def extract_features_and_generate_model(essays, algorithm=util_functions.AlgorithmTypes.regression):
"""
Feed in an essay set to get feature vector and classifier
essays must be an essay set object
additional array is an optional argument that can specify
a numpy array of values to add in
returns a trained FeatureExtractor object and a trained classifier
"""
f = feature_extractor.FeatureExtractor()
f.initialize_dictionaries(essays)
train_feats = f.gen_feats(essays)
set_score = numpy.asarray(essays._score, dtype=numpy.int)
if len(util_functions.f7(list(set_score)))>5:
algorithm = util_functions.AlgorithmTypes.regression
else:
algorithm = util_functions.AlgorithmTypes.classification
clf,clf2 = get_algorithms(algorithm)
cv_error_results=get_cv_error(clf2,train_feats,essays._score)
try:
clf.fit(train_feats, set_score)
except ValueError:
log.exception("Not enough classes (0,1,etc) in sample.")
set_score[0]=1
set_score[1]=0
clf.fit(train_feats, set_score)
return f, clf, cv_error_results
def extract_features_and_generate_model(essays, type=util_functions.AlgorithmTypes.regression):
"""
Feed in an essay set to get feature vector and classifier
essays must be an essay set object
additional array is an optional argument that can specify
a numpy array of values to add in
returns a trained FeatureExtractor object and a trained classifier
"""
f = feature_extractor.FeatureExtractor()
f.initialize_dictionaries(essays)
train_feats = f.gen_feats(essays)
set_score = numpy.asarray(essays._score, dtype=numpy.int)
if len(util_functions.f7(list(set_score))) > 5:
type = util_functions.AlgorithmTypes.regression
else:
type = util_functions.AlgorithmTypes.classification
clf, clf2 = get_algorithms(type)
cv_error_results = get_cv_error(clf2, train_feats, essays._score)
try:
clf.fit(train_feats, set_score)
except ValueError:
log.exception("Not enough classes (0,1,etc) in sample.")
set_score[0] = 1
set_score[1] = 0
clf.fit(train_feats, set_score)
return f, clf, cv_error_results
def select_algorithm(score_list):
#Decide what algorithm to use (regression or classification)
try:
#Count the number of unique score points in the score list
if len(util_functions.f7(list(score_list))) > 5:
algorithm = util_functions.AlgorithmTypes.regression
else:
algorithm = util_functions.AlgorithmTypes.classification
except:
algorithm = util_functions.AlgorithmTypes.regression
return algorithm
def select_algorithm(score_list):
#Decide what algorithm to use (regression or classification)
try:
#Count the number of unique score points in the score list
if len(util_functions.f7(list(score_list)))>5:
algorithm = util_functions.AlgorithmTypes.regression
else:
algorithm = util_functions.AlgorithmTypes.classification
except:
algorithm = util_functions.AlgorithmTypes.regression
return algorithm
def create(text,score,prompt_string):
"""
Creates a machine learning model from input text, associated scores, a prompt, and a path to the model
TODO: Remove model path argument, it is needed for now to support legacy code
text - A list of strings containing the text of the essays
score - a list of integers containing score values
prompt_string - the common prompt for the set of essays
"""
#Initialize a results dictionary to return
results = {'errors': [],'success' : False, 'cv_kappa' : 0, 'cv_mean_absolute_error': 0,
'feature_ext' : "", 'classifier' : "", 'algorithm' : util_functions.AlgorithmTypes.classification,
'score' : score, 'text' : text, 'prompt' : prompt_string}
if len(text)!=len(score):
msg = "Target and text lists must be same length."
results['errors'].append(msg)
log.exception(msg)
return results
#Decide what algorithm to use (regression or classification)
try:
#Count the number of unique score points in the score list
if len(util_functions.f7(list(score)))>5:
type = util_functions.AlgorithmTypes.regression
else:
type = util_functions.AlgorithmTypes.classification
except:
type = util_functions.AlgorithmTypes.regression
try:
#Create an essay set object that encapsulates all the essays and alternate representations (tokens, etc)
e_set = model_creator.create_essay_set(text, score, prompt_string)
except:
msg = "essay set creation failed."
results['errors'].append(msg)
log.exception(msg)
try:
#Gets features from the essay set and computes error
feature_ext, classifier, cv_error_results = model_creator.extract_features_and_generate_model(e_set, type=type)
results['cv_kappa']=cv_error_results['kappa']
results['cv_mean_absolute_error']=cv_error_results['mae']
results['feature_ext']=feature_ext
results['classifier']=classifier
results['algorithm'] = type
results['success']=True
except:
msg = "feature extraction and model creation failed."
results['errors'].append(msg)
log.exception(msg)
return results
def create(text, score, prompt_string):
"""
Creates a machine learning model from input text, associated scores, a prompt, and a path to the model
TODO: Remove model path argument, it is needed for now to support legacy code
text - A list of strings containing the text of the essays
score - a list of integers containing score values
prompt_string - the common prompt for the set of essays
"""
# Initialize a results dictionary to return
results = {
'errors': [],
'success': False,
'cv_kappa': 0,
'cv_mean_absolute_error': 0,
'feature_ext': "",
'classifier': "",
'algorithm': util_functions.AlgorithmTypes.classification,
'score': score,
'text': text,
'prompt': prompt_string
}
if len(text) != len(score):
msg = "Target and text lists must be same length."
results['errors'].append(msg)
log.exception(msg)
return results
# Decide what algorithm to use (regression or classification)
try:
# Count the number of unique score points in the score list
if len(util_functions.f7(list(score))) > 5:
type = util_functions.AlgorithmTypes.regression
else:
type = util_functions.AlgorithmTypes.classification
except:
type = util_functions.AlgorithmTypes.regression
try:
# Create an essay set object that encapsulates all the essays and alternate representations (tokens, etc)
e_set = model_creator.create_essay_set(text, score, prompt_string)
except:
msg = "essay set creation failed."
results['errors'].append(msg)
log.exception(msg)
try:
# Gets features from the essay set and computes error
feature_ext, classifier, cv_error_results = model_creator.extract_features_and_generate_model(
e_set, type=type)
results['cv_kappa'] = cv_error_results['kappa']
results['cv_mean_absolute_error'] = cv_error_results['mae']
results['feature_ext'] = feature_ext
results['classifier'] = classifier
results['algorithm'] = type
results['success'] = True
except:
msg = "feature extraction and model creation failed."
results['errors'].append(msg)
log.exception(msg)
return results
