---
license: apache-2.0
datasets:
- briefai/LongShort-Dataset
language:
- en
pipeline_tag: text-generation
tags:
- pytorch
- llama-2
- Gen-AI
- Finance
- KPI Extraction
---

<p align="center">
<img src="https://github.com/brief-ai-uchicago/Branding/blob/main/SVG/brief_logo_white_circle.svg" height="200px" width="200px"/>
</p>

# LongShort-Llama-2-7B
- Model creator: [Brief AI](https://huggingface.co/briefai)
- Original model: [Llama 2 7B Chat](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf)

### Model Description

This model leverages Llama-2-7B architecture to extract financial KPIs from the earnings call documents. The model is fine-tuned on 5K data samples.

### Dataset Description
Data Source: Factiva
Data Description: 28K+ Earnings Call Documents
Data Scope: 1K+ public companies, 
Fine Tuning Data: Collection of 60K+ samples.

![image/png](https://cdn-uploads.huggingface.co/production/uploads/65370806ca70ffcd7214ee26/ho_C9XOE6iF27X8hpy_EM.png)
## Prompt template: LongShort-Llama-2-7B

```
[INST]Given the context, answer the question.

### Question:
Extract all the finance-based performance indicators and evaluation metrics.

### Context:
{context}

### Answer:
[/INST]

```

## Uses

Financial KPI and Metric Extraction


## Evaluation

![image/png](https://cdn-uploads.huggingface.co/production/uploads/65370806ca70ffcd7214ee26/FHrvR-sgdW0A1goJVVrp0.png)

LongShort-Llama-2-7B is giving 44.4% accuracy on a validation set of 10% of the original training dataset.

## Thanks