|
--- |
|
license: apache-2.0 |
|
language: |
|
- cpp |
|
metrics: |
|
- bleu |
|
library_name: transformers |
|
pipeline_tag: text-generation |
|
tags: |
|
- code-generation |
|
- code-completion |
|
- competitive-programming |
|
--- |
|
|
|
|
|
# CodeLanderAI Model |
|
|
|
This model is a fine-tuned version of the `CodeT5` model specifically designed for code completion in competitive programming. It was trained on a custom dataset of 12 million code samples derived from 2 million source code files. |
|
|
|
## Intended Use |
|
|
|
The model is intended for generating code completions based on the context provided by the user. It only supports cpp programming languages commonly used in competitive programming. |
|
|
|
### Languages Supported |
|
- C++ |
|
|
|
|
|
### Metrics |
|
|
|
The model was evaluated using the following metrics: |
|
|
|
- **BLEU Score:** Measures the quality of generated code against reference code. |
|
- **CodeBLEU:** A metric tailored for code generation, considering syntax and structure. |
|
- **Accuracy:** How often the model provides the correct code completion. |
|
- **Perplexity:** Indicates how well the model predicts the next token in a sequence. |
|
|
|
### Datasets |
|
|
|
The model was fine-tuned on a custom dataset containing code samples from competitive programming platforms. |
|
|
|
|
|
|
