def startWorkerNode(self):
'''Receives updated parameters from the master, keeps updating local parameters using its own subset of training data, pushes local parameters to
master, all asynchronously '''
self.initMVRNN() #also initializes training and test sets
self.func = costFn
self.fprime = None
self.optimizer = StochasticGradientDescent(niter=self.nIter , learning_rate=0.01, learningrateFactor=1.0, printAt10Iter='.', printAt100Iter='\n+')
genObj = None
while True:
#fetch action and parameters from masters
action, master_rnn = self.pop('fromMaster_' + str(self.index))
if(action == "finish"):
print "Node: ",self.index," Received finish action from master, exiting..."
return
if self.debug:
print "Node: ", self.index, " Fetched new action & parameters from Master.. " + str(master_rnn.W[0,0])
self.params.resetNumReducedWords() #to unroll all the words
# W, WO, Wcat, Wv, Wo = unroll_theta(theta, self.params)
self.copy_into_rnn(master_rnn.getParamsList()) #copy to local rnn to global
theta = self.rnn.getTheta(self.params, self.all_train_idx)
self.params.setNumReducedWords(len(self.all_train_idx)) #set number of reduced words, to be used by costfn for unrolling
start_time = time.clock()
try:
genObj.next()
except (AttributeError, StopIteration) :
genObj = self.optimizer.minimizeBatchesPll(rnn=self.rnn, rnnData_train=self.rnnData_train, allTrainSentIdx=self.all_train_idx,
params=self.params, x0=theta, func=costFn, fprime=None,
rnnData_test=self.rnnData_dev, initialSetSize=1, niter=1, seed=self.seed,
modelFileName='', printStatistics=False, modelSaveIter=10, nIterInPart=1,
nFetch=self.nFetch, rnd=self.rnd, nodeid=self.index) #optimize this theta and save it in self.rnn
except:
raise
end_time = time.clock()
# [W, WO, Wcat, Wv, Wo] = self.rnn
# theta = np.concatenate((W.flatten(), WO.flatten(), Wcat.flatten(), Wv.flatten(), Wo.flatten())) # current local optimal theta
#push local theta to master
self.push('toMaster_' + str(self.index), (self.rnn, None)) #push the theta value #pushing a tuple object #playdoh bug
if self.debug:
print "Node:", self.index, " Execution time for ", self.nFetch, " minibatches: ", (end_time - start_time)/60, 'minutes'
print "Node:", self.index, " Pushed local parameters to Master.."
def train():
SEED = 13742
load_model = False
custom_load = False
np.random.seed(SEED)
#get sentences, trees and labels
nExamples = 5
print "loading data.."
rnnData_train = RNNDataCorpus()
rnnData_train.load_data_srl(load_file=config.train_data_srl,
nExamples=nExamples)
rnnData_dev = RNNDataCorpus()
rnnData_dev.load_data_srl(load_file=config.dev_data_srl,
nExamples=nExamples)
print "Number of sentences loaded in training data: ", rnnData_train.ndoc()
#initialize params
print "initializing params"
params = Params(data=rnnData_train, wordSize=52, rankWo=2)
n = params.wordSize
fanIn = params.fanIn
nWords = params.nWords
nLabels = params.categories
rank = params.rankWo
if (load_model):
modelfile = config.saved_params_file + 'SGD_SLL300'
rnn = MVRNNSLL.load(modelfile)
print 'loaded model : ', modelfile
elif (custom_load):
modelfile = config.saved_params_file + 'SGD_SLL300'
print "loading customized model..", modelfile
# d = 2#extra features for wordvectors
# Wo = 0.01*np.random.randn(n + 2*n*rank, nWords) #Lm, as in paper
# Wo[:n,:] = np.ones((n,Wo.shape[1])) #Lm, as in paper
# Wcat = 0.005*np.random.randn(nLabels, fanIn) #Wlabel, as in paper
# Wv = 0.01*np.random.randn(n, nWords)
# WO = 0.01*np.random.randn(n, 2*n)
# W = 0.01*np.random.randn(n, 2*n+1)
#load pre-trained weights here
oldrnn = MVRNNSLL.load(modelfile)
# Wv[:-d,:] = oldrnn.Wv
categories = [x.strip() for x in rnnData_train.categories]
Tran = init_transitions(dict(zip(categories, range(len(categories)))),
'iob')
rnn = MVRNNSLL(oldrnn.W, oldrnn.WO, oldrnn.Wcat, oldrnn.Wv, oldrnn.Wo,
Tran)
else:
#define theta
#one vector for all the parameters of mvrnn model: W, Wm, Wlabel, L, Lm
# n = params.wordSize; fanIn = params.fanIn; nWords = params.nWords; nLabels = params.categories; rank=params.rankWo
# Wo = 0.01*np.random.randn(n + 2*n*rank, nWords) #Lm, as in paper
# Wo[:n,:] = np.ones((n,Wo.shape[1])) #Lm, as in paper
# Wcat = 0.005*np.random.randn(nLabels, fanIn) #Wlabel, as in paper
# #load pre-trained weights here
## mats = sio.loadmat(config.saved_params_file)
# oldrnn = MVRNNSLL.load(modelfile)
# Wv = oldrnn.Wv #L, as in paper
# W = oldrnn..get('W') #W, as in paper
# WO = mats.get('WO') #Wm, as in paper
Wo = 0.01 * np.random.randn(n + 2 * n * rank, nWords) #Lm, as in paper
Wo[:n, :] = np.ones((n, Wo.shape[1])) #Lm, as in paper
Wcat = 0.005 * np.random.randn(nLabels, fanIn) #Wlabel, as in paper
Wv = 0.01 * np.random.randn(n, nWords)
WO = 0.01 * np.random.randn(n, 2 * n)
W = 0.01 * np.random.randn(n, 2 * n + 1)
categories = [x.strip() for x in rnnData_train.categories]
Tran = init_transitions(dict(zip(categories, range(len(categories)))),
'iob')
rnn = MVRNNSLL(W, WO, Wcat, Wv, Wo, Tran)
# rnn = MVRNNSLL(W, WO, Wcat, Wv, Wo, Tran)
[_, _, all_train_idx
] = getRelevantWords(rnnData_train, rnn.Wv, rnn.Wo,
params) #sets nWords_reduced, returns new arrays
params.setNumReducedWords(len(all_train_idx))
theta = rnn.getTheta(params, all_train_idx)
#optimize
print "starting training using SGD..."
