def main():
print("Getting features for valid papers from the database")
#data = data_io.get_features_db("ValidPaper")
data = data_io.get_precomputed_features("ValidFeatures")
author_paper_ids = [x[:2] for x in data]
features = [x[2:] for x in data]
print("Loading the classifier")
classifier = data_io.load_model()
print("Making predictions")
predictions = classifier.predict_proba(features)[:, 1]
predictions = list(predictions)
author_predictions = defaultdict(list)
paper_predictions = {}
for (a_id, p_id), pred in zip(author_paper_ids, predictions):
author_predictions[a_id].append((pred, p_id))
for author_id in sorted(author_predictions):
paper_ids_sorted = sorted(author_predictions[author_id], reverse=True)
paper_predictions[author_id] = [x[1] for x in paper_ids_sorted]
print("Writing predictions to file")
data_io.write_submission(paper_predictions)
def main():
print("Getting features for valid papers from the database")
#data = data_io.get_features_db("ValidPaper")
data = data_io.get_precomputed_features("ValidFeatures")
author_paper_ids = [x[:2] for x in data]
features = [x[2:] for x in data]
print("Loading the classifier")
classifier = data_io.load_model()
print("Making predictions")
predictions = classifier.predict_proba(features)[:,1]
predictions = list(predictions)
author_predictions = defaultdict(list)
paper_predictions = {}
for (a_id, p_id), pred in zip(author_paper_ids, predictions):
author_predictions[a_id].append((pred, p_id))
for author_id in sorted(author_predictions):
paper_ids_sorted = sorted(author_predictions[author_id], reverse=True)
paper_predictions[author_id] = [x[1] for x in paper_ids_sorted]
print("Writing predictions to file")
data_io.write_submission(paper_predictions)
def main():
print("Getting features for deleted papers from the database")
#features_deleted = data_io.get_features_db("TrainDeleted")
features_deleted = data_io.get_precomputed_features("DeletedFeatures")
print("Getting features for confirmed papers from the database")
#features_conf = data_io.get_features_db("TrainConfirmed")
features_conf = data_io.get_precomputed_features("ConfirmedFeatures")
print("Getting features for deleted papers from the database")
#valid_features_deleted = data_io.get_features_db("ValidDeleted")
valid_features_deleted = data_io.get_precomputed_features(
"ValidDeletedFeatures")
print("Getting features for confirmed papers from the database")
#valid_features_conf = data_io.get_features_db("ValidConfirmed")
valid_features_conf = data_io.get_precomputed_features(
"ValidConfirmedFeatures")
features = [
x[2:] for x in features_deleted + features_conf +
valid_features_deleted + valid_features_conf
]
target = [0 for x in range(len(features_deleted))] + [1 for x in range(len(features_conf))] \
+ [0 for x in range(len(valid_features_deleted))] + [1 for x in range(len(valid_features_conf))]
print("Training the Classifier")
clfRF = RandomForestClassifier(n_estimators=100,
verbose=2,
n_jobs=10,
min_samples_split=10,
compute_importances=True,
random_state=1)
clfGBM = GradientBoostingClassifier(n_estimators=100,
verbose=2,
min_samples_split=10,
random_state=1)
classifier = clfRF
classifier.fit(features, target)
print("Saving the classifier")
#data_io.save_model(clfier)
print "Feature importance", classifier.feature_importances_
def main():
print("Getting features for deleted papers from the database")
#features_deleted = data_io.get_features_db("TrainDeleted")
features_deleted = data_io.get_precomputed_features("DeletedFeatures")
print("Getting features for confirmed papers from the database")
#features_conf = data_io.get_features_db("TrainConfirmed")
features_conf = data_io.get_precomputed_features("ConfirmedFeatures")
print("Getting features for deleted papers from the database")
#valid_features_deleted = data_io.get_features_db("ValidDeleted")
valid_features_deleted = data_io.get_precomputed_features("ValidDeletedFeatures")
