Doing saliency maps for a regression problem

[amunozj]

Hi! first of all I wanted to thank you for putting this together it's pretty awesome. Is there a quick hack I could do use your saliency visualizations for a regression problem. That is, instead of providing and index you provide a value.

I tried doing something like this:

heatmap = visualize_saliency(model, layer_idx, filter_indices=0.00054109, seed_input=dfnp)

But I get an error:

Value passed to parameter 'begin' has DataType float32 not in list of allowed values: int32, int64

thank you!

[raghakot]

See the self-driving car example in the applications folder. It shows how to handle regression use-cases