VAL_NUM = len(VAL_ANSWER)
VAL_BATCHSIZE = max(int(VAL_NUM/100),1)
VAL_INPUT = [ VAL_INPUT[i:i+VAL_BATCHSIZE] for i in range(0,VAL_NUM,VAL_BATCHSIZE)]
VAL_ANSWER = [ VAL_ANSWER[i:i+VAL_BATCHSIZE] for ii in range(0,VAL_NUM,VAL_BATCHSIZE)]
#pdb.set_trace()
print "Start training......"
totaltime = 0
for epoch in range(MAX_EPOCH):
tStart = time.time()
cost = 0
for batch_x,batch_y in zip(BATCHED_INPUT,BATCHED_OUTPUT):
cost += nn.train(batch_x,batch_y,
LEARNING_RATE,
MOMENTUM)
tEnd = time.time()
totaltime += tEnd - tStart
if (epoch+1 != MAX_EPOCH) and ((epoch+1) % L_RATE_DECAY_STEP == 0):
print "learning rate annealed at epoch {0}".format(epoch+1)
LEARNING_RATE /= 10
if epoch+1 != MAX_EPOCH and (epoch+1) % SAVE_MODEL_EPOCH == 0:
fh = open(MODEL_ROOT+MODEL+"_at_{0}".format(epoch+1),'wb')
saved_params = (nn.layers, nn.W, nn.b)
pickle.dump(saved_params, fh)
fh.close()
#pdb.set_trace()
val_x = np.asarray(Val_Feats,dtype='float32').T
val_label = [ LABEL_DICT[ID] for ID in Val_IDs ]
# phone48to39 = load_dict_48to39()
# pdb.set_trace()
totaltime = 0
print "Start training......"
for epoch in range(100,MAX_EPOCH):
tStart = time.time()
cost = 0
for batched_inputs,batched_outputs in zip(BATCHED_INPUT,BATCHED_OUTPUT):
cost += nn.train(batched_inputs,batched_outputs,
LEARNING_RATE,
MOMENTUM,
RETAIN_PROB,
INPUT_R_PROB)
tEnd = time.time()
totaltime += tEnd - tStart
if (epoch+1 != MAX_EPOCH) and ((epoch+1) % L_RATE_DECAY_STEP == 0):
print "learning rate annealed at epoch {0}".format(epoch+1)
LEARNING_RATE /= 10
if epoch+1 != MAX_EPOCH and (epoch+1) % SAVE_MODEL_EPOCH == 0:
fh = open(MODEL_ROOT+MODEL+"_at_{0}".format(epoch+1),'wb')
saved_params = (nn.layers, nn.W, nn.b)
pickle.dump(saved_params, fh)
fh.close()
# Calculate Validation Set Error
val_batch = 500
print "Start training"
# phone48to39 = load_dict_48to39()
# pdb.set_trace()
p48list = load_list39to48()
p48to39dict = load_dict_48to39()
totaltime = 0
for epoch in range(MAX_EPOCH):
tStart = time.time()
cost = 0
for batched_inputs,batched_outputs in zip(BATCHED_INPUT,BATCHED_OUTPUT):
cost += nn.train(batched_inputs,batched_outputs)
tEnd = time.time()
totaltime += tEnd - tStart
# Calculate Validation Error
valsum = 0
for val in validationNlabel:
val_x = np.transpose( np.asarray([val[1]],dtype='float32') )
p_feat = nn.test(val_x)
pos = np.argmax(p_feat)
p_48 = p48list[pos]
p_39 = p48to39dict[p_48]
if val[2] == p_48:
valsum += 1
print "valdiating:",val[2],p_48
valcorrect = float(valsum)/len(validationNlabel)
# phone48to39 = load_dict_48to39()
# pdb.set_trace()
totaltime = 0
val_batch = 500
val_output = []
print "Start training......"
for epoch in range(MAX_EPOCH):
tStart = time.time()
cost = 0
for idx in range(0,len(BATCHED_INPUT)):
cost += nn.train(BATCHED_INPUT[idx],BATCHED_OUTPUT[idx],
LEARNING_RATE,
MOMENTUM,
RETAIN_PROB,
1)
tEnd = time.time()
del BATCHED_INPUT
del BATCHED_OUTPUT
totaltime += tEnd - tStart
if (epoch+1 != MAX_EPOCH) and ((epoch+1) % L_RATE_DECAY_STEP == 0):
print "learning rate annealed at epoch {0}".format(epoch+1)
LEARNING_RATE*=0.5
if epoch+1 != MAX_EPOCH and (epoch+1) % SAVE_MODEL_EPOCH == 0:
fh = open(MODEL_ROOT+MODEL+"_at_{0}".format(epoch+1),'wb')
saved_params = (nn.layers, nn.W, nn.b)
pickle.dump(saved_params, fh)
fh.close()
Val_IDs, Val_Feats = SepIDnFeat(LABELED_VALIDATION_SET)
val_x = np.asarray(Val_Feats, dtype='float32').T
val_label = [LABEL_DICT[ID] for ID in Val_IDs]
# phone48to39 = load_dict_48to39()
# pdb.set_trace()
totaltime = 0
print "Start training......"
for epoch in range(100, MAX_EPOCH):
tStart = time.time()
cost = 0
for batched_inputs, batched_outputs in zip(BATCHED_INPUT, BATCHED_OUTPUT):
cost += nn.train(batched_inputs, batched_outputs, LEARNING_RATE,
MOMENTUM, RETAIN_PROB, INPUT_R_PROB)
tEnd = time.time()
totaltime += tEnd - tStart
if (epoch + 1 != MAX_EPOCH) and ((epoch + 1) % L_RATE_DECAY_STEP == 0):
print "learning rate annealed at epoch {0}".format(epoch + 1)
LEARNING_RATE /= 10
if epoch + 1 != MAX_EPOCH and (epoch + 1) % SAVE_MODEL_EPOCH == 0:
fh = open(MODEL_ROOT + MODEL + "_at_{0}".format(epoch + 1), 'wb')
saved_params = (nn.layers, nn.W, nn.b)
pickle.dump(saved_params, fh)
fh.close()
# Calculate Validation Set Error
val_batch = 500
val_output = []
for i in xrange((val_x.shape[1] - 1) / val_batch + 1):
print "Start training"
# phone48to39 = load_dict_48to39()
# pdb.set_trace()
p48list = load_list39to48()
p48to39dict = load_dict_48to39()
totaltime = 0
for epoch in range(MAX_EPOCH):
tStart = time.time()
cost = 0
for batched_inputs, batched_outputs in zip(BATCHED_INPUT, BATCHED_OUTPUT):
cost += nn.train(batched_inputs, batched_outputs)
tEnd = time.time()
totaltime += tEnd - tStart
# Calculate Validation Error
valsum = 0
for val in validationNlabel:
val_x = np.transpose(np.asarray([val[1]], dtype='float32'))
p_feat = nn.test(val_x)
pos = np.argmax(p_feat)
p_48 = p48list[pos]
p_39 = p48to39dict[p_48]
if val[2] == p_48:
valsum += 1
print "valdiating:", val[2], p_48
valcorrect = float(valsum) / len(validationNlabel)