print("Getting features for confirmed papers from the database")
#valid_features_conf = data_io.get_features_db("ValidConfirmed")
valid_features_conf = data_io.get_precomputed_features("ValidConfirmedFeatures")
features = [x[2:] for x in features_deleted + features_conf + valid_features_deleted + valid_features_conf]
target = [0 for x in range(len(features_deleted))] + [1 for x in range(len(features_conf))] \
+ [0 for x in range(len(valid_features_deleted))] + [1 for x in range(len(valid_features_conf))]
print("Training the Classifier")
clfRF = RandomForestClassifier(n_estimators=100,
verbose=2,
n_jobs=10,
min_samples_split=10,
compute_importances=True,
random_state=1)
clfGBM = GradientBoostingClassifier(n_estimators=100,
verbose=2,
min_samples_split=10,
random_state=1)
classifier = clfRF
classifier.fit(features, target)
print("Saving the classifier")
#data_io.save_model(clfier)
print "Feature importance", classifier.feature_importances_
def main():
print("Getting features for deleted papers from the database")
#features_deleted = data_io.get_features_db("TrainDeleted")
features_deleted = data_io.get_precomputed_features("DeletedFeatures")
print("Getting features for confirmed papers from the database")
#features_conf = data_io.get_features_db("TrainConfirmed")
features_conf = data_io.get_precomputed_features("ConfirmedFeatures")
print("Getting features for deleted papers from the database")
#valid_features_deleted = data_io.get_features_db("ValidDeleted")
valid_features_deleted = data_io.get_precomputed_features("ValidDeletedFeatures")
print("Getting features for confirmed papers from the database")
#valid_features_conf = data_io.get_features_db("ValidConfirmed")
valid_features_conf = data_io.get_precomputed_features("ValidConfirmedFeatures")
features = [x[2:] for x in features_deleted + features_conf] #+ valid_features_deleted + valid_features_conf
target = [0 for x in range(len(features_deleted))] + [1 for x in range(len(features_conf))]
#+ [0 for x in range(len(valid_features_deleted))] + [1 for x in range(len(valid_features_conf))]
print("Training the Classifier")
RF = RandomForestClassifier(n_estimators=50,
verbose=2,
n_jobs=1,
min_samples_split=10,
compute_importances=True,
random_state=1)
GBM = GradientBoostingClassifier(n_estimators=100,
verbose=2,
min_samples_split=10,
random_state=1)
classifier = RF
classifier.fit(features, target)
# Validation
author_paper_ids = [x[:2] for x in valid_features_conf+valid_features_deleted]
features = [x[2:] for x in valid_features_conf+valid_features_deleted]
print("Making predictions")
predictions = classifier.predict_proba(features)[:,1]
predictions = list(predictions)
author_predictions = defaultdict(list)
paper_predictions = {}
for (a_id, p_id), pred in zip(author_paper_ids, predictions):
author_predictions[str(a_id)].append((pred,str(p_id)))
for author_id in sorted(author_predictions):
paper_ids_sorted = sorted(author_predictions[author_id], reverse=True)
paper_predictions[author_id] = [x[1] for x in paper_ids_sorted]
predicted = paper_predictions.items()
predicted.sort()
#Now I have sorted predictions for each author_id
#Need to get the ground truth for the validation set:
valid_confirmed_data = [row for row in csv.reader(open("ValidSolution.csv"))] #TrainConfirmed.csv
valid_confirmed_papers = [(row[0],row[1].split()) for row in valid_confirmed_data[1:]]
valid_confirmed_papers.sort()
print predicted[0]
print valid_confirmed_papers[0]
import ml_metrics as metrics
print metrics.mapk([row[1] for row in valid_confirmed_papers], [row[1] for row in predicted],10000)
def main():
print("Getting features for deleted papers from the database")
#features_deleted = data_io.get_features_db("TrainDeleted")
features_deleted = data_io.get_precomputed_features("DeletedFeaturester")
print("Getting features for confirmed papers from the database")
