def write_solver_file(solver_file, train_model, test_models, type, base_lr,
momentum, weight_decay, lr_policy, gamma, power,
random_seed, max_iter, clip_gradients, snapshot_prefix):
'''Writes a solver prototxt file with parameters set to the
corresponding argument values. In particular, the train_net
parameter is set to train_model, and a test_net parameter is
added for each of test_models, which should be a list.'''
param = SolverParameter()
param.train_net = train_model
for test_model in test_models:
param.test_net.append(test_model)
param.test_iter.append(0) #don't test automatically
param.test_interval = max_iter
param.type = type
param.base_lr = base_lr
param.momentum = momentum
param.weight_decay = weight_decay
param.lr_policy = lr_policy
param.gamma = gamma
param.power = power
param.display = 0 #don't print solver iterations
param.random_seed = random_seed
param.max_iter = max_iter
if clip_gradients > 0:
param.clip_gradients = clip_gradients
param.snapshot_prefix = snapshot_prefix
print "WRITING", solver_file
with open(solver_file, 'w') as f:
f.write(str(param))
def get_solver(args):
from caffe.proto.caffe_pb2 import SolverParameter
from google.protobuf.text_format import Merge
solver = SolverParameter()
Merge(open(args.solver, 'r').read(), solver)
return solver
def lenet_solver():
"""A simple version of LeNet's solver proto"""
parser = argparse.ArgumentParser()
parser.add_argument(
'--train_net',
default='../../Lenet/lenet_auto_train.prototxt',
help=
'path to train net prototxt. [DEFAULT=../../Section4/caffenet_train.prototxt]'
)
parser.add_argument(
'--test_net',
default='../../Lenet/lenet_auto_test.prototxt',
help=
'path to validation net prototxt. [DEFAULT=../../Section4/caffenet_valid.prototxt]'
)
parser.add_argument('--solver_target_folder',
default='../../Lenet/',
help='solver target FOLDER. [DEFAULT=../../Section5/]')
parser.add_argument(
'--solver_filename',
default='Lenet_solver.prototxt',
help='solver prototxt NAME. [DEFAULT=caffenet_solver.prototxt]')
parser.add_argument(
'--snapshot_target_folder',
default='../../Lenet/',
help='snapshot target FOLDER. [DEFAULT=../../Section6/')
parser.add_argument('--snapshot_prefix',
default='Lenet',
help='snapshot NAME prefix, [DEFAULT=caffenet]')
args = parser.parse_args()
SOLVER_FULL_PATH = args.solver_target_folder + args.solver_filename
SNAPSHOT_FULL_PATH = args.snapshot_target_folder + args.snapshot_prefix
os.system('rm -rf ' + SOLVER_FULL_PATH)
os.system('rm -rf ' + SNAPSHOT_FULL_PATH + '*')
solver = SolverParameter()
solver.train_net = 'lenet_auto_train.prototxt'
solver.test_net.append('lenet_auto_test.prototxt')
solver.test_iter.append(100)
solver.test_interval = 500
solver.base_lr = 0.01
solver.momentum = 0.9
solver.weight_decay = 0.0005
solver.lr_policy = 'inv'
solver.gamma = 0.0001
solver.power = 0.75
# solver.stepsize = 2500
solver.display = 100
solver.max_iter = 10000
solver.snapshot = 5000
solver.snapshot_prefix = SNAPSHOT_FULL_PATH
solver.solver_mode = SolverParameter.GPU
with open(args.solver_filename, 'w') as f: # generating prototxt
f.write(str(solver))
def main():
# Use first line of file docstring as description if a file docstring exists.
parser = argparse.ArgumentParser(
description=__doc__.split('\n')[0] if __doc__ else '',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('--bitmask', type=str, required=True)
parser.add_argument('--output', type=str, required=True)
parser.add_argument('--snapshot_prefix',
default='{output}/models/alexnet_{bitmask}')
parser.add_argument('--solver',
default='models/bvlc_alexnet/solver.prototxt')
parser.add_argument('--trainval',
default='models/bvlc_alexnet/train_val.prototxt')
# Do we need the deploy?
parser.add_argument('--deploy',
default='models/bvlc_alexnet/deploy.prototxt')
args = parser.parse_args()
args.snapshot_prefix = args.snapshot_prefix.format(output=args.output,
bitmask=args.bitmask)
bitmask = args.bitmask
assert len(bitmask) == len(LAYERS), (
'Expected {} bits in bitmask, received {}.'.format(
len(LAYERS), len(bitmask)))
try:
int(bitmask)
except ValueError:
raise Exception('Invalid bitmask: {}'.format(bitmask))
zeroed_layers = set(layer for i, layer in enumerate(LAYERS)
if bitmask[i] == '0')
# Parse solver
solver = SolverParameter()
parse_prototxt(solver, args.solver)
solver.snapshot_prefix = args.snapshot_prefix
solver.net = '{}/train_val.prototxt'.format(args.output)
# Parse trainval NetParameter
trainval = NetParameter()
parse_prototxt(trainval, args.trainval)
for layer in trainval.layer:
if layer.name in zeroed_layers:
print('Zeroing {}'.format(layer.name))
for param in layer.param:
param.lr_mult = 0
param.decay_mult = 0
mkdir_p(args.output)
write_prototxt(trainval, '{}/train_val.prototxt'.format(args.output))
write_prototxt(solver, '{}/solver.prototxt'.format(args.output))
copyfile(args.deploy, '{}/deploy.prototxt'.format(args.output))
def init_solver(self):
""" Helper method to initialize the solver. """
solver_param = SolverParameter()
solver_param.snapshot_prefix = self._hyperparams['weights_file_prefix']
solver_param.display = 0 # Don't display anything.
