Introduction to Experiment Hyperparams
As part of monitoring and managing your experiments, you might need to provide hyperparams such as dropout rate, whether a CPU or GPU is used, batch size, and so on, to aid in hyperparam tuning.
The following image shows the panes below the Overview tab of an experiment that uses hyperparameters.
Each framework has a list of parameters that are automatically fetched from the model with or without using this function to set parameters. Study the relevant topic for each framework to see the list of these parameters.
If you have performed basic integration with the MissingLink SDK, learn how to set hyperparams for the supported frameworks: