Convert a deep learning code written in TensorFlow 1 to TensorFlow 2 -- 2
Budget: $750 – $1,500 AUD
There is a code written in TensorFlow 1 for developing a deep learning model. I want to convert the code to TensorFlow 2. The new code should:
1. Train the model as fast as the original code (This is the crucial condition about this project)
2. Generate exactly the same quality of results as the original code (with the same dataset that is used for the original code)
3. Be like it is "genuinely" written in TensorFlow 2 and not only TensorFlow 2 compatible. placeholders and sessions should be removed.
This is the original code:
https://github.com/sanderlab/CellBox/tree/master/cellbox/cellbox
At the end of the project I will assess the speed of the new code, the performance of the model training process and output of the code.
Thanks!
1. Train the model as fast as the original code (This is the crucial condition about this project)
2. Generate exactly the same quality of results as the original code (with the same dataset that is used for the original code)
3. Be like it is "genuinely" written in TensorFlow 2 and not only TensorFlow 2 compatible. placeholders and sessions should be removed.
This is the original code:
https://github.com/sanderlab/CellBox/tree/master/cellbox/cellbox
At the end of the project I will assess the speed of the new code, the performance of the model training process and output of the code.
Thanks!