Open kdmcdrm opened 4 years ago
Follow up on this, the difference in output values was because I forgot to put the model in evaluate mode in PyTorch (stupid!). The main issue outlined here is still present, I had to rename the output layers to allow the model to load.
Describe the bug Trying to convert a siamese Resnet18 model from Pytorch to Tensorflow but I get the following error in tf.loadModel
I tried modifying the names of these layers manually in the model.json file, which corrects this loading issue, but in that case I found the model output for an input of all ones differed between the raw and converted models. I'm not sure if this second part is related or a separate issue.
To Reproduce Here's the network, pretty simple.
Expected behavior No loading error.
Logs
Environment (please complete the following information):