config.prototxt

# directory where saving the output files
workspace: "deephi/miniVggNet/pruning"

gpu: "0"

# the number of iterations to run in test phase: Max num =validation set / batch_size = 9000 / 50
test_iter: 180

# the accuracy name of the model output you care about
acc_name: "top-1"

# model for the training
model:   "deephi/miniVggNet/pruning/train_val.prototxt"

## first run
weights: "deephi/miniVggNet/pruning/float.caffemodel"

## second run
#weights: "deephi/miniVggNet/pruning/regular_rate_0.1/snapshots/_iter_40000.caffemodel"
## third run
#weights: "deephi/miniVggNet/pruning/regular_rate_0.2/snapshots/_iter_40000.caffemodel"
## 4-th run
#weights: "deephi/miniVggNet/pruning/regular_rate_0.3/snapshots/_iter_40000.caffemodel"
## 5-th run
#weights: "deephi/miniVggNet/pruning/regular_rate_0.4/snapshots/_iter_40000.caffemodel"
## 6-th run
#weights: "deephi/miniVggNet/pruning/regular_rate_0.5/snapshots/_iter_40000.caffemodel"
## 7-th run
#weights: "deephi/miniVggNet/pruning/regular_rate_0.6/snapshots/_iter_40000.caffemodel"

# solver used for fine tuning
solver:  "deephi/miniVggNet/pruning/solver.prototxt"

# expected compression ratio: 0.1 means 10%
rate: 0.1  #first run
#rate: 0.2 #second run
#rate: 0.3 #third run
#rate: 0.4 # 4-th run
#rate: 0.5 # 5-th run
#rate: 0.6 # 6-th run
#rate: 0.7 # 7-th run

#compression method used (only REGULAR is available)
pruner { method: REGULAR }