search

zllrunning / SiameseX.PyTorch

A simplified PyTorch implementation of Siamese networks for tracking: SiamFC, SiamRPN, SiamRPN++, SiamVGG, SiamDW, SiamRPN-VGG.
437 stars 94 forks source link

Problem with pretrained weights #10

Open lemmersj opened 4 years ago

lemmersj commented 4 years ago

I've recently been attempting to run the demo_rpn.py script with the pretrained weights you have made available. Unfortunately, it appears that the weights don't match the model.

Placing a breakpoint at line 14 of demo_rpn_utils/utils.py and examining the model/loaded weights, I notice a disparity in the number of parameters:

For SiamRPNPPRes50 Weights in loaded dict: 458 Weights in state dict: 384

For SiamRPNResNeXt22 Weights in loaded dict: 180 Weights in state dict: 151

Any thoughts?

ljh9999 commented 4 years ago

I encountered a problem similar to yours. It may be a problem with the pre-trained model provided by the author. Have you solved it?

lemmersj commented 4 years ago

I haven't. I ended up using DaSiamRPN, which was pretty much plug-and-play for me.