Develop a hyper parameter optimization method for TF models
Budget: $250 – $750 USD
Hi there!
I’m looking for a Data Scientist/Machine Learning Engineer with python proficiency. We need to build and test a general hyper parameter optimization method for TF models classification/regression models. The method need to run several training attempts (but not all the possible combinations) over a universe of hyper parameters like:
• Number of layers of the NN
• Learning rate
• Number of dropout layers
• Activation functions
• Neurons per layer
• Others
We expect a method that can optimize the runs in order to decrease the number of trainings.
For this project the following tasks are required:
1. Design the method (input, output parameters).
2. Implement the method using python and tensorlfow.
3. Test the method over two well-known datasets (i.e. petale / house prices / etc).
We expect a python script with the all the code and a brief explanation about the code.
Feel free to ask for more information.
I’m looking for a Data Scientist/Machine Learning Engineer with python proficiency. We need to build and test a general hyper parameter optimization method for TF models classification/regression models. The method need to run several training attempts (but not all the possible combinations) over a universe of hyper parameters like:
• Number of layers of the NN
• Learning rate
• Number of dropout layers
• Activation functions
• Neurons per layer
• Others
We expect a method that can optimize the runs in order to decrease the number of trainings.
For this project the following tasks are required:
1. Design the method (input, output parameters).
2. Implement the method using python and tensorlfow.
3. Test the method over two well-known datasets (i.e. petale / house prices / etc).
We expect a python script with the all the code and a brief explanation about the code.
Feel free to ask for more information.