def Listen():
try:
signal.signal(signal.SIGALRM, handler)
signal.alarm(100)
flag = checkConn()
if flag == True:
getData()
signal.alarm(0)
except AssertionError:
exit(0)
def Listen():
try:
signal.signal(signal.SIGALRM, handler)
signal.alarm(100)
flag = checkConn();
if flag == True:
getData()
signal.alarm(0)
except AssertionError:
exit(0)
def linearRegreSin(url,degree):
[a,b] = getData(url)
trainA = a[0:139]
trainB = b[0:139]
testA = a[140:]
testB = b[140:]
poly = PolynomialFeatures(degree)
trainA = np.float64(poly.fit_transform(trainA))
testA = np.float64(poly.fit_transform(testA))
theta = np.dot(np.dot(np.linalg.inv(np.dot(trainA.T,trainA)),trainA.T),trainB)
plt.figure(1)
plt.xlabel('x')
plt.ylabel('y')
plt.title('data')
plt.plot(trainA[:,1],trainB,"r*")
y=np.dot(trainA, theta)
print(pow(sum((y-trainB)**2),1/2)/140) #print MSE
y=np.dot(testA, theta)
#plt.plot(testA[:,1], testB, "r.")
plt.plot(testA[:,1],y,"k*")
print(pow(sum((y-testB)**2),1/2)/60) #print MSE
plt.show()
print(theta)
def main():
# Get data
training_set, test_set = getData()
# predict
prediction = modeling(trainx, trainy, testx, testy).predict(testx)
# Transform
prediction_i = scaler.inverse_transform(prediction.reshape(-1, 1))
test_y_inv = scaler.inverse_transform(testy.reshape(-1, 1))
prediction2_i = np.array(prediction_i[:, 0][1:])
test2_y_inv = np.array(test_y_inv[:, 0])
# Train
_, predictions = flow(data,
split,
modeling,
rmse,
n_train=600,
n_test=60)
# Get dif
mean, _ = validate(data, split, modeling, rmse, flow)
# Show chart
getChart(test_Dates, test2_y_inv, predictions - mean)
def bestMatch(data): #finds other jobs for which your personality is a match
jobnames = data.titles
currAvg = data.avg
totalComp = []
betterJobs = []
for i in range(len(data.textJobnames)):
comparisons = 0
jobname = jobnames[data.z + i]
# print(jobname)
txt = findJob(jobname)
callPersonalityInsights(txt, jobname)
a, b, c = getData(jobname + ".json")
things = [a[0], a[1], b]
for ele in range(len(things)):
#print(candidateData[ele])
match, res = Comparing(data.candidatedata[ele], things[ele])
comparisons += res
# print(ele)
totalAvg = comparisons / len(things)
totalComp += [(jobname, totalAvg)]
for element in totalComp:
name, avg = element
if currAvg < avg:
betterJobs += [element]
data.bestJobs = betterJobs
def train(f_hr=8, b_hr=3, thr=40, method="HCR"):
p = DATA_PATH
# Set logger
logger = generateLogger(p + "log.log")
log(
"--------------------------------------------------------------------------",
logger)
log(
"--------------------------------- Train --------------------------------",
logger)
# Get data
end_dt = datetime.now() - timedelta(hours=24)
start_dt = end_dt - timedelta(hours=8000)
log("Get data from " + str(start_dt) + " to " + str(end_dt), logger)
df_esdr_array_raw, df_smell_raw = getData(start_dt=start_dt,
end_dt=end_dt,
logger=logger)
df_esdr, df_smell = preprocessData(df_esdr_array_raw=df_esdr_array_raw,
df_smell_raw=df_smell_raw,
logger=logger)
# Compute features
df_X, df_Y, df_C = computeFeatures(df_esdr=df_esdr,
df_smell=df_smell,
f_hr=f_hr,
b_hr=b_hr,
thr=thr,
is_regr=False,
add_inter=False,
add_roll=False,
add_diff=False,
logger=logger,
out_p_mean=p + "mean.csv",
out_p_std=p + "std.csv")
# Select features
# NOTE: currently, the best model uses all the features
#df_X, df_Y = selectFeatures(df_X, df_Y, logger=logger, out_p=p+"feat_selected.csv")
# Train, save, and evaluate model
model = trainModel({
"X": df_X,
"Y": df_Y,
"C": df_C
},
method=method,
out_p=p + "model.pkl",
logger=logger)
metric = computeMetric(df_Y, model.predict(df_X, df_C), False)
for m in metric:
log(metric[m], logger)
def GET(self, name):
data = ''
i = web.input()
user = i.get('user')
if user == "explorer":
latitude = i.get("latitude")
longitude = i.get("longitude")
data = getData(latitude, longitude)
else:
data = "Name Error"
if data == None:
data = "No Data"
