Training
Train your neural network model with IBM Spectrum Conductor Deep Learning Impact 1.2.2.
- Create a training model
Create a training model in IBM Spectrum Conductor Deep Learning Impact. IBM Spectrum Conductor Deep Learning Impact 1.2.2 supports Caffe, PyTorch and TensorFlow models. - Train a model
After you create your training model, you can start training it. Before each training run you can adjust the weight files of your model, or specify the hyperparameters. - Tune hyperparameters
Submit a hyperparameter tuning job for your training model. - Validate a trained model
After a successful training run, validate your trained model against a validation dataset. - Create an inference model
Create an inference model from the cluster management console. - Test an inference model
After creating an inference model, you can test it to see prediction results of the model.