This repository contains the TensorFlow models and its associated data.
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
Please make sure to update tests as appropriate.
Contains the existing sample datasets in csv format and key embeddings of the data.
The model used for the recommendation system is the ScaNN implemented through Tensorflow to allow serving with Tensorflow Serving service. The model creation process each time requires different steps:
- Preprocessing
- Simple text preprocessing
- Keyword extraction
- Embedding
- Training of the ScaNN
- Testing
- Saving and serving
Sample demo application that narrows down the job-search based on location, reputation, etc through a flask api. For more, see the README.md in the subfolder.
The model is deployed to Tensorflow Serving. Its design is outlined in the image below.
TODO
Access on the Colab notebook to test the trained model interactivelly.