Generate_features_pred room for optimization?

[rvarun7777]

In the Generate_features_pred function as pointed below:

FeatDepth/mono/model/mono_fm/net.py

Line 80 in 3b5bf79

outputs = self.generate_features_pred(inputs, outputs)

I see that you are using only the scale 0 features. But the calculation takes place for all the 4 scales.
I believe, it can be calculated only once.

FeatDepth/mono/model/mono_fm/net.py

Line 168 in 3b5bf79

disp = outputs[("disp", 0, 0)]

FeatDepth/mono/model/mono_fm/net.py

Line 184 in 3b5bf79

outputs[("feature", frame_id, 0)] = F.grid_sample(src_f, pix_coords, padding_mode="border")

[sconlyshootery]

Yes, you're right, thank you for your advice.