ubaid0405/tinyllama-1.1b-docstring-gen

Model Description

This model is a LoRA fine-tuned version of TinyLlama-1.1B-Chat-v1.0, trained to generate Python docstrings for given function code.

• —Base model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
• —Dataset: codesearchnet (Python subset)
• —Method: Supervised Fine-Tuning (SFT) with LoRA
• —Developed by: Ubaid

Training Results

How to Use

\\\`python from peft import AutoPeftModelForCausalLM from transformers import AutoTokenizer

repo = "ubaid0405/tinyllama-1.1b-docstring-gen" model = AutoPeftModelForCausalLM.frompretrained(repo, devicemap="auto") tok = AutoTokenizer.from_pretrained(repo)

code = "def add(a, b):\n return a + b" prompt = f"[INST] Write a docstring:\n{code} [/INST]"

ids = tok(prompt, returntensors="pt").inputids.to(model.device) out = model.generate(ids, maxnewtokens=100) print(tok.decode(out[0][ids.shape[1]:], skipspecialtokens=True)) \\\`