Expose hyper-parameter settings for deep-learning scenarios in cli

[LittleLittleCloud]

Text-Classification

• epoch
• batchsize
• device

Sentence-Similarity

• epoch
• batchsize
• device

Image Classification

• Arch
• batchsize
• epoch
• device

[luisquintanilla]

Is Arch in image classification which base model to use?

[LittleLittleCloud]

@luisquintanilla Yep, will be one of resnet101, resnet50, resnet18, alexnet

[zewditu]

@LittleLittleCloud could you have some explanation of what value will be added by exposing hyper-parameter settings to the user?

[luisquintanilla]

@LittleLittleCloud are we okay to close this one? Did this make it into the latest release?