SqueezeNet out of memory with batch size (512) smaller than AlexNet (1024) about squeezenet HOT 11 CLOSED

You can compare the networks using Netscope here:
AlexNet, SqueezeNet v1.0, SqueezeNet v1.1

This table gives a summary:

Problem 2) is definitely because SqueezeNet uses more activation memory than AlexNet (5x as much in v1.0 and 3x as much in v1.1). The largest output feature map in SqNet v1.1 is 817k pixels, in AlexNet 290k pixels.

Problem 1) could have different reasons:

• SqueezeNet v1.0 has approximately the same computational complexity as AlexNet (860 million vs. 1140 million multiply-accumulate operations), SqueezeNet v1.1 should be less complex (390 million MACCs).
• SqueezeNet has many more layers than AlexNet (26 CONV layers versus 5 CONV layers), and switching the layers takes some time on the CPU and GPU
• SqueezeNet has larger output feature maps (compare Problem 2 above). Moving this data around between GPU and Memory takes some time, too.

So I don't think you're doing anything wrong, this is probably the price you have to pay for the increased network complexity. I you haven't done already, definitely give SqueezeNet v1.1 a try!

from squeezenet.

You can use the Netscope CNN Analyzer tool: https://dgschwend.github.io/netscope/
The actual code for the calculation is here:
https://github.com/dgschwend/netscope/blob/gh-pages/src/analyzer.coffee
The formulas are pretty self-explanatory.

from squeezenet.

Maybe it is caused because of concat layers?

from squeezenet.

@dgschwend Thank u so much for so many details. The Netscope is such a cool tool! I was using v1.1. I guess the problem comes from the increased activations and layers as you pointed and the concat layers as @Grabber mentioned.

from squeezenet.

You are welcome, glad I could help! ;)

from squeezenet.

@dgschwend , is there any document about how to compute the macc, memory of a network? tks

from squeezenet.

@dgschwend Nice work on that table. Is "Table 4.1" part of a longer paper that I could read (and cite)?

from squeezenet.

@forresti Yes, I just published my Master Thesis.

The project report and all code for "ZynqNet: An FPGA-Accelerated Embedded Convolutional Neural Network" are public. The whole work is based on SqueezeNet, and the report contains a detailed analysis + comparison of prior CNN topologies.

from squeezenet.

@dgschwend hi, can you share the reference to "SqueezeNet++" in your table? couldn't find it

from squeezenet.

That's the work-in-progress codename of ZynqNet CNN from my Master Thesis (see post above). See https://github.com/dgschwend/zynqnet

from squeezenet.

Have the same issue that SqeezeNet v1.1 is slower then AlexNet, i.e. 3.91 ms vs 3.01 ms.
https://github.com/mrgloom/kaggle-dogs-vs-cats-solution

from squeezenet.