Model Card for Model ID

A BERT-like model pre-trained on Java buggy code.

Model Details

Model Description

A BERT-like model pre-trained on Java buggy code.

Developed by: André Nascimento
Shared by: Hugging Face
Model type: Fill-Mask
Language(s) (NLP): Java (EN)
License: [More Information Needed]
Finetuned from model: BERT Base Uncased

Uses

Direct Use

Fill-Mask.

Downstream Use [optional]

The model can be used for other tasks, like Text Classification.

Out-of-Scope Use

[More Information Needed]

Bias, Risks, and Limitations

[More Information Needed]

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model.

from transformers import pipeline
unmasker = pipeline('fill-mask', model='bert-java-bfp_single')
unmasker(java_code) # Replace with Java code; Use '[MASK]' to mask tokens/words in the code.

[More Information Needed]

Training Details

Training Data

The model was trained on 236040 Java methods, containing the code before and after the bug fix was applied. The whole dataset was built from Extracted Bug-Fix Pairs (BFP), extracting single file/single method commits, and keeping only method with less than 512 tokens. An 80/20 train/validation split was applied afterwards.

Training Procedure

Preprocessing [optional]

Remove comments and replace consecutive whitespace characters by a single space.

Training Hyperparameters

Training regime: fp16 mixed precision

Speeds, Sizes, Times [optional]

[More Information Needed]

Evaluation

Testing Data, Factors & Metrics

Testing Data

The model was evaluated on 59024 Java methods, from the 20% split of the dataset mentioned in Training Data

[More Information Needed]

Factors

[More Information Needed]

Metrics

Perplexity

Results

1.73

Summary

Model Examination [optional]

[More Information Needed]

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Hardware Type: [More Information Needed]
Hours used: [More Information Needed]
Cloud Provider: [More Information Needed]
Compute Region: [More Information Needed]
Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]