Starter Template

---

For Truefoundry MLE internship assignment

Info: https://docs.google.com/document/d/1rPyN-IwjRXoxTkHP8xiZ3gByGiYn96eorZfYnyUPvL0/edit

1. You can run the fastapi with uvicorn as uvicorn main:app --host 0.0.0.0 --port 8000
2. Then you can run the test runner as python test_all.py http://0.0.0.0:8000/predict - Note the TASK_TO_URL contains urls to models hosted internally by Truefoundry

Finally you can deploy your fastapi app on Truefoundry and re run the test runner as python test_all.py <YOUR URL>

Note the object detection model might take 15-30s to respond