Update README.md

README.md

@@ -1,100 +1,37 @@
-# 🧠 Mori: Your Friendly Data Science Assistant
-
-**Mori** is a conversational assistant trained to answer questions about data science, AI concepts, and related topics.  
-It uses a dual-response system: one technical model and one social model, allowing it to switch between factual answers and casual interactions.
-
----
-
-# Instalación
-
-## Entorno recomendado
-
-Crea y activa un entorno virtual para aislar dependencias. Se recomienda usar la carpeta en la **raíz del proyecto** (ej. `.venv`):
-
-```bash
-# Windows (PowerShell)
-python -m venv .venv
-.\.venv\Scripts\Activate.ps1
-
-# Windows (CMD)
-python -m venv .venv
-.\.venv\Scripts\activate.bat
-
-# macOS / Linux (bash/zsh)
-python3 -m venv .venv
-source .venv/bin/activate
-```
-
-## Requisitos
-- **Python** 3.10  
-- (Opcional, GPU) **NVIDIA Driver** actualizado  
-- (Opcional) **CUDA/cuDNN** a nivel sistema **solo si los necesitas**.  
-  > Si instalas PyTorch con un wheel que ya incluye CUDA (p. ej., `cu118`), no requieres el toolkit local; basta con el **driver**:
-  - pip3 install torch --index-url https://download.pytorch.org/whl/cu118
-
-## Instalación (GPU)
-
-- python -m pip install --upgrade pip
-- pip install torch --index-url https://download.pytorch.org/whl/cuXXX
-- pip install -r requirements.txt
-
-## Instalación (CPU)
-
-- python -m pip install --upgrade pip
-- pip install torch --index-url https://download.pytorch.org/whl/cpu
-- pip install -r requirements.txt
-
-
-# Uso
-
-***Es necesario utilizar el siguiente comando:***
-
-- streamlit run app.py --server.port=8501 --server.address=0.0.0.0
-
-## 🚀 Try it out!
-
-Just type your question in the input box and let Mori guide you!  
-You can ask things like:
-- *What is overfitting?*
-- *Can you explain PCA?*
-- *Tell me a data science joke!*
-- *I'm feeling tired...*
-
----
-
-## 🛠️ How it works
-
-Mori is powered by two custom fine-tuned models based on the T5 architecture:
-- `Mori Técnico`: answers technical questions about data science and machine learning
-- `Mori Social`: replies in a friendly, supportive tone for casual conversation
-
-It uses **Streamlit** for the interface and **Hugging Face Transformers** to load and run the models.
-
----
-
-## 📁 Project structure
-
-. ├── app.py # Streamlit app ├── requirements.txt # Required packages ├── Models/ # Fine-tuned models (Técnico + Social) └── Statistics/ # Logs generated on user interaction (not persistent)
-
-
----
-
-## 📊 Logging
-
-All interactions are logged locally in the `Statistics/` folder as `.csv` and `.jsonl` files.  
-*Note: Hugging Face Spaces do not persist these files between restarts.*
-
----
-
-## 👤 Author
-
-Created by **[hazutecuhtli](https://huggingface.co/hazutecuhtli)**  
-PhD in Electronic Engineering, passionate about data, modeling, and sharing knowledge.
-
----
-
-## 💡 Future Plans
-
-- Add model explanation outputs
-- Enable user-uploaded datasets
-- Connect logs to Google Sheets for persistent stats
+---
+title: Mori – Asistente Técnico (Procesamiento de Datos)
+emoji: 🤖
+colorFrom: indigo
+colorTo: green
+sdk: streamlit
+sdk_version: "1.38.0"
+app_file: mori_tech_only.py
+pinned: false
+license: mit
+---
+
+# 🧠 Mori: Your Friendly Data Science Assistant
+
+**Mori** is a conversational assistant trained to answer questions about data science, AI concepts, and related topics.  
+It now runs **tech-only**: just the technical model + optional RAG (FAISS on HF), no social model, no classifier.
+
+---
+
+## Instalación
+
+### Entorno recomendado
+
+Crea y activa un entorno virtual para aislar dependencias:
+
+```bash
+# Windows (PowerShell)
+python -m venv .venv
+.\.venv\Scripts\Activate.ps1
+
+# Windows (CMD)
+python -m venv .venv
+.\.venv\Scripts\activate.bat
+
+# macOS / Linux (bash/zsh)
+python3 -m venv .venv
+source .venv/bin/activate