def __init__(self, normalize = False):
self.normalize = normalize
if self.normalize == True:
self.norm = normalizer()
import re
import jieba
from sklearn.preprocessing import MultiLabelBinarizer
from tqdm import tqdm
import h5py
from sklearn.feature_extraction.text import TfidfVectorizer as TFIDF
from sklearn.svm import LinearSVC
from predictor import data
import bz2
from normalize import normalizer
dim = 500000
special_character_removal = re.compile(
r'[@#$%^&*,.【】[]{}；‘，。、？!?“”‘’; \\/"\']', re.IGNORECASE)
replace_numbers = re.compile(r'\d+', re.IGNORECASE)
normalizer = normalizer('word.txt')
word_len = 2
# cut = thulac.thulac(seg_only=True)
def seg(text):
seg_list = jieba.cut(text.strip())
seg_list = [word for word in seg_list if len(word) >= word_len]
return " ".join(seg_list)
def text_to_wordlist(text):
return " ".join(normalizer.seg_one_text(text, 2))
# text = special_character_removal.sub('', text)
# text = replace_numbers.sub('NUMBERREPLACE', text)
@author: saurabh
"""
import numpy as np
from sklearn.linear_model import ElasticNetCV
from sklearn.metrics import r2_score, mean_squared_error
from load_data import load_data
from normalize import normalizer
import os
data_path = os.path.join("..", "ml-100k", "u.data")
X, y = load_data(data_path, test_percentage=10)
num_users, num_movies = X.shape
binary = (X > 0)
norm = normalizer()
X = norm.fit_transform(X)
predicted = X.copy()
clf = ElasticNetCV(alphas=[0.0125, 0.025, 0.05, .1, .125, .5, 1.0, 2.0, 4.0])
#clf = ElasticNetCV(alphas = [ .1])
for user in range(num_users):
#bool array for movies rated by user
movie_user = binary[user]
#which users to consider as attributes for regression, in this case all except current user
neighbors = np.ones((num_users), dtype=bool)
neighbors[user] = False
X_train_user = X[neighbors]
X_train_user = X_train_user[:, movie_user].T
def main():
fin = sys.argv[1]
fin2 = sys.argv[2]
readin(fin, fin2)
input2 = open(fin2, 'r')
nparr = np.loadtxt(fin2,delimiter=',')
study = normal.normalizer(nparr)
target = np.zeros((np.shape(study)[0],2))
index = np.where(study[:,14] == 0)
target[index,0] = 1
index = np.where(study[:,14] == 1)
target[index,1] = 1
study = study[:,:-1]
order = range(np.shape(study)[0])
np.random.shuffle(order)
study = study[order,:]
target = target[order,:]
thelen = 5
Carray = []
Tarray=[]
TP = 0
FN = 0
FP = 0
TN = 0
for i in range(5):
Carray.append([])
Tarray.append([])
ind = 0
for i in range(len(study)):
if (ind <5):
Carray[ind].append(study[i])
Tarray[ind].append(target[i])
ind = ind +1
else:
ind = 0
Errors = []
Cons= []
for i in range(20):
for j in range(5):
Target1 = np.empty((0,2))
Target2 = np.empty((0,2))
x1 = np.empty((0,14))
x2 = np.empty((0,14))
for k in range(5):
if (k!= j):
for h in range(len(Carray[k])):
x1 = np.vstack((x1,Carray[k][h]))
Target1 = np.vstack((Target1,Tarray[k][h]))
x2 = np.vstack((x2,Carray[j]))
Target2= np.vstack((Target2,Tarray[j]))
net = mlp(x1,Target1,5,momentum = 0.1, outtype='softmax')
Errors.append(net.earlystopping(x1,Target1,x2,Target2,0.3))
num = net.confmat(x2,Target2)
Cons.append(num)
print("\n")
print("The CONS",Cons)
Errors = np.array(Errors)
Max = np.amax(Errors)
print("The max error:", Max)
Min = np.amin(Errors)
print("the min error:", Min)
Mean = np.mean(Errors)
print("the mean error is:", Mean)
Std = np.std(Errors)
print("the STD is :", Std)
for i in range(len(Cons)):
for j in range(2):
if (j == 1):
F1 = F1 + Cons[i][j][0]
T2 = T2 + Cons[i][j][1]
if (j == 0):
T1 = T1 + Cons[i][j][0]
F2 = F2 + Cons[i][j][1]
T1 = T1 / len(Cons)
F2 = F2 / len(Cons)
F1 = F1 / len(Cons)
T2 = T2 / len(Cons)
Accuracy = ((T1 + T2)/(T1+T2+F1+F2))
Sensitivity = (T1/(T1+F2))
Specificity = (T2/(T2+F1))
print("The Accuracy", Accuracy)
print("The Sensitivity",Sensitivity)
print("The Specificity", Specificity)
import pandas as pd
import re
import logging
import multiprocessing
from normalize import normalizer
import os
import json
import config as cf
logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s',
level=logging.INFO)
SOME_FIXED_SEED = 20
replace_numbers = re.compile(r'\d+', re.IGNORECASE)
normalizer_ = normalizer(
'E:/pycharm/judicial_doc_measurement/utils/lda/word.txt')
word_len = 2
progress = 0
#获取单篇文章的法条
def get_law(node):
law_tmp = ''
law_list = []
for subnode in node:
if subnode.tag == 'FLFTFZ':
name = ''
for subsubnode in subnode:
if subsubnode.tag == 'MC':
name = subsubnode.get('value')
elif subsubnode.tag == 'T':
import numpy as np
import pandas as pd
import re
import logging
import multiprocessing
from normalize import normalizer
import os
import json
logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)
SOME_FIXED_SEED = 20
replace_numbers = re.compile(r'\d+', re.IGNORECASE)
normalizer_ = normalizer('./word.txt')
word_len = 2
progress = 0
#获取单篇文章的法条
def get_law(node):
law_tmp = ''
law_list = []
for subnode in node:
if subnode.tag == 'FLFTFZ':
name = ''
for subsubnode in subnode:
if subsubnode.tag == 'MC':
name = subsubnode.get('value')
elif subsubnode.tag == 'T':
from normalize import normalizer
from queue import Queue
import multiprocessing as mp
from method_handlers import *
import commands
import sys
import os
ERROR_FREE_CODE = True
if len(sys.argv) != 2:
print("Usage: python3 parser.py </path/to/GCode_file>")
sys.exit(1)
normalizer(sys.argv[1])
class Machine(object):
ROOM_TEMP = 27
def __init__(self,
instruction_queue,
message_queue,
current_position=(0, 0, 0),
origin=(0, 0, 0),
feed_rate=0,
extrusion_rate=0,
fan_on=False,
bed_temp=ROOM_TEMP,