def plot():
train_time,trainingX,trainingY = read_data(0,70000)
test_time,testingX,testingY = read_data(130000,162000)
plt.plot(trainingX[:,2], scale*trainingY, 'b.', markersize=6, label=u'Observations', color=BLUE)
plt.xlabel('Time [s]')
plt.ylabel('Acceleration [g]')
plt.legend(loc='upper left')
# Plot the training time series
fig = plt.figure()
plt.plot(train_time, scale*trainingY, 'b', label=u'Observations', color=BLUE)
plt.xlabel('Time [s]')
plt.ylabel('Acceleration [g]')
plt.title('Predictions for Training Set')
# Plot the testing time series
fig = plt.figure()
plt.plot(test_time, scale*testingY, 'b', label=u'Observations', color=BLUE)
plt.xlabel('Time [s]')
plt.ylabel('Acceleration [g]')
plt.title('Bridge Acceleration')
plt.legend(['Recorded time series','Predicted Values'])
plt.xlim([660,700])
plt.savefig('test.png')
def run(self):
# Get command line arguments and initialize test_dir with directory
self.init_arguments()
self.test_dir = self.args.directory
# Get test_set by calling read_data from test_data.py
test_set = test_data.read_data(self.test_dir)
for test_doc in test_set.values():
# Initially creates a .key file from default rake and adds default keywords to the file
self.get_keywords_for_key_file(test_doc)
# Gets optimum parameters by using .key file created in previous step, and creates
# a rake object using these parameters
self.get_final_keywords(test_doc, test_set)
print('Keywords will be stored in a .key file with the same name as the input file.')
__author__ = 'a_medelyan'
import test_data
import rake
import sys
# reading a directory with test documents
input_dir = sys.argv[1]
# number of top ranked keywords to evaluate
top = int(sys.argv[2])
test_set = test_data.read_data(input_dir)
best_fmeasure = 0
best_vals = []
for min_char_length in range(3, 8):
for max_words_length in range(3, 6):
for min_keyword_frequency in range(1, 7):
rake_object = rake.Rake("SmartStoplist.txt", min_char_length,
max_words_length, min_keyword_frequency)
total_fmeasure = 0
for test_doc in test_set.values():
keywords = rake_object.run(test_doc.text)
num_manual_keywords = len(test_doc.keywords)
correct = 0
try:
for i in range(0, min(top, len(keywords))):
if keywords[i][0] in set(test_doc.keywords):
correct += 1
from __future__ import absolute_import
from __future__ import print_function
from six.moves import range
__author__ = 'a_medelyan'
import test_data
import rake
import sys
# reading a directory with test documents
input_dir = sys.argv[1]
# number of top ranked keywords to evaluate
top = int(sys.argv[2])
test_set = test_data.read_data(input_dir)
best_fmeasure = 0
best_vals = []
for min_char_length in range(3,8):
for max_words_length in range(3,6):
for min_keyword_frequency in range(1,7):
rake_object = rake.Rake("SmartStoplist.txt", min_char_length, max_words_length, min_keyword_frequency)
total_fmeasure = 0
for test_doc in test_set.values():
keywords = rake_object.run(test_doc.text)
num_manual_keywords = len(test_doc.keywords)
correct = 0
try:
for i in range(0,min(top, len(keywords))):