This project scrapes content from fashion, Bollywood, Hollywood, and entertainment websites, and uses a local LLM via Ollama to generate clear, concise, and engaging summaries.
- 🔍 Scrapes web content using
requestsandBeautifulSoup - 🧼 Cleans up irrelevant HTML elements like scripts, images, and styles
- 🧠 Sends cleaned text to a locally running LLM model (via Ollama)
- 📄 Generates summaries in plain text or Markdown format
- 🧑🎤 Focused on:
- Fashion trends
- Celebrity style
- TV and movie gossip
- Pop culture updates
| Tool | Purpose |
|---|---|
| 🐍 Python | Core programming language |
| 🧼 BeautifulSoup | Web scraping and HTML parsing |
| 🧠 OpenAI (with Ollama backend) | Language model interaction |
| 💻 Jupyter Notebook | Interactive development and testing |
| 🌐 Ollama | Local LLM hosting (like llama3) |
- Python 3.8+
- Ollama installed & running locally
- Model pulled (e.g.
llama3.2): - ollama run llama3.2
pip install openai requests beautifulsoup4 lxml python-dotenv
### ▶️ Run the Notebook
1. Start Ollama in terminal:
```bash
ollama run llama3-
Launch Jupyter Notebook:
jupyter notebook
-
Open the
.ipynbnotebook and run each cell in order.
🔗 https://www.pinkvilla.com
📄 Summary:
- Kareena Kapoor stuns in a denim jacket at the airport.
- Vogue India highlights trending 90s throwback looks.
- Alia Bhatt attends a wedding in elegant pastels.
📦 fashion-summarizer/
├── Untitled.ipynb # Main notebook with scraping + summarization
├── README.md # Project documentation (this file)
Vanshika Mahajan
Passionate about Generative AI, fashion-tech, and practical NLP apps.
(Add your LinkedIn, GitHub, or portfolio link here if you want)
This project is open source and available under the MIT License.