these functions. However, I am not sure how to generate prediction on this sample data. I tried replacing the test file containing original test data with this sample test data and then ran this command python run_exp.py --model_tag codet5_small --task concode --sub_task none This command starts with training, then evaluating and finally testing. However, I am interested in only prediction. Isn't there any way to directly generate predictions from fine-tuned model on concode ? Kindly let me know if I am doing something wrong.

Hi, the newly added AI coding assistant demo is cool! I have a few questions about it:

1. Did you make the codeT5 model into VS Code plugin? How did you do that?

2. When demonstrating code generation, editing the comment generates the corresponding code snippet. Aren't the inputs to the code generation model natural Language Description and class Environment?

The format of the data in the Concode dataset is

{
    "code": "int function ( double [ ] arg0 , double [ ] arg1 ) { int loc0 = arg0 . length - arg1 . length ; outer : for ( int loc1 = 0 ; loc1 <= loc0 ; loc1 ++ ) { for ( int loc2 = 0 ; loc2 < arg1 . length ; loc2 ++ ) { if ( ne ( arg0 [ loc1 + loc2 ] , arg1 [ loc2 ] ) ) { continue outer ; } } return ( loc1 ) ; } return ( - 1 ) ; }",
    "nl": "searches for the first subsequence of a that matches sub elementwise . elements of sub are considered to match elements of a if they pass the #eq test . concode_field_sep double max_ratio concode_elem_sep double min_ratio concode_elem_sep boolean off concode_field_sep boolean isElemMatch concode_elem_sep int compare concode_elem_sep boolean isSubset concode_elem_sep boolean ne concode_elem_sep boolean lt concode_elem_sep boolean gte concode_elem_sep void set_rel_diff concode_elem_sep boolean eq concode_elem_sep boolean lte concode_elem_sep boolean gt"
}

Is it possible to generate accurate code snippet by typing only comments without class Environment? Doesn't the loss of context information affect the quality of the generated code?

I've finetuned CodeT5 large on a small python dataset(~1700) data points. I see that the results are more or less correct but the code is not always compilable(due to inconsistent spacing and new line characters). Any idea on fixing this? And how CodeBLEU work if the code generated by the model isn't compilable? The model might generate non compilable code during initial phases of the training right?

Hello, I have spent quite some time trying to run the fine-tuned model you kindly provided, on the clone detection task, with no luck. Could you provide minimal python scripts to load and run the model on each of the downstream tasks, as was done for ntp and summarization on 🤗?

Thanks a lot!

Hi.

Me and my two colleagues are interested in replicating the results of CodeT5-base on code generation task with our own dataset. However we're having a few hiccups on preprocessing data, and we hope you don't mind a few questions.

Mainly, we're wondering how you dealt with the Google BigQuery data alongside CodeSearchNet. From our knowledge, CodeSearchNet data is consisted of codes that are nicely isolated blocks of function codes, while the BigQuery data and other extra data of C/C# from open-source Github repositories is, as far as we guess, isn't presented in such a convenient matter.

Our own data is also in a similar position where none of the codes are either isolated to blocks of function codes, but rather mostly a complete file of itself. We were wondering if you did any preprocessing of your own so that the extra data of C/C# would match that of CodeSearchNet, or if you just used it raw.

And if you did use it raw, has it affected the performance compared to when the model was trained only with CodeSearchNet data? Thank you in advance.

P.S. My colleagues are also wondering how you dealt with the whitespace, arguing that the paper wasn't so clear with that. One argues that you discarded whitespace all-together, while the other argues that you only removed duplicates of whitespce into one instance. ex) A. '\s\s\s' --> '' B. '\s\s\s' --> '\s'