def test_weak_classifier_adjective(test_feature_objects, adjective,
classifiers, scalers, file_ptr,
comma_file_ptr):
'''
takes in a dictionary of all features
returns a dictionary of weak classifiers for each feature
'''
# specify feature to be extracted
feature_name_list = [
"pdc_rise_count", "pdc_area", "pdc_max", "pac_energy", "pac_sc",
"pac_sv", "pac_ss", "pac_sk", "tac_area", "tdc_exp_fit", "gripper_min",
"gripper_mean", "transform_distance", "electrode_polyfit"
]
motion_scores = dict()
# for each motion (slide, squeeze, etc.)
for motion in test_feature_objects:
motion_test_set = test_feature_objects[motion]
# pull out the features specified as a vector
test_feature_vector, test_label_dict = utilities.feature_obj_2_feature_vector(
motion_test_set, feature_name_list)
# create scaler
test_feature_vector_scaled = scalers[motion].transform(
test_feature_vector)
clf = classifiers[motion]
results = clf.predict(test_feature_vector_scaled)
print classification_report(test_label_dict[1][adjective], results)
file_ptr.write('Adjective: ' + adjective + ' Motion name: ' +
motion)
file_ptr.write(
'\n' +
classification_report(test_label_dict[1][adjective], results) +
'\n\n')
motion_scores[motion] = f1_score(test_label_dict[1][adjective],
results)
# Pull out the best motion
best_motion, best_score = utilities.get_best_motion(motion_scores)
# Write results to comma split file to load
comma_file_ptr.write(adjective + ',' + str(motion_scores['tap']) + ',' +
str(motion_scores['squeeze']) + ',' +
str(motion_scores['thermal_hold']) + ',' +
str(motion_scores['slide']) + ',' +
str(motion_scores['slide_fast']) + ',' + best_motion +
'\n')
def test_adj_motion_classifier(test_feature_objects, adjective, classifiers,
scalers, feature_name_list, file_ptr,
comma_file_ptr):
'''
takes in a dictionary of all features
returns a dictionary of weak classifiers for each feature
'''
motion_scores = dict()
# for each motion (slide, squeeze, etc.)
for motion in test_feature_objects:
motion_test_set = test_feature_objects[motion]
# pull out the features specified as a vector
test_feature_vector, test_label_dict = utilities.feature_obj_2_feature_vector(
motion_test_set, feature_name_list)
# create scaler
test_feature_vector_scaled = scalers[motion].transform(
test_feature_vector)
clf = classifiers[motion]
results = clf.predict(test_feature_vector_scaled)
print classification_report(test_label_dict[1][adjective], results)
file_ptr.write('Adjective: ' + adjective + ' Motion name: ' +
motion)
file_ptr.write(
'\n' +
classification_report(test_label_dict[1][adjective], results) +
'\n\n')
motion_scores[motion] = f1_score(test_label_dict[1][adjective],
results)
# Pull out the best motion
best_motion, best_score = utilities.get_best_motion(motion_scores)
# Write results to comma split file to load
comma_file_ptr.write(adjective + ',' + str(motion_scores['tap']) + ',' +
str(motion_scores['squeeze']) + ',' +
str(motion_scores['thermal_hold']) + ',' +
str(motion_scores['slide']) + ',' +
str(motion_scores['slide_fast']) + ',' + best_motion +
'\n')
def test_weak_classifier_adjective(test_feature_objects, adjective, classifiers, scalers, file_ptr, comma_file_ptr):
'''
takes in a dictionary of all features
returns a dictionary of weak classifiers for each feature
'''
# specify feature to be extracted
feature_name_list = ["pdc_rise_count", "pdc_area", "pdc_max", "pac_energy", "pac_sc", "pac_sv", "pac_ss", "pac_sk", "tac_area", "tdc_exp_fit", "gripper_min", "gripper_mean", "transform_distance", "electrode_polyfit"]
motion_scores = dict()
