https://github.com/datastaxdevs/workshop-ai-as-api
https://github.com/datastaxdevs/workshop-ai-as-api
astradb fastapi pandas python
Last synced: 6 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/datastaxdevs/workshop-ai-as-api
- Owner: datastaxdevs
- Created: 2022-02-08T18:05:16.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-04-05T16:27:41.000Z (over 2 years ago)
- Last Synced: 2024-11-27T03:34:29.864Z (11 months ago)
- Topics: astradb, fastapi, pandas, python
- Language: Python
- Homepage:
- Size: 12.5 MB
- Stars: 133
- Watchers: 3
- Forks: 55
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Workshop: AI as an API (W026)
**The full course, including hands-on instructions, is on [DataStax Academy](https://datastax.academy/course/view.php?id=10). There,
you will learn how to work in the Interactive Lab**.
_For best results: (1) Check the course [start page](https://datastax.academy/course/view.php?id=10)._
_(2) Enroll in the course. (3) Start **learning** and **practicing**!_
### Learn to build your own NLP text classifier and expose it as an API using the following technologies:
- AI-based text analysis with Tensorflow/Keras;
- Astra DB, a Database-as-a-Service built on Apache Cassandra®;
- FastAPI, the high-performance Python framework for creating APIs;
- Many useful Python libraries and packages including `pandas`, `pydantic`, `dotenv`, `sklearn`, `uvicorn`, and more.
### During this hands-on workshop, you will:
- prepare the labeled dataset for model training;
- train the model to classify any input text;
- export the trained model and test it interactively;
- create your free NoSQL database for data storage;
- set up and start an API exposing the classifier as a reusable class;
- learn how to speed up the API with DB-based caching;
- inspect how a streaming response is handled in the API.
#### Prerequisites:
- Familiarity with Python;
- A Github account is required for the hands-on.