Backend for PetVac. Written in Python using Flask.

Please install these libraries first through pip:

pip install flask pandas scikit-learn ollama pydantic flask_cors

Add pickle to pip if needed.

Then, download Ollama from its website here. After installation, pull an LLM model through cmd (this project uses llama3.2:3b-instruct-q6_K) by:

ollama pull <model_name>

Run the server via:

python main.py

Note that this has no UI defined at /.

There are two routes:

POST /llm

• Returns a JSON output containing the pets and their allergies.
• Requires an input.

POST /suggest

• Returns the url, product_name, and the image_url of the suggested products.
• Needs num (number of suggestions) and the list of dogs similar in format from /llm.