nIter = 500
optimizer = StochasticGradientDescent(niter=nIter,
learning_rate=0.01,
learningrateFactor=1.0,
printAt10Iter='.',
printAt100Iter='\n+')
optimizer.minimizeBatches(rnn=rnn,
rnnData_train=rnnData_train,
allTrainSentIdx=all_train_idx,
params=params,
x0=theta,
func=costFn,
fprime=None,
rnnData_test=rnnData_dev,
initialSetSize=1,
niter=nIter,
seed=SEED,
modelFileName=config.saved_params_file +
'SGD_SLL',
printStatistics=True,
modelSaveIter=100,
nIterInPart=1,
nFetch=-1,
rnd=None,
nodeid=-1)
def train():
SEED = 131742
load_model = False
custom_load = True #loads model from previously saved model except Wcat
np.random.seed(SEED)
#get sentences, trees and labels
nExamples = 5
print "loading data.."
rnnData_train = RNNDataCorpus()
rnnData_train.load_data_srl(load_file=config.train_data_srl,
nExamples=nExamples)
rnnData_dev = RNNDataCorpus()
rnnData_dev.load_data_srl(load_file=config.dev_data_srl,
nExamples=nExamples)
print "Number of sentences loaded in training data: ", rnnData_train.ndoc()
#initialize params
print "initializing params"
params = Params(data=rnnData_train, wordSize=52, rankWo=2)
n = params.wordSize
fanIn = params.fanIn
nWords = params.nWords
nLabels = params.categories
rank = params.rankWo
if (load_model):
with open(config.saved_params_file + "_45", 'r') as loadfile:
rnn = cPickle.load(loadfile)
elif (custom_load):
d = 2 #extra features for wordvectors
Wo = 0.01 * np.random.randn(n + 2 * n * rank, nWords) #Lm, as in paper
Wo[:n, :] = np.ones((n, Wo.shape[1])) #Lm, as in paper
Wcat = 0.005 * np.random.randn(nLabels, fanIn) #Wlabel, as in paper
Wv = 0.01 * np.random.randn(n, nWords)
WO = 0.01 * np.random.randn(n, 2 * n)
W = 0.01 * np.random.randn(n, 2 * n + 1)
#load pre-trained weights here
with open(config.saved_params_file + '_65', 'r') as loadfile:
oldrnn = cPickle.load(loadfile)
Wv[:-d, :] = oldrnn.Wv
# WO[:-d,:] = oldrnn.WO
rnn = MVRNN(W, WO, Wcat, Wv, Wo)
else:
Wo = 0.01 * np.random.randn(n + 2 * n * rank, nWords) #Lm, as in paper
Wo[:n, :] = np.ones((n, Wo.shape[1])) #Lm, as in paper
Wcat = 0.005 * np.random.randn(nLabels, fanIn) #Wlabel, as in paper
Wv = 0.01 * np.random.randn(n, nWords)
WO = 0.01 * np.random.randn(n, 2 * n)
W = 0.01 * np.random.randn(n, 2 * n + 1)
rnn = MVRNN(W, WO, Wcat, Wv, Wo)
[_, _, all_train_idx
] = getRelevantWords(rnnData_train, rnn.Wv, rnn.Wo,
params) #sets nWords_reduced, returns new arrays
params.setNumReducedWords(len(all_train_idx))
theta = rnn.getTheta(params, all_train_idx)
#optimize
print "starting training using SGD..."
nIter = 500
optimizer = StochasticGradientDescent(niter=nIter,
learning_rate=0.01,
learningrateFactor=1.0,
printAt10Iter='.',
printAt100Iter='\n+')
optimizer.minimizeBatches(rnn=rnn,
rnnData_train=rnnData_train,
allTrainSentIdx=all_train_idx,
params=params,
x0=theta,
func=costFn,
fprime=None,
rnnData_test=rnnData_dev,
initialSetSize=1,
niter=nIter,
seed=17,
modelFileName=config.saved_params_file +
'SGD_SRL',
printStatistics=True,
modelSaveIter=1,
nIterInPart=1,
nFetch=-1,
rnd=None,
nodeid=-1)
print "Finished training! "