return data
def getProportions(location):
flatten = lambda l: [item for sublist in l for item in sublist]
fullData = getData(location)
newData = []
for x in range(len(fullData)):
if x not in exclusions:
newData.append(fullData[x])
data = flatten(newData)
proportions = {
-1: data.count('-1'),
0: data.count('0'),
1: data.count('1')
}
proportions['total'] = proportions[-1] + proportions[0] + proportions[1]
return proportions
def GET(self,name):
data = ''
i = web.input()
user = i.get('user')
if user == "explorer":
latitude = i.get("latitude")
longitude = i.get("longitude")
data = getData(latitude, longitude)
else:
data = "Name Error"
if data == None:
data = "No Data"
return data
def get_data():
data = ast.literal_eval(getData()["GPSData:"])
text_data = " "
for i in range(len(data)):
if i % 2 == 1:
text_data += data[i] + " "
else:
text_data += data[i]
with open('current_log.txt', 'a') as fi:
fi.write(text_data + "\n")
with open('log.txt', 'a') as f:
f.write(text_data + "\n")
with open('take_screenshot.pickle', 'wb') as f:
pickle.dump('True', f)
return text_data
async def i(ctx, *args):
errorColor=0xed4337
mainColor=0xfde107
if len(args) > 1:
embed = embedFunc("Hata","Lütfen tek kelime giriniz.",errorColor)
elif len(args) == 0:
embed = embedFunc("Hata", "Lütfen kelime giriniz",errorColor)
else:
title, gif, description = getData(args[0])
if title != "0":
embed = embedFunc(title,description,mainColor)
embed.set_image(url=gif)
embed.set_footer(text="isaretce.com")
else:
title="Kelime Bulunamadı"
embed = embedFunc(title,args[0],errorColor)
await ctx.send(embed=embed)
def analyzeStimulus(location):
stimulusData = getData(location)
def analyzeNeurons(data, neuronNum1, neuronNum2, delay):
neuron1 = data[neuronNum1]
neuron2 = data[neuronNum2]
if delay > 0:
neuron1 = neuron1[delay:]
neuron2 = neuron2[:-delay]
elif delay < 0:
neuron1 = neuron1[:delay]
neuron2 = neuron2[-delay:]
neuron1 = map(float, neuron1)
neuron2 = map(float, neuron2)
correlationVal = scipy.stats.pearsonr(neuron1, neuron2)[0]
tempCorVals = 0
tempNeuron2 = neuron2[:]
lessThanCounter = 0.0
return [neuronNum1 + 1, neuronNum2 + 1, delay, correlationVal]
analyzedNeurons = []
for combo in neuronCombinations:
for delay in range(-3, 4):
analyzedNeurons.append(
analyzeNeurons(stimulusData, combo[0], combo[1], delay))
for x in range(len(analyzedNeurons)):
if math.isnan(analyzedNeurons[x][-1]):
analyzedNeurons[x][-1] = 0
analyzedNeuronsSorted = sorted(analyzedNeurons, key=lambda cor: cor[3])
highPositiveCorrelations = analyzedNeuronsSorted[len(analyzedNeuronsSorted
) * 95 / 100:]
highNegativeCorrelations = analyzedNeuronsSorted[
0:len(analyzedNeuronsSorted) * 5 / 100]
return highPositiveCorrelations, highNegativeCorrelations
def __init__(self, params):
"""
Constructor method of the class which initializes variables and calls function to read image names and labels
"""
self._data_file_path = params['data_file_path']
self._num_of_artists = params['num_of_artists']
self._num_of_images_per_artist = params['num_of_images_per_artist']
self._saveData_to_file = params['saveData_to_file']
self._read_prefetched_Data = params['read_prefetched_Data']
self._batch_size = params['batch_size']
self._data_dir = params['data_dir']
self._height = 256
self._width = 256
self._left_image = {}
self._right_image = {}
self._labels = {}
self._params = {}
self._trainFlag = True
# Object for the getData class which reads filenames and labels from csv/text file. Defined in the getData.py
_csv_obj = getData()
# Fetches the training data either from csv or text file.
if not self._read_prefetched_Data:
print('Loading file names and labels from csv.')
self.data, self.label = _csv_obj.read_csv(
self._data_file_path, self._data_dir, self._num_of_artists,
self._num_of_images_per_artist, self._saveData_to_file)
else:
print('Loading existing file names and labels from text file.')
