⚙️ Fine-Tuning for if
Statements
Figure 1: The output for my teammates' and my original Python code versus my teammates and my Python code with masked if
statements.
This project explores code completing suitable if
statements in Python functions, leveraging CodeT5 Transformer model from Hugging Face (small-sized model). CodeT5 is a pre-trained encoder-decoder Transformer model designed for code understanding and generation. It has been trained on a large corpus of code across multiple programming languages and supports a range of downstream tasks such as code completion, summarization, translation, and generation.