aditismile/resume_enhnaced

phi3-full-resume-enhancer

This is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct for resume enhancement and professional writing.

Model Description

This model transforms unstructured, informal resumes into professional, well-formatted resumes with:

• Quantified achievements
• Action-oriented language
• Professional formatting
• Enhanced skill descriptions
• Structured sections

Training Data

The model was fine-tuned on 5 high-quality examples demonstrating transformation from casual/unstructured resumes to professional formats across various roles (Software Developer, Marketing Professional, Data Analyst, Project Manager, Software Engineer).

How to Use

Installation

pip install transformers peft torch

Basic Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel
import torch

# Load model and tokenizer
base_model_id = "microsoft/Phi-3-mini-4k-instruct"
adapter_model_id = "aditismile/resume_enhnaced"

tokenizer = AutoTokenizer.from_pretrained(base_model_id, trust_remote_code=True)
base_model = AutoModelForCausalLM.from_pretrained(
    base_model_id,
    device_map="auto",
    torch_dtype=torch.float16,
    trust_remote_code=True
)

model = PeftModel.from_pretrained(base_model, adapter_model_id)
model.eval()

# Prepare input
resume_text = """John Doe
john@email.com

Summary: Developer with some experience

Work:
- Coded stuff at Company XYZ
- Fixed bugs

Skills: Python, JavaScript"""

prompt = f"""<|system|>
You are an expert resume writer. Transform the following resume into a professional format.<|end|>
<|user|>
{resume_text}<|end|>
<|assistant|>
"""

# Generate enhanced resume
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(
    **inputs,
    max_new_tokens=800,
    temperature=0.7,
    top_p=0.9,
    do_sample=True
)

result = tokenizer.decode(outputs[0], skip_special_tokens=True)
enhanced_resume = result.split("<|assistant|>")[-1].strip()
print(enhanced_resume)

Training Details

• Base Model: microsoft/Phi-3-mini-4k-instruct
• Fine-tuning Method: LoRA (Low-Rank Adaptation)
• Training Examples: 5 full resume transformations
• Max Length: 2048 tokens
• Epochs: 5
• Learning Rate: 2e-4
• LoRA Rank: 16
• LoRA Alpha: 32

Limitations

• Trained on a small dataset (5 examples) - may benefit from additional training data
• Best suited for technical and professional roles
• May hallucinate quantifiable metrics if not present in original resume
• Designed for English resumes only

Intended Use

This model is intended for:

• Resume enhancement and professional writing assistance
• Career coaching tools
• Job application preparation
• Educational purposes in resume writing

Citation

If you use this model, please cite:

@misc{phi3_full_resume_enhancer,
  author = {aditismile},
  title = {{Phi-3 Resume Enhancement Model}},
  year = {{2024}},
  publisher = {{Hugging Face}},
  howpublished = {{\url{{https://huggingface.co/aditismile/resume_enhnaced}}}}
}

License

This model inherits the license from microsoft/Phi-3-mini-4k-instruct. The fine-tuned adapters are released under MIT license.

Contact

For questions or feedback, please open an issue on the model repository.