def main(_):
print 'reading npy...'
data = np.load('../1st.npy')
test_order = np.load('../test.npy')
jpg_list = np.load('64bin.npy')
print 'reading finished'
sess = tf.Session()
print 'building network...'
hg = vae_margin_test.vae(is_training=False)
global_step = tf.Variable(0,name='global_step',trainable=False)
merged_summary = tf.summary.merge_all()
summary_writer = tf.summary.FileWriter(FLAGS.summary_dir,sess.graph)
saver = tf.train.Saver(max_to_keep=None)
saver.restore(sess,FLAGS.checkpoint_path)
print 'restoring from '+FLAGS.checkpoint_path
def test_step():
print 'Testing...'
all_recon_loss = 0
all_output = []
all_label = []
batch_size = FLAGS.batch_size
for i in range(int(len(test_order)/batch_size)):
input_nlcd = get_data.get_nlcd(data,test_order[batch_size*i:batch_size*(i+1)])
input_image = get_data.get_jpg_test(jpg_list,test_order[batch_size*i:batch_size*(i+1)])
input_label = get_data.get_label(data,test_order[batch_size*i:batch_size*(i+1)])
feed_dict={}
feed_dict[hg.input_nlcd]=input_nlcd
feed_dict[hg.input_image]=input_image
feed_dict[hg.input_label]=input_label
feed_dict[hg.keep_prob]=1.0
output= sess.run(hg.output,feed_dict)
for i in output:
all_output.append(i)
for i in input_label:
all_label.append(i)
#recon_loss = all_recon_loss/(5094.0/batch_size)
all_output = np.array(all_output)
all_label = np.array(all_label)
# auc = roc_auc_score(all_label,all_output)
# all_output=np.reshape(all_output,(-1))
# all_label=np.reshape(all_label,(-1))
# ap = average_precision_score(all_label,all_output)
# time_str = datetime.datetime.now().isoformat()
# new_auc = roc_auc_score(all_label,all_output)
# tempstr = "{}: auc {:g}, ap {:g}, recon_loss {:g}, new_auc {:g}".format(time_str, auc, ap, recon_loss, new_auc)
# print(tempstr)
all_output=np.reshape(all_output,(-1,100))
all_label=np.reshape(all_label,(-1,100))
return all_output,all_label
# auc = roc_auc_score(all_label,all_output)
# print "new_auc {:g}".format(auc)
all_output = 0.0
all_label = 0.0
all_fuck = 0.0
num = 10000.0
for i in range(int(num)):
print i
output,label= test_step()
#all_output.append(output)
all_output+=output
all_label=label
#f**k = np.multiply.reduce(label,1)
f**k = tang_reduce(output,label)
all_fuck+=f**k
# tang = 0.0
# for i in all_output:
# tang+=i
# tang = tang/num
tang = all_output/num
f**k = all_fuck/num
epsilon = 0
#float_labels = tf.cast(labels, tf.float32)
loss = all_label * np.log(tang + epsilon) + (1 - all_label) * np.log(1 - tang + epsilon)
loss = np.sum(loss,axis=1)
loss = np.mean(loss)
f**k = np.mean(np.log(f**k))
#print 'another final:'+str(loss)
auc = roc_auc_score(all_label,all_output)
tang=np.reshape(tang,(-1))
all_label=np.reshape(all_label,(-1))
np.save('all_output_1wavg.npy',tang)
np.save('all_label.npy',all_label)
ap = average_precision_score(all_label,tang)
time_str = datetime.datetime.now().isoformat()
new_auc = roc_auc_score(all_label,tang)
print 'margin results'
tempstr = "{}: auc {:g}, ap {:g}, recon_loss {:g}, new_auc {:g}".format(time_str, auc, ap, loss, new_auc)
print(tempstr)
print 'average log sigma:'+str(f**k)
def main(_):
print 'reading npy...'
data = np.load('../data/1st.npy')
test_order = np.load('../data/test.npy')
print 'reading finished'
sess = tf.Session()
print 'building network...'
hg = vae_margin_test.vae(is_training=False)
global_step = tf.Variable(0,name='global_step',trainable=False)
merged_summary = tf.summary.merge_all()
summary_writer = tf.summary.FileWriter(FLAGS.summary_dir,sess.graph)
saver = tf.train.Saver(max_to_keep=None)
saver.restore(sess,FLAGS.checkpoint_path)
print 'restoring from '+FLAGS.checkpoint_path
def test_step():
print 'Testing...'
all_recon_loss = 0
all_output = []
all_label = []
batch_size = 283*6*3
for i in range(int(len(test_order)/batch_size)):
input_nlcd = get_data.get_nlcd(data,test_order[batch_size*i:batch_size*(i+1)])
input_label = get_data.get_label(data,test_order[batch_size*i:batch_size*(i+1)])
feed_dict={}
feed_dict[hg.input_nlcd]=input_nlcd
feed_dict[hg.input_label]=input_label
feed_dict[hg.keep_prob]=1.0
output= sess.run(hg.output,feed_dict)
for i in output:
all_output.append(i)
for i in input_label:
all_label.append(i)
#recon_loss = all_recon_loss/(5094.0/batch_size)
all_output = np.array(all_output)
all_label = np.array(all_label)
# auc = roc_auc_score(all_label,all_output)
# all_output=np.reshape(all_output,(-1))
# all_label=np.reshape(all_label,(-1))
# ap = average_precision_score(all_label,all_output)
# time_str = datetime.datetime.now().isoformat()
# new_auc = roc_auc_score(all_label,all_output)
# tempstr = "{}: auc {:g}, ap {:g}, recon_loss {:g}, new_auc {:g}".format(time_str, auc, ap, recon_loss, new_auc)
# print(tempstr)
all_output=np.reshape(all_output,(-1,100))
all_label=np.reshape(all_label,(-1,100))
return all_output,all_label
# auc = roc_auc_score(all_label,all_output)
# print "new_auc {:g}".format(auc)
all_output = 0.0
all_label = 0.0
all_fuck = 0.0
num = 10000.0
for i in range(int(num)):
print i
output,label= test_step()
#all_output.append(output)
all_output+=output
all_label=label
#f**k = np.multiply.reduce(label,1)
f**k = tang_reduce(output,label)
all_fuck+=f**k
# tang = 0.0
# for i in all_output:
# tang+=i
# tang = tang/num
tang = all_output/num
f**k = all_fuck/num
epsilon = 0
#float_labels = tf.cast(labels, tf.float32)
loss = all_label * np.log(tang + epsilon) + (1 - all_label) * np.log(1 - tang + epsilon)
loss = np.sum(loss,axis=1)
loss = np.mean(loss)
f**k = np.mean(np.log(f**k))
#print 'another final:'+str(loss)
auc = roc_auc_score(all_label,all_output)
tang=np.reshape(tang,(-1))
all_label=np.reshape(all_label,(-1))
np.save('all_output_1wavg.npy',tang)
np.save('all_label.npy',all_label)
ap = average_precision_score(all_label,tang)
time_str = datetime.datetime.now().isoformat()
new_auc = roc_auc_score(all_label,tang)
print 'margin results'
tempstr = "{}: auc {:g}, ap {:g}, recon_loss {:g}, new_auc {:g}".format(time_str, auc, ap, loss, new_auc)
print(tempstr)
print 'average log sigma:'+str(f**k)
