Upload 2 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+HEA_query.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,14 +1,108 @@
+# 🔬 HEA Query
+
+**LLM-Powered Research Assistant for High Entropy Alloys**  
+*Hackathon: LLM for Materials Science*
+
+---
+
+## 🧠 Summary
+
+**HEA Query** is a research assistant built to enable intelligent access to both unstructured scientific literature and structured datasets related to **High Entropy Alloys (HEAs).**
+
+We combine:
+
+- 🔍 A semantic search engine over thousands of research paper chunks  
+- 📊 Cleaned and normalized tabular datasets containing alloy properties  
+- 🤖 A powerful LLM (Mistral-7B) to generate natural language answers  
+- 🖥️ An interactive Gradio interface for easy exploration
+
+This assistant helps materials scientists ask questions like:
+
+> 🧪 _"Which FCC alloys have Vickers hardness above 200?"_  
+> 📈 _"Show me alloys with high entropy mixing and low modulus."_  
+> 📚 _"What do recent papers say about yield strength trends in BCC HEAs?"_
+
+---
+
+## 🚀 Features
+
+- 🔍 **Semantic paper search** using FAISS + BAAI embeddings  
+- 📊 **Smart dataset filtering** using canonical and synonymous property names  
+- 🤖 **LLM-based reasoning** via Mistral-7B  
+- 🧾 **Unified prompt generation** from papers + structured data  
+- 🖥️ **Gradio app** with answer + tables + raw paper context
+
+---
+
+## 📂 Resources Used
+
+### 📚 Resource 1: Literature Corpus
+
+- ~1800 open-access PDFs on HEAs
+- Extracted sections: `abstract`, `introduction`, `methods`, `conclusion`
+- Chunked with LangChain `RecursiveCharacterTextSplitter`
+- Embedded with `BAAI/bge-base-en`
+- Indexed using `FAISS`
+
+### 📊 Resource 2: Structured Datasets
+
+| Dataset   | Description                                | Format              |
+|-----------|--------------------------------------------|---------------------|
+| MPEA      | Experimental data (HV, UTS, density, etc.) | `dataset1_clean.csv` |
+| ML Pred   | Design parameters + predictions            | `dataset2_clean.csv` |
+| Achief    | Thermodynamic + phase data                 | `dataset3_clean.csv` |
+
+➡️ Each dataset was cleaned, normalized, and reformatted for query integration.
+
+---
+
+## 🧠 Model Details
+
+- **LLM**: `Mistral-7B-Instruct v0.3`
+- **Embeddings**: `BAAI/bge-base-en`
+- **Frameworks**: LangChain, Transformers, FAISS, Gradio
+
 ---
-title: HEA Query
-emoji: 🏢
-colorFrom: gray
-colorTo: indigo
-sdk: gradio
-sdk_version: 5.45.0
-app_file: app.py
-pinned: false
-license: mit
-short_description: LLM-Powered assistant to query HEA papers and datasets
+
+## 💬 How It Works
+
+1. User submits a question.
+2. The system:
+   - Uses **FAISS** to retrieve relevant paper chunks.
+   - Filters HEA datasets based on numeric or categorical queries.
+   - Constructs a unified prompt combining both results.
+3. **Mistral LLM** generates a structured, domain-aware response.
+
+### Outputs:
+- 📄 **LLM Answer**
+- 📊 **Matching rows from datasets**
+- 📚 **Raw context from papers**
+
+---
+
+## 🖥️ Gradio Demo UI
+
+| Panel         | Description                                 |
+|---------------|---------------------------------------------|
+| 🧠 LLM Answer | Natural language response from Mistral      |
+| 📊 CSV Matches | Tabular results matching query filters      |
+| 📚 FAISS Context | Raw text from relevant research papers      |
+
+---
+
+## 🧑‍💻 Team
+
+- **Taradutt Pattnaik** – [University of Connecticut, Storrs]  
+- **Sanjeev K Nayak** – [University of Connecticut, Storrs]  
+- **Alexander Horvath** – [University of Connecticut, Storrs]  
+
+---
+
+## 📚 Dataset References _(to be added)_
+
+- [1] MPEA Dataset – *C. Borg, “Expanded dataset of mechanical properties and observed phases of multi-principal element alloys”. figshare, 12-Jul-2020, doi: 10.6084/m9.figshare.12642953.v9* 
+- [2] ML Pred Dataset – *R. Machaka, G. T. Motsi, L. M. Raganya, P. M. Radingoana, and S. Chikosha, “Machine learning-based prediction of phases in high-entropy alloys: A data article,” Data in Brief, vol. 38, p. 107346, Oct. 2021, doi: https://doi.org/10.1016/j.dib.2021.107346.*
+- [3] Achief Dataset – *C. E. Precker, A. Gregores Cotoand S. Muíños Landín, “Materials for Design Open Repository. High Entropy Alloys”. Zenodo, Aug. 03, 2021. doi: 10.5281/zenodo.5155150.*  
+
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference