Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
OpenFace Pre-Trained Model Incorrect DepthConcat #351
Context of the issue.
The OpenFace pre-trained model nn4.small2.v1.t7 on your website may be trained incorrectly, or be outputting results that were not intended.
Inception 3a (and other Inception layers) should be outputting a tensor with W = 12 and H = 12, as outlined in your paper.
Printing out the output tensor size of Inception(3a) after a forward pass of nn4.small2.v1.t7:
Where Sequential modules between the equal signs are concatenated by Inception(3a). We see that nn.SpatialMaxPooling has an output size of (1 x 32 x 5 x 5). Going into the Torch documentation and viewing the raw output, the DepthConcat layer used to make the Inception module actually adds padding to nn.SpatialMaxPooling to make it's dimensions W x H = 12 x 12. Here is the sample of the output I saw, from a forward pass of random images:
Steps to reproduce.
Script used to print out results: