def process(search_code):
data = fetch_data(search_code)
matched_sessions = filter(data)
if matched_sessions:
playsound("src/Rooster.mp3")
else:
print("No Match found")
def process(district_code):
data = fetch_data(district_code)
matched_sessions = filter(data)
if matched_sessions:
pass
# filename = '/tmp/sessions.csv'
# write_data(filename, matched_sessions)
# send_email(filename)
# send_data(matched_sessions)
else:
print("No Match found")
def post_rec():
if request.args.get('max_price', '') != '':
max_price = float(re.sub("[\$]", "", request.args.get('max_price',
'')))
else:
max_price = None
if request.args.get('min_price', '') != '':
min_price = float(re.sub("[\$]", "", request.args.get('min_price',
'')))
else:
min_price = None
if request.args.get('movein', '') != '':
movein = datetime.strptime(request.args.get('movein', ''), '%Y-%m-%d')
else:
movein = None
if request.args.get('work', '') != '':
work = str(request.args.get('work', ''))
else:
work = None
# print str(request.args.get('local', '')), len(str(request.args.get('local', '')))
if request.args.get('local', '') in types:
local = str(request.args.get('local', ''))
else:
local = None
neighborhoods = request.args.getlist('neighborhood[]')
attributes = request.args.getlist('attributes[]')
description = re.sub(r'\\[ntr]', ' ', request.args.get('description', ''))
description = re.sub('\\s+', ' ', description)
input = [
max_price, min_price, movein, neighborhoods, attributes, description,
work, local
]
keep = filter(listings, input)
filtered_listings = listings.ix[keep]
match = cluster(filtered_listings, description, keep)
df_match = listings.ix[match]
df_match = df_match[[
'title', 'price', 'neighborhood', 'movein', 'attributes',
'description', 'latitude', 'longitude'
]]
# print input[6], input[7]
if bool(input[6] and input[7]):
final_listings = add_gmaps_cols(df_match, input[6], input[7])
json = final_listings.to_json()
return recpage(json)
else:
json = df_match.to_json()
return recpage(json)
def main():
[X_train, y_train, X_test, y_test] = example0()
valid_col = filter(X_train)
print len(valid_col)
valid_col = dimension_reduce(X_train, valid_col)
print len(valid_col)
X_train = format(X_train, valid_col)
X_test = format(X_test, valid_col)
N, sl = X_train.shape
num_classes = len(np.unique(y_train))
"""Hyperparamaters"""
batch_size = 128
max_iterations = 1000
dropout = 0.8
config = {
'num_layers': 3, #number of layers of stacked RNN's
'hidden_size': 128, #memory cells in a layer
'max_grad_norm': 5, #maximum gradient norm during training
'batch_size': batch_size,
'learning_rate': .003,
'sl': sl,
'num_classes': num_classes
}
epochs = np.floor(batch_size * max_iterations / N)
print('Train %.0f samples in approximately %d epochs' % (N, epochs))
model = Model(config)
"""Session time"""
sess = tf.Session()
sess.run(model.init_op)
cost_train_ma = -np.log(
1 / float(num_classes) + 1e-9) #Moving average training cost
acc_train_ma = 0.0
for i in range(max_iterations):
X_batch, y_batch = sample_batch(X_train, y_train, batch_size)
cost_train, acc_train, _ = sess.run(
[model.cost, model.accuracy, model.train_op],
feed_dict={
model.input: X_batch,
model.labels: y_batch,
model.keep_prob: dropout
})
cost_train_ma = cost_train_ma * 0.99 + cost_train * 0.01
acc_train_ma = acc_train_ma * 0.99 + acc_train * 0.01
#Evaluate validation performance
if i % 20 == 0:
X_batch, y_batch = sample_batch(X_test, y_test, batch_size)
cost_val, summ, acc_val = sess.run(
[model.cost, model.merged, model.accuracy],
feed_dict={
model.input: X_batch,
model.labels: y_batch,
model.keep_prob: 1.0
})
print(
'At %5.0f/%5.0f: COST %5.3f/%5.3f(%5.3f) -- Acc %5.3f/%5.3f(%5.3f)'
% (i, max_iterations, cost_train, cost_val, cost_train_ma,
acc_train, acc_val, acc_train_ma))
plt.plot(i, acc_train, 'r*')
plt.plot(i, cost_train, 'kd')
cost_val, summ, acc_val = sess.run(
[model.cost, model.merged, model.accuracy],
feed_dict={
model.input: X_test,
model.labels: y_test,
model.keep_prob: 1.0
})
epoch = float(i) * batch_size / N
print('Trained %.1f epochs, accuracy is %5.3f and cost is %5.3f' %
(epoch, acc_val, cost_val))
plt.xlabel('Iteration')
plt.ylabel('Accuracy & Cost')
plt.show()