# -*- coding: utf-8 -*- import sys from collections import defaultdict from datetime import datetime import common_method as cm import array_method as am import MeCab import math import numpy as np from six.moves import cPickle reload(sys) sys.setdefaultencoding('utf-8') from sklearn.svm import LinearSVC if __name__ == '__main__': array = cPickle.load('/home/c-yukadotami/pkl_data/array.pkl') data_training = [x["array"] for x in array] label_training = [int(cm.label2id(x["answer"])) for x in array] estimator = LinearSVC(C=1.0) estimator.fit(data_training, label_training) label_prediction = estimator.predict(data_test) print(label_prediction)
import math import numpy as np from six.moves import cPickle reload(sys) sys.setdefaultencoding('utf-8') from sklearn.ensemble import RandomForestClassifier if __name__ == '__main__': with open('/home/c-yukadotami/pkl_data/array.pkl', 'rb') as pickle_file: array = cPickle.load(pickle_file) data_training = np.array([]) for x in xrange(1,300): np.append(data_training, array[x]["array"]) label_training = np.array([]) for x in xrange(1..300): np.append(label_training,int(cm.label2id(array[x]["answer"]))) # for x in array: # np.append(data_training, x["array"]) # label_training = np.array([]) # for x in array: # np.append(label_training,int(cm.label2id(x["answer"]))) print data_training print label_training estimator = RandomForestClassifier() estimator.fit(data_training, label_training) label_prediction = estimator.predict([1,35]) for id in label_prediction: print(cm.id2label(id))