# for each motion (slide, squeeze, etc.)
for motion in test_feature_objects:
motion_test_set = test_feature_objects[motion]
# pull out the features specified as a vector
test_feature_vector, test_label_dict = utilities.feature_obj_2_feature_vector(motion_test_set, feature_name_list)
# create scaler
test_feature_vector_scaled = scalers[motion].transform(test_feature_vector)
clf = classifiers[motion]
results = clf.predict(test_feature_vector_scaled)
print classification_report(test_label_dict[1][adjective], results)
file_ptr.write('Adjective: ' + adjective + ' Motion name: '+motion)
file_ptr.write('\n'+classification_report(test_label_dict[1][adjective], results)+ '\n\n')
motion_scores[motion] = f1_score(test_label_dict[1][adjective], results)
# Pull out the best motion
best_motion, best_score = utilities.get_best_motion(motion_scores)
# Write results to comma split file to load
comma_file_ptr.write(adjective+','+str(motion_scores['tap'])+',' +str(motion_scores['squeeze'])+','+str(motion_scores['thermal_hold'])+','+str(motion_scores['slide'])+','+str(motion_scores['slide_fast'])+','+best_motion +'\n')
def test_adj_motion_classifier(test_feature_objects, adjective, classifiers, scalers, feature_name_list, file_ptr, comma_file_ptr):
'''
takes in a dictionary of all features
returns a dictionary of weak classifiers for each feature
'''
motion_scores = dict()
# for each motion (slide, squeeze, etc.)
for motion in test_feature_objects:
motion_test_set = test_feature_objects[motion]
# pull out the features specified as a vector
test_feature_vector, test_label_dict = utilities.feature_obj_2_feature_vector(motion_test_set, feature_name_list)
# create scaler
test_feature_vector_scaled = scalers[motion].transform(test_feature_vector)
clf = classifiers[motion]
results = clf.predict(test_feature_vector_scaled)
print classification_report(test_label_dict[1][adjective], results)
file_ptr.write('Adjective: ' + adjective + ' Motion name: '+motion)
file_ptr.write('\n'+classification_report(test_label_dict[1][adjective], results)+ '\n\n')
motion_scores[motion] = f1_score(test_label_dict[1][adjective], results)
# Pull out the best motion
best_motion, best_score = utilities.get_best_motion(motion_scores)
# Write results to comma split file to load
comma_file_ptr.write(adjective+','+str(motion_scores['tap'])+',' +str(motion_scores['squeeze'])+','+str(motion_scores['thermal_hold'])+','+str(motion_scores['slide'])+','+str(motion_scores['slide_fast'])+','+best_motion +'\n')
def main(classifiers_pkl, all_classifiers_pkl, test_feature_pkl, ensemble_test_feature_pkl, scaler_pkl, final_classifier_pkl):
# Load data into pipeline
print "Loading data from file"
file_ptr = open(scaler_pkl)
scaler_dict = cPickle.load(file_ptr)
file_ptr = open(test_feature_pkl)
test_feature_dict = cPickle.load(file_ptr)
file_ptr = open(ensemble_test_feature_pkl)
ensemble_feature_dict = cPickle.load(file_ptr)
classifier_text_report = open("all_classifier_report_svm.txt", "w")
final_classification_text_report = open("final_classifier_report_svm.txt", "w")
feature_name_list = ["pdc_rise_count", "pdc_area", "pdc_max", "pac_energy", "pac_sc", "pac_sv", "pac_ss", "pac_sk", "tac_area", "tdc_exp_fit", "gripper_min", "gripper_mean", "transform_distance", "electrode_polyfit"]
if final_classifier_pkl != None:
file_ptr = open(final_classifier_pkl)
final_classifiers_dict = cPickle.load(file_ptr)
if all_classifiers_pkl == None:
file_ptr = open(classifiers_pkl)
classifiers_dict = cPickle.load(file_ptr)
# Parse the adjective classifier name being tested
adjective_name = classifiers_pkl.split('/')[-1].split('_')[0]
test_adj_motion_classifier(classifiers_dict, adjective_name, test_feature_dict, feature_name_list, scaler_dict)
else:
file_ptr = open(all_classifiers_pkl)
classifiers_dict = cPickle.load(file_ptr)
# Get the name of the classifier
#classifier_name = all_classifiers_pkl.split('_')[1]
classifier_name = "svm"
# Open up text file to store best
report_best_classifier = open("best_classifier_report_" + classifier_name+ ".txt", "w")
# Store the best classifier
best_classifiers = dict()
report_best_classifier.write('Adjective'+','+'tap'+','+'squeeze'+','+'thermal_hold'+','+'slide'','+'slide_fast,'+'best_motion'+'\n')
results_prediction = dict()
# Go through all of the adjective classifiers
for adj in classifiers_dict:
