You can deploy this model and start earning money today!

arnoxide/
boxcode-model

safetensors

gpt2

Views

No views yet

Fine-tuned Code Model for Vite, React, Next.js, Node.js, Python, and Go

This model is a fine-tuned version of microsoft/CodeGPT-small-java-adaptedgpt2 adapted for code understanding and generation tasks across several technologies: Vite, React, Next.js, Node.js, Python, and Go.

The model was fine-tuned on code data extracted from private GitHub repositories.

How to use

You can load and use this model with the transformers library:

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_path = "." # Or the repository ID after pushing to Hugging Face Hub
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(model_path)

# Example inference
input_text = "def fibonacci("
input_ids = tokenizer(input_text, return_tensors="pt").input_ids
output_ids = model.generate(input_ids, max_length=50)
output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
print(output_text)

Fine-tuning Details

Base Model: microsoft/CodeGPT-small-java-adaptedgpt2
Fine-tuning Data: Code from private GitHub repositories focusing on Vite, React, Next.js, Node.js, Python, and Go.
Training Parameters: (Include key parameters from TrainingArguments here, e.g., epochs, batch size)
- Number of Epochs: 3
- Per Device Train Batch Size: 2
- Learning Rate: 2e-5

Limitations

This model was fine-tuned on a limited private dataset and may not generalize well to all code scenarios or languages.
The model size is optimized for deployment on Hugging Face Spaces with a 2GB limit, which might impact its performance compared to larger models.

arnoxide/boxcode-model

Fine-tuned Code Model for Vite, React, Next.js, Node.js, Python, and Go

How to use

Fine-tuning Details

Limitations

arnoxide/
boxcode-model