#features_conf = data_io.get_features_db("TrainConfirmed")
features_conf = data_io.get_precomputed_features("ConfirmedFeaturester")
print("Getting features for deleted papers from the database")
#valid_features_deleted = data_io.get_features_db("ValidDeleted")
valid_features_deleted = data_io.get_precomputed_features("DeletedValidFeaturester")
print("Getting features for confirmed papers from the database")
#valid_features_conf = data_io.get_features_db("ValidConfirmed")
valid_features_conf = data_io.get_precomputed_features("ConfirmedValidFeaturester")
all_features = [x for x in features_deleted + features_conf + valid_features_deleted + valid_features_conf]
all_target = [0 for x in range(len(features_deleted))] + [1 for x in range(len(features_conf))] \
+ [0 for x in range(len(valid_features_deleted))] + [1 for x in range(len(valid_features_conf))]
print "Load ground truth"
valid_confirmed_data = [row for row in csv.reader(open("ValidSolution.csv"))]
valid_confirmed_papers = [(row[0],row[1].split()) for row in valid_confirmed_data[1:]]
valid_confirmed_papers.sort()
train_confirmed_data = [row for row in csv.reader(open("Train.csv"))]
train_confirmed_papers = [(row[0],row[1].split()) for row in train_confirmed_data[1:]]
train_confirmed_papers.sort()
ground_truth = valid_confirmed_papers + train_confirmed_papers
authors = [row[0] for row in ground_truth]
scaling = False
mp = []
for k in xrange(10):
# Now split authors anyway you like.
print "Split data"
authors_train, authors_test = cross_validation.train_test_split(authors,test_size=0.1, random_state=k)
print "Build training set"
train_indices = [i for (i,x) in enumerate(all_features) if str(x[0]) in authors_train]
test_indices = [i for (i,x) in enumerate(all_features) if str(x[0]) in authors_test]
train_features = [map(float,all_features[i][2:]) for i in train_indices]
print len(train_features)
print train_features[0]
if scaling:
scaler = preprocessing.StandardScaler().fit(train_features)
train_features = scaler.transform(train_features)
train_targets = [all_target[i] for i in train_indices]
print len(train_targets)
print "Build test set"
author_paper_ids = [all_features[i][:2] for i in test_indices]
test_features = [map(float,all_features[i][2:]) for i in test_indices]
if scaling:
test_features = scaler.transform(test_features)
test_targets = [all_target[i] for i in test_indices]
test_ground_truth = [row for row in ground_truth if row[0] in authors_test]
test_ground_truth.sort()
print("Training the Classifier")
RF = RandomForestClassifier(n_estimators=100, verbose=1, n_jobs=10, min_samples_split=10, compute_importances=True, random_state=1)
SVM = svm.SVC(cache_size=1000, verbose=True)
knn = neighbors.KNeighborsClassifier()
GBM = GradientBoostingClassifier(n_estimators=100, verbose=1, min_samples_split=10, random_state=1)
log = linear_model.LogisticRegression(random_state=1)
classifier = log
classifier.fit(train_features, train_targets)
if classifier == RF or classifier == GBM:
print "Feature importance", classifier.feature_importances_
print("Making predictions")
predictions = classifier.predict_proba(test_features)[:,1]
predictions = list(predictions)
author_predictions = defaultdict(list)
paper_predictions = {}
for (a_id, p_id), pred in zip(author_paper_ids, predictions):
author_predictions[str(a_id)].append((pred,str(p_id)))
for author_id in sorted(author_predictions):
paper_ids_sorted = sorted(author_predictions[author_id], reverse=True)
paper_predictions[author_id] = [x[1] for x in paper_ids_sorted]
predicted = paper_predictions.items()
predicted.sort()
print [x[0] for x in predicted[:5]]
def main():
print("Getting features for deleted papers from the database")
#features_deleted = data_io.get_features_db("TrainDeleted")
features_deleted = data_io.get_precomputed_features("DeletedFeaturester")