# Test the adjective scores
probability_vector, motion_scores, prediction, truth_vector = test_adj_motion_classifier(classifiers_dict[adj], adj, test_feature_dict, feature_name_list, scaler_dict, classifier_text_report)
# Compute the best scores
best_motion, best_score = utilities.get_best_motion(motion_scores)
#report_best_classifier.write('Adjective: '+adj)
#report_best_classifier.write('\nMotion Scores:\n' + str(motion_scores))
#report_best_classifier.write('\n\nBest Motion is: ' +best_motion+ "\n\n")
report_best_classifier.write(adj+','+str(motion_scores['tap'])+','+str(motion_scores['squeeze'])+','+str(motion_scores['thermal_hold'])+','+str(motion_scores['slide'])+','+str(motion_scores['slide_fast'])+','+best_motion+'\n')
results_prediction[adj] = prediction[best_motion]
best_classifiers[adj] = (classifiers_dict[adj][best_motion], best_motion)
# Test the final paper
#import pdb; pdb.set_trace()
#final_results = final_classifiers_dict[adj].predict(probability_vector)
#final_score = f1_score(truth_vector,final_results)
#print final_score
classifier_text_report.close()
# Store the pickle file
cPickle.dump(best_classifiers, open("best_classifiers_"+classifier_name+".pkl", "w"))
report_best_classifier.close()
def main(classifiers_pkl, all_classifiers_pkl, test_feature_pkl,
ensemble_test_feature_pkl, scaler_pkl, final_classifier_pkl):
# Load data into pipeline
print "Loading data from file"
file_ptr = open(scaler_pkl)
scaler_dict = cPickle.load(file_ptr)
file_ptr = open(test_feature_pkl)
test_feature_dict = cPickle.load(file_ptr)
file_ptr = open(ensemble_test_feature_pkl)
ensemble_feature_dict = cPickle.load(file_ptr)
classifier_text_report = open("all_classifier_report_svm.txt", "w")
final_classification_text_report = open("final_classifier_report_svm.txt",
"w")
feature_name_list = [
"pdc_rise_count", "pdc_area", "pdc_max", "pac_energy", "pac_sc",
"pac_sv", "pac_ss", "pac_sk", "tac_area", "tdc_exp_fit", "gripper_min",
"gripper_mean", "transform_distance", "electrode_polyfit"
]
if final_classifier_pkl != None:
file_ptr = open(final_classifier_pkl)
final_classifiers_dict = cPickle.load(file_ptr)
if all_classifiers_pkl == None:
file_ptr = open(classifiers_pkl)
classifiers_dict = cPickle.load(file_ptr)
# Parse the adjective classifier name being tested
adjective_name = classifiers_pkl.split('/')[-1].split('_')[0]
test_adj_motion_classifier(classifiers_dict, adjective_name,
test_feature_dict, feature_name_list,
scaler_dict)
else:
file_ptr = open(all_classifiers_pkl)
classifiers_dict = cPickle.load(file_ptr)
# Get the name of the classifier
#classifier_name = all_classifiers_pkl.split('_')[1]
classifier_name = "svm"
# Open up text file to store best
report_best_classifier = open(
"best_classifier_report_" + classifier_name + ".txt", "w")
# Store the best classifier
best_classifiers = dict()
report_best_classifier.write('Adjective' + ',' + 'tap' + ',' +
'squeeze' + ',' + 'thermal_hold' + ',' +
'slide'
',' + 'slide_fast,' + 'best_motion' +
'\n')
results_prediction = dict()
# Go through all of the adjective classifiers
for adj in classifiers_dict:
# Test the adjective scores
probability_vector, motion_scores, prediction, truth_vector = test_adj_motion_classifier(
classifiers_dict[adj], adj, test_feature_dict,
feature_name_list, scaler_dict, classifier_text_report)
# Compute the best scores
best_motion, best_score = utilities.get_best_motion(motion_scores)
#report_best_classifier.write('Adjective: '+adj)
#report_best_classifier.write('\nMotion Scores:\n' + str(motion_scores))
#report_best_classifier.write('\n\nBest Motion is: ' +best_motion+ "\n\n")
report_best_classifier.write(adj + ',' +
str(motion_scores['tap']) + ',' +
str(motion_scores['squeeze']) + ',' +
str(motion_scores['thermal_hold']) +
',' + str(motion_scores['slide']) +
',' +
str(motion_scores['slide_fast']) +
',' + best_motion + '\n')
results_prediction[adj] = prediction[best_motion]
best_classifiers[adj] = (classifiers_dict[adj][best_motion],
best_motion)
# Test the final paper
#import pdb; pdb.set_trace()
#final_results = final_classifiers_dict[adj].predict(probability_vector)
#final_score = f1_score(truth_vector,final_results)
#print final_score
classifier_text_report.close()
# Store the pickle file
cPickle.dump(best_classifiers,
open("best_classifiers_" + classifier_name + ".pkl", "w"))
report_best_classifier.close()
