def main():
# Argument Parser
parser = argparse.ArgumentParser()
parser.add_argument("-w",
"--weights",
type=str,
required=True,
help="model weights")
parser.add_argument("-m",
"--model",
type=str,
required=True,
help="model json")
parser.add_argument("-p",
"--prediction",
type=str,
required=True,
help="prediction file")
args = parser.parse_args()
# Load data
print 'Loading data...'
X_test = load_data()[1][0]
TEST_ID = '../Data/pkl/img_q_id_test'
TEST_ID_PKL = pickle.load(open(TEST_ID + '.pkl', 'rb'))
# Create Model
print 'Loading model...'
model = model_from_json(open(args.model).read())
model.load_weights(args.weights)
# Predict
print 'Predicting...'
probs = model.predict(X_test, batch_size=128)
ids = map(nameToId,
[TEST_ID_PKL[idx][1] for idx in range(len(TEST_ID_PKL))])
answers = map(numToC, np.argmax(probs[:, :5], axis=1).tolist())
prediction = zip(ids, answers)
# Write to CSV
print 'Writing to CSV...'
with open(args.prediction, 'wb') as fout:
c = csv.writer(fout, delimiter=',')
c.writerow(['q_id', 'ans'])
c.writerows(prediction)
print 'Done'
def main():
# Argument Parser
parser = argparse.ArgumentParser()
parser.add_argument("-w", "--weights", type=str, required=True, help="model weights")
parser.add_argument("-m", "--model", type=str, required=True, help="model json")
parser.add_argument("-p", "--prediction", type=str, required=True, help="prediction file")
args = parser.parse_args()
# Load data
print "Loading data..."
X_test = load_data()[1][0]
TEST_ID = "../Data/pkl/img_q_id_test"
TEST_ID_PKL = pickle.load(open(TEST_ID + ".pkl", "rb"))
# Create Model
print "Loading model..."
model = model_from_json(open(args.model).read())
model.load_weights(args.weights)
# Predict
print "Predicting..."
probs = model.predict(X_test, batch_size=128)
ids = map(nameToId, [TEST_ID_PKL[idx][1] for idx in range(len(TEST_ID_PKL))])
answers = map(numToC, np.argmax(probs[:, :5], axis=1).tolist())
prediction = zip(ids, answers)
# Write to CSV
print "Writing to CSV..."
with open(args.prediction, "wb") as fout:
c = csv.writer(fout, delimiter=",")
c.writerow(["q_id", "ans"])
c.writerows(prediction)
print "Done"
# Parameters
batch_size = 200
nb_epoch = 20
verbose = 1
validation_split = 0.1
shuffle = True
show_accuracy = True
MODEL_ROOT = '../models/elephas/'
PREDICTION_ROOT = '../predictions/'
MODEL = 'sixlayeradagrad_noLSTM_batch_{}'.format(batch_size)
print 'Loading data...'
(X_train,Y_train),(X_test,Y_test) = load_data()
TEST_ID = '../Data/pkl/img_q_id_test'
TEST_ID_PKL = pickle.load(open(TEST_ID+'.pkl','rb'))
ids = map(nameToId,[ TEST_ID_PKL[idx][1] for idx in range(len(TEST_ID_PKL)) ])
print 'Building model...'
model = Akar.keras_model(1)
#print 'Defining callbacks...'
#checkpoint = ModelCheckpoint('../models/elephas/checkpoint_'+MODEL+'.h5', monitor='val_loss', verbose=0, save_best_only=True, mode='auto')
#earlystopping = EarlyStopping(monitor='val_loss', patience=2, verbose=0)
print 'Start training...'
for epoch in [20,40,60,80,100,120,140,160,180,200]:
model.fit( X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch, verbose=verbose, callbacks=[],validation_split=validation_split, shuffle=shuffle,show_accuracy=show_accuracy)
model.save_weights(MODEL_ROOT+MODEL+"_{}.h5".format(epoch))
import argparse
import cPickle as pickle
import csv
import numpy as np
import pdb
import sys
import Keras_model_deep_noLSTM as Akar
import vqa
(X_train,Y_train),(X_test,Y_test) = vqa.load_data()
PREDICTION_FILE_NAME = '../predictions/test_elephat_test'
MODEL_NAME = '../models/elephas/overfit_noLSTM_100.h5'
TEST_ID = '../Data/pkl/img_q_id_test'
TEST_ID_PKL = pickle.load(open(TEST_ID+'.pkl','rb'))
print "start making model..."
model = Akar.keras_model(1)
model.load_weights(MODEL_NAME)
print "Start testing..."
prediction = model._predict([X_test])
def nameToId(ans_string):
return '{0:{fill}{align}7}'.format(ans_string,fill='0',align='>')
def numToC(ans_int):
if ans_int == 0:
# Parameters
batch_size = 200
nb_epoch = 20
verbose = 1
validation_split = 0.1
shuffle = True
show_accuracy = True
MODEL_ROOT = '../models/elephas/'
PREDICTION_ROOT = '../predictions/'
MODEL = 'attention_noLSTM_batch_{}'.format(batch_size)
print 'Loading data...'
