tum-ai-academy

Deploy AI Models - Making AI models accessible to users.

How-to

To get started with making ML applications reachable to users:

Step 1: Clone this repo

1. Click <>Code
2. Copy the link from HTTPS
3. Open your IDE
4. Start a new repo from VCS
5. Paste the link copied in Step 2

Step 2: Install Requirements

For PyCharm/IntelliJ users:

1. On the lower left-hand side, click Python Packages
2. Search for fastapi
3. On the right-hand side, click Install package
4. Repeat Steps 2 and 3 for uvicorn, torch and transformers

For VSCode users:

1. Go to Python on the left-hand side
2. At Workspace Environment, go to Packages
3. Click on the search icon
4. Search for fastapi and click on the first item
5. Repeat Step 4 for uvicorn, torch and transformers

For terminal users:

pip install -r requirements.txt

Step 3: Run hello-world

1. Open ex1.py
2. Right-click and go to Run ex1.py
3. Follow this link: http://localhost:8000/

Step 4: Run Classification Model

1. Open ex2.py
2. Right-click and go to Run ex2.py
3. Follow this link: http://localhost:8000/analysis?query=I am very happy with the purchase

Step 5: Run Generative Model

1. Open ex3.py
2. Right-click and go to Run ex3.py
3. Follow this link: http://localhost:8000/generate?prompt=translate English to German That is good

Step 6: Clean Up

1. Open cleanup.py
2. Right-click and go to Run cleanup.py

Made with ❤️ by datamax.ai.