print("Getting features for confirmed papers from the database")
#features_conf = data_io.get_features_db("TrainConfirmed")
features_conf = data_io.get_precomputed_features("ConfirmedFeaturester")
print("Getting features for deleted papers from the database")
#valid_features_deleted = data_io.get_features_db("ValidDeleted")
valid_features_deleted = data_io.get_precomputed_features(
"DeletedValidFeaturester")
print("Getting features for confirmed papers from the database")
#valid_features_conf = data_io.get_features_db("ValidConfirmed")
valid_features_conf = data_io.get_precomputed_features(
"ConfirmedValidFeaturester")
all_features = [
x for x in features_deleted + features_conf + valid_features_deleted +
valid_features_conf
]
all_target = [0 for x in range(len(features_deleted))] + [1 for x in range(len(features_conf))] \
+ [0 for x in range(len(valid_features_deleted))] + [1 for x in range(len(valid_features_conf))]
print "Load ground truth"
valid_confirmed_data = [
row for row in csv.reader(open("ValidSolution.csv"))
]
valid_confirmed_papers = [(row[0], row[1].split())
for row in valid_confirmed_data[1:]]
valid_confirmed_papers.sort()
train_confirmed_data = [row for row in csv.reader(open("Train.csv"))]
train_confirmed_papers = [(row[0], row[1].split())
for row in train_confirmed_data[1:]]
train_confirmed_papers.sort()
ground_truth = valid_confirmed_papers + train_confirmed_papers
authors = [row[0] for row in ground_truth]
scaling = False
mp = []
for k in xrange(10):
# Now split authors anyway you like.
print "Split data"
authors_train, authors_test = cross_validation.train_test_split(
authors, test_size=0.1, random_state=k)
print "Build training set"
train_indices = [
i for (i, x) in enumerate(all_features)
if str(x[0]) in authors_train
]
test_indices = [
i for (i, x) in enumerate(all_features)
if str(x[0]) in authors_test
]
train_features = [
map(float, all_features[i][2:]) for i in train_indices
]
print len(train_features)
print train_features[0]
if scaling:
scaler = preprocessing.StandardScaler().fit(train_features)
train_features = scaler.transform(train_features)
train_targets = [all_target[i] for i in train_indices]
print len(train_targets)
print "Build test set"
author_paper_ids = [all_features[i][:2] for i in test_indices]
test_features = [map(float, all_features[i][2:]) for i in test_indices]
if scaling:
test_features = scaler.transform(test_features)
test_targets = [all_target[i] for i in test_indices]
test_ground_truth = [
row for row in ground_truth if row[0] in authors_test
]
test_ground_truth.sort()
print("Training the Classifier")
RF = RandomForestClassifier(n_estimators=100,
verbose=1,
n_jobs=10,
min_samples_split=10,
compute_importances=True,
random_state=1)
SVM = svm.SVC(cache_size=1000, verbose=True)
knn = neighbors.KNeighborsClassifier()
GBM = GradientBoostingClassifier(n_estimators=100,
verbose=1,
min_samples_split=10,
random_state=1)
log = linear_model.LogisticRegression(random_state=1)
classifier = log
classifier.fit(train_features, train_targets)
if classifier == RF or classifier == GBM:
print "Feature importance", classifier.feature_importances_
print("Making predictions")
predictions = classifier.predict_proba(test_features)[:, 1]
predictions = list(predictions)
author_predictions = defaultdict(list)
paper_predictions = {}
for (a_id, p_id), pred in zip(author_paper_ids, predictions):
author_predictions[str(a_id)].append((pred, str(p_id)))
for author_id in sorted(author_predictions):
paper_ids_sorted = sorted(author_predictions[author_id],
reverse=True)
paper_predictions[author_id] = [x[1] for x in paper_ids_sorted]
predicted = paper_predictions.items()
predicted.sort()
print[x[0] for x in predicted[:5]]
print[x[0] for x in test_ground_truth[:5]]
mp.append(
metrics.mapk([row[1] for row in test_ground_truth],
[row[1] for row in predicted], 10000))
print mp[k]
print numpy.mean(mp)
print numpy.std(mp)