(X_train,Y_train),(X_test,Y_test) = load_data()
TEST_ID = '../Data/pkl/img_q_id_test'
TEST_ID_PKL = pickle.load(open(TEST_ID+'.pkl','rb'))
ids = map(nameToId,[ TEST_ID_PKL[idx][1] for idx in range(len(TEST_ID_PKL)) ])
print 'Building model...'
model = Akar.keras_model(1)
#print 'Defining callbacks...'
#checkpoint = ModelCheckpoint('../models/elephas/checkpoint_'+MODEL+'.h5', monitor='val_loss', verbose=0, save_best_only=True, mode='auto')
#earlystopping = EarlyStopping(monitor='val_loss', patience=2, verbose=0)
print 'Start training...'
for epoch in [20,40,60,80,100]:
model.fit( X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch, verbose=verbose, callbacks=[],validation_split=validation_split, shuffle=shuffle,show_accuracy=show_accuracy)
model.save_weights(MODEL_ROOT+MODEL+"_{}.h5".format(epoch))
def main():
model_root = '../models/'
model_json = '../models/json/'
# Models to ensemble
# ( 'model_name','model_type')
model_types = [
('elephas/overfit_noLSTM_100.h5','deep_noLSTM_1.json'),
#('no_lstm_5layer_250.h5','noLSTM_1.json'),
('elephas/overfit_noLSTM_batch_100_100.h5','deep_noLSTM_1.json'),
('elephas/overfit_noLSTM_batch_500_80.h5','deep_noLSTM_1.json'),
('elephas/overfit_noLSTM_batch_1000_60.h5','deep_noLSTM_1.json')
]
# Load data
print 'Loading data...'
X_test = load_data()[1][0]
TEST_ID = '../Data/pkl/img_q_id_test'
TEST_ID_PKL = pickle.load(open(TEST_ID+'.pkl','rb'))
numoftest = X_test.shape[0]
# Load Models
print 'Loading models...'
models = []
for weights, json in model_types:
m = model_from_json(open(model_json+json).read())
m.load_weights(model_root+weights)
models.append(m)
# Predict
print 'Predicting...'
grouped_answers = []
for model in models:
probs = model.predict(X_test,batch_size=128)
answers= map(numToC,np.argmax(probs,axis=1).tolist())
grouped_answers.append(answers)
# Ensemble
print 'Ensembling...'
answers = []
for idx in xrange(numoftest):
curlist = []
for m in range(len(models)):
curlist.append(grouped_answers[m][idx])
max_cnt = Counter(curlist)
m = max( v for _, v in max_cnt.iteritems())
r = [ k for k, v in max_cnt.iteritems() if v == m ]
shuffle(r)
answers.append(r[0])
# Write to CSV
ids = map(nameToId,[ TEST_ID_PKL[idx][1] for idx in range(len(TEST_ID_PKL)) ])
prediction = zip(ids,answers)
print 'Writing to CSV...'
with open('test_ensemble.csv','wb') as fout:
c = csv.writer(fout,delimiter =',')
c.writerow(['q_id','ans'])
c.writerows(prediction)
print 'Done'
def main():
model_root = '../models/'
model_json = '../models/json/'
# Models to ensemble
# ( 'model_name','model_type')
model_types = [
('elephas/overfit_noLSTM_100.h5', 'deep_noLSTM_1.json'),
#('no_lstm_5layer_250.h5','noLSTM_1.json'),
('elephas/overfit_noLSTM_batch_100_100.h5', 'deep_noLSTM_1.json'),
('elephas/overfit_noLSTM_batch_500_80.h5', 'deep_noLSTM_1.json'),
('elephas/overfit_noLSTM_batch_1000_60.h5', 'deep_noLSTM_1.json')
]
# Load data
print 'Loading data...'
X_test = load_data()[1][0]
TEST_ID = '../Data/pkl/img_q_id_test'
TEST_ID_PKL = pickle.load(open(TEST_ID + '.pkl', 'rb'))
numoftest = X_test.shape[0]
# Load Models
print 'Loading models...'
models = []
for weights, json in model_types:
m = model_from_json(open(model_json + json).read())
m.load_weights(model_root + weights)
models.append(m)
# Predict
print 'Predicting...'
grouped_answers = []
for model in models:
probs = model.predict(X_test, batch_size=128)
answers = map(numToC, np.argmax(probs, axis=1).tolist())
grouped_answers.append(answers)
# Ensemble
print 'Ensembling...'
answers = []
for idx in xrange(numoftest):
curlist = []
for m in range(len(models)):
curlist.append(grouped_answers[m][idx])
max_cnt = Counter(curlist)
m = max(v for _, v in max_cnt.iteritems())
r = [k for k, v in max_cnt.iteritems() if v == m]
shuffle(r)
answers.append(r[0])
# Write to CSV
ids = map(nameToId,
[TEST_ID_PKL[idx][1] for idx in range(len(TEST_ID_PKL))])
prediction = zip(ids, answers)
print 'Writing to CSV...'
with open('test_ensemble.csv', 'wb') as fout:
c = csv.writer(fout, delimiter=',')
c.writerow(['q_id', 'ans'])
c.writerows(prediction)
print 'Done'
