README.md

metadata

license: llama3.2
language:
  - en
base_model:
  - meta-llama/Llama-3.2-1B
pipeline_tag: text-classification
library_name: transformers
tags:
  - regression
  - story-point-estimation
  - software-engineering
datasets:
  - duracloud
metrics:
  - mae
  - mdae
model-index:
  - name: llama-3.2-1b-story-point-estimation
    results:
      - task:
          type: regression
          name: Story Point Estimation
        dataset:
          type: duracloud
          name: duracloud Dataset
          split: test
        metrics:
          - type: mae
            value: 1.118
            name: Mean Absolute Error (MAE)
          - type: mdae
            value: 0.944
            name: Median Absolute Error (MdAE)

LLAMA 3 Story Point Estimator - duracloud

This model is fine-tuned on issue descriptions from duracloud and tested on duracloud for story point estimation.

Model Details

• Base Model: LLAMA 3.2 1B

• Training Project: duracloud

• Test Project: duracloud

• Task: Story Point Estimation (Regression)

• Architecture: PEFT (LoRA)

• Input: Issue titles

• Output: Story point estimation (continuous value)

Usage

from transformers import AutoModelForSequenceClassification, AutoTokenizer
from peft import PeftModel

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("DEVCamiloSepulveda/0-LLAMA3SP-duracloud")
model = AutoModelForSequenceClassification.from_pretrained("DEVCamiloSepulveda/0-LLAMA3SP-duracloud")

# Prepare input text
text = "Your issue description here"
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=20, padding="max_length")

# Get prediction
outputs = model(**inputs)
story_points = outputs.logits.item()

Training Details

• Fine-tuning method: LoRA (Low-Rank Adaptation)
• Sequence length: 20 tokens
• Best training epoch: 13 / 20 epochs
• Batch size: 32
• Training time: 288.061 seconds
• Mean Absolute Error (MAE): 1.118
• Median Absolute Error (MdAE): 0.944