self.data, self.label = _csv_obj.read_from_file(
self._data_file_path)
# Splits the training data into training and validation sets (80-20)
self.train_data = self.data[:int(len(self.data) * 0.8)]
self.train_labels = self.label[:int(len(self.label) * 0.8)]
self.valid_data = self.data[int(len(self.data) * 0.8):]
self.valid_labels = self.label[int(len(self.label) * 0.8):]
matEps=[lamb,eps.real,eps.imag]
return matEps
import matplotlib.pyplot as plt
#Plotting Ag
#hag=getData("database/main/Ag/Hagemann.yml");
range=(0.280,0.790)
lambdasAg=np.linspace(range[0],range[1],100)
ag=getData("database/main/Ag/Johnson.yml",lambdasAg);
#print ag
lambdasTiO2=np.linspace(0.280,0.790)
#tio2=getData("database/main/TiO2/Devore-o.yml",lambdasTiO2);
tio2=getData("database/main/SiO2/Malitson.yml",lambdasTiO2);
fig=plt.figure()
plt.plot(lambdasTiO2*1e3,[ x.real for x in tio2],'g-',label=r'Real($n_{SiO_{2}}$)')
plt.legend()
plt.title(r"Współczynnik załamania $SiO_2$ ($n$)");
plt.ylabel('')
plt.xlabel('wavelength [nm]')
# @Time : 2020/9/3 11:23
# @Author : Jianyuan Lu
# @FileName: Homework2.1.py
# @Software: PyCharm
import numpy as np
import math
import matplotlib.pyplot as plt
import scipy.io as spio
import scipy.sparse.linalg as ll
import sklearn.preprocessing as skpp
from getData import *
A = getData('../data/food-consumption.csv')
stdA = np.std(A, axis=0) #沿行计算标准差
stdA = skpp.normalize(stdA.reshape(
1, -1)) # 使得所有标准差的平方和等于1 the normalize is different from MATLAB's
stdA2 = np.std(A, axis=1) #沿列计算标准差
stdA2 = skpp.normalize(stdA2.reshape(1, -1))
meanA = (np.mean(A, axis=0))
mean2A = (np.mean(A, axis=1))
temp = A - meanA
temp2 = (A.T - mean2A)
Anew = (A - meanA) @ np.diag(
np.ones(stdA.shape[1]) / stdA[0]) #Anew每个元素除以对应的标准差
Anew2 = temp2 @ np.diag(np.ones(stdA2.shape[1]) / stdA2[0]) #Anew每个元素除以对应的标准差
Anew = Anew.T #归一化，转置后的A
def drawResult(canvas, data): #RESULT PAGE 1
canvas.create_rectangle(0, 0, data.width, data.height, fill="slategray4")
margin = 50
canvas.create_text(data.width // 2,
margin,
text="RESULTS",
font="Times 20 bold")
personality, needs, values = getData(
"comparison.json") #data.candidate +".json"
first, second = personality[0], personality[1]
data.comparisondata = []
comparisonData = {}
things = [first, second, needs]
for dictionary in things:
comphell = {}
for key in dictionary:
comphell[key] = dictionary[key]
comparisonData[key] = dictionary[key]
data.comparisondata += [comphell]
personalitycomp, needscomp, valuescomp = getData("candidate.json")
firstcomp, secondcomp = personalitycomp[0], personalitycomp[1]
things1 = [firstcomp, secondcomp, needscomp]
data.candidatedata = []
candidateData = {}
for dictionary1 in things1:
candhell = {}
for key1 in dictionary1:
candhell[key1] = dictionary1[key1]
candidateData[key1] = dictionary1[key1]
data.candidatedata += [candhell]
match, data.avg = Comparing(candidateData, comparisonData)
canvas.create_text(data.width // 2,
3 * margin,
text="You are a %s match to the job" % data.avg,
font="Times 15")
margin = 30
border = 25
width = 4 * margin
centeringx1, centeringy = 1.7, 1.5
canvas.create_rectangle(data.width - width,
data.height - 2 * margin,
data.width - border,
data.height - border,
fill="pink",
outline="white")
canvas.create_text(data.width - width / centeringx1,
data.height - centeringy * margin,
text="DETAILS")
data.result1bounds = (data.width - width, data.height - 2 * margin,
data.width - border, data.height - border)
data.candidateList = []
for dictionary1 in things1:
candidatedata = {}
for key1 in dictionary1:
candidatedata[key1] = dictionary1[key1]
data.candidateList += [[candidatedata]]
canvas.create_text(
data.width // 2,
4 * margin,
text=
"To conside similar and better suited jobs press 'r', your match to the job is also displayed:"
)
i = