Group2255/programming-help-chatbot
0
1---2title: PromptPilot (AML-3304)3emoji: ๐ค4colorFrom: blue5colorTo: indigo6sdk: streamlit7sdk_version: 1.20.08app_file: app.py9pinned: false10---11 12# ๐ค PromptPilot โ Unified AI Chatbot (AML-3304)13 14This is an MVP AI chatbot developed for the **AML-3304 assignment**:15 16**"From Tokens to Transformers: Building AI Systems with Embeddings, Generative Models, and MLOps."**17 18PromptPilot combines five intelligent chat modes powered by popular Transformer-based models, enabling users to explore code generation, factual Q&A, probabilistic reasoning, and paraphrasing.19 20---21 22## ๐ Chat Modes23 24| Mode | Description | Model Used |25|-------------------------|-------------------------------------------------------------|----------------------------------------|26| ๐งโ๐ป Code Generator | Generate Python code from plain language | `Salesforce/codegen-350M-mono` |27| ๐ General Q&A | Factual question answering and explanations | `google/flan-t5-base` |28| ๐ Bayesian Q&A | Uncertainty-based reasoning using sampling | `google/flan-t5-base` + top-k sampling |29| ๐ Deep Coder | Technical Q&A and code completion tasks | `deepseek-ai/deepseek-coder-6.7b-instruct` |30| ๐ Transformer Explorer | Paraphrasing and text transformation using sequence models | `google/flan-t5-base` |31 32---33 34## ๐ Live Demo35 36Try the chatbot here: 37**๐ [https://huggingface.co/spaces/Group2255/programming-help-chatbot](https://huggingface.co/spaces/Group2255/programming-help-chatbot)**38 39---40 41## ๐งฉ Features42 43- Multi-mode UI for interactive exploration 44- Efficient model switching using `st.session_state` 45- GPU-compatible on Hugging Face Spaces 46- Supports both deterministic and sampling-based generation 47- Lightweight models for fast response and free-tier compatibility 48 49---50 51## ๐ป Run Locally52 53### โ
Option 1: Python + Streamlit54 55```bash56git clone https://huggingface.co/spaces/Group2255/programming-help-chatbot57cd programming-help-chatbot58pip install -r requirements.txt59streamlit run app.py60 