• Automation framework: CK
  
• Development repository: ck-caffe2
  
• Source: GitHub
  
• Tags: caffe2,weights,caffemodel2,bvlc,googlenet,channel-stable
  
• Support for host OS: linux, windows
  
• Support for target OS: linux, windows
  
• Version: trunk
  
• Software detection plugin: model.caffe2
  
• How to get the stable version from this portal via the CK client:
  
  pip install cbench
  cb download package:caffemodel2-bvlc-googlenet --version=1.0.0 --all
  ck install package:caffemodel2-bvlc-googlenet
  ck help package
  
• How to get the development version:
  
  pip install ck
  ck pull repo:ck-caffe2
  ck install package:caffemodel2-bvlc-googlenet
  ck help package
  
• Useful CK automation actions (CLI with the Python CK API and JSON IO):
  
  ck show env
  ck find env:{ID from above}
  ck virtual env:{ID from above}
  
• CLI and Python API: module:package
  

name: BVLC GoogleNet Model caffemodel: bvlc_googlenet.caffemodel caffemodel_url: http://dl.caffe.berkeleyvision.org/bvlc_googlenet.caffemodel license: unrestricted sha1: 405fc5acd08a3bb12de8ee5e23a96bec22f08204 caffe_commit: bc614d1bd91896e3faceaf40b23b72dab47d44f5

This model is a replication of the model described in the GoogleNet publication. We would like to thank Christian Szegedy for all his help in the replication of GoogleNet model.

Differences: - not training with the relighting data-augmentation; - not training with the scale or aspect-ratio data-augmentation; - uses "xavier" to initialize the weights instead of "gaussian"; - quick_solver.prototxt uses a different learning rate decay policy than the original solver.prototxt, that allows a much faster training (60 epochs vs 250 epochs);

The bundled model is the iteration 2,400,000 snapshot (60 epochs) using quick_solver.prototxt

This bundled model obtains a top-1 accuracy 68.7% (31.3% error) and a top-5 accuracy 88.9% (11.1% error) on the validation set, using just the center crop. (Using the average of 10 crops, (4 + 1 center) * 2 mirror, should obtain a bit higher accuracy.)

Timings for bvlc_googlenet with cuDNN using batch_size:128 on a K40c: - Average Forward pass: 562.841 ms. - Average Backward pass: 1123.84 ms. - Average Forward-Backward: 1688.8 ms.

This model was trained by Sergio Guadarrama @sguada

License

This model is released for unrestricted use.

Sample validation output

      I0712 11:53:19.627121 109762 caffe.cpp:292] loss1/loss1 = 1.86667 (* 0.3 = 0.560002 loss)
      I0712 11:53:19.627127 109762 caffe.cpp:292] loss1/top-1 = 0.55522
      I0712 11:53:19.627131 109762 caffe.cpp:292] loss1/top-5 = 0.804981
      I0712 11:53:19.627136 109762 caffe.cpp:292] loss2/loss1 = 1.50183 (* 0.3 = 0.45055 loss)
      I0712 11:53:19.627140 109762 caffe.cpp:292] loss2/top-1 = 0.629759
      I0712 11:53:19.627142 109762 caffe.cpp:292] loss2/top-5 = 0.856821
      I0712 11:53:19.627147 109762 caffe.cpp:292] loss3/loss3 = 1.25635 (* 1 = 1.25635 loss)
      I0712 11:53:19.627153 109762 caffe.cpp:292] loss3/top-1 = 0.689299
      I0712 11:53:19.627156 109762 caffe.cpp:292] loss3/top-5 = 0.891441

• 1.0.0 (2019-10-08)

• .cm/desc.json (3 bytes)
• .cm/info.json (404 bytes)
• .cm/meta.json (2050 bytes)
• deploy.prototxt (35828 bytes)
• predict_net.pb (16221 bytes)
• quick_solver.prototxt (331 bytes)
• readme.md (2255 bytes)
• solver.prototxt (344 bytes)
• train_val.prototxt (40002 bytes)