solver_param.base_lr = self._hyperparams['lr']
solver_param.lr_policy = self._hyperparams['lr_policy']
solver_param.momentum = self._hyperparams['momentum']
solver_param.weight_decay = self._hyperparams['weight_decay']
solver_param.type = self._hyperparams['solver_type']
solver_param.random_seed = self._hyperparams['random_seed']
# Pass in net parameter either by filename or protostring.
if isinstance(self._hyperparams['network_model'], basestring):
self.solver = caffe.get_solver(self._hyperparams['network_model'])
else:
network_arch_params = self._hyperparams['network_arch_params']
network_arch_params['dim_input'] = self._dO
network_arch_params['dim_output'] = self._dU
network_arch_params['batch_size'] = self.batch_size
network_arch_params['phase'] = TRAIN
solver_param.train_net_param.CopyFrom(
self._hyperparams['network_model'](**network_arch_params)
)
# For running forward in python.
network_arch_params['batch_size'] = 1
network_arch_params['phase'] = TEST
solver_param.test_net_param.add().CopyFrom(
self._hyperparams['network_model'](**network_arch_params)
)
# For running forward on the robot.
network_arch_params['batch_size'] = 1
network_arch_params['phase'] = 'deploy'
solver_param.test_net_param.add().CopyFrom(
self._hyperparams['network_model'](**network_arch_params)
)
# These are required by Caffe to be set, but not used.
solver_param.test_iter.append(1)
solver_param.test_iter.append(1)
solver_param.test_interval = 1000000
f = tempfile.NamedTemporaryFile(mode='w+', delete=False)
f.write(MessageToString(solver_param))
f.close()
self.solver = caffe.get_solver(f.name)
def _init_solver(self): """ Helper method to initialize the solver. """ solver_param = SolverParameter() solver_param.display = 0 # Don't display anything. solver_param.base_lr = self._hyperparams['lr'] solver_param.lr_policy = self._hyperparams['lr_policy'] solver_param.momentum = self._hyperparams['momentum'] solver_param.weight_decay = self._hyperparams['weight_decay'] solver_param.type = self._hyperparams['solver_type'] solver_param.random_seed = self._hyperparams['random_seed'] # Pass in net parameter by protostring (could add option to input prototxt file). network_arch_params = self._hyperparams['network_arch_params'] network_arch_params['dim_input'] = self._dO network_arch_params['demo_batch_size'] = self._hyperparams['demo_batch_size'] network_arch_params['sample_batch_size'] = self._hyperparams['sample_batch_size'] network_arch_params['T'] = self._T network_arch_params['phase'] = TRAIN solver_param.train_net_param.CopyFrom( self._hyperparams['network_model'](**network_arch_params) ) # For running forward in python. network_arch_params['phase'] = TEST solver_param.test_net_param.add().CopyFrom( self._hyperparams['network_model'](**network_arch_params) ) network_arch_params['phase'] = 'forward_feat' solver_param.test_net_param.add().CopyFrom( self._hyperparams['network_model'](**network_arch_params) ) # These are required by Caffe to be set, but not used. solver_param.test_iter.append(1) solver_param.test_iter.append(1) solver_param.test_interval = 1000000 f = tempfile.NamedTemporaryFile(mode='w+', delete=False) f.write(MessageToString(solver_param)) f.close() self.solver = caffe.get_solver(f.name)
def lenet_solver_simple():
"""A simple version of LeNet's solver proto"""
solver = SolverParameter()
solver.train_net = 'lenet_auto_train.prototxt'
solver.test_net.append('lenet_auto_test.prototxt')
solver.test_iter.append(100)
solver.test_interval = 500
solver.base_lr = 0.01
solver.momentum = 0.9
solver.weight_decay = 0.0005
solver.lr_policy = 'inv'
solver.gamma = 0.0001
solver.power = 0.75
# solver.stepsize = 2500
solver.display = 100
solver.max_iter = 10000
solver.snapshot = 5000
solver.snapshot_prefix = 'SNAPSHOT_FULL_PATH'
solver.solver_mode = SolverParameter.GPU
with open('SOLVER_FULL_PATH', 'w') as f: # generating prototxt
f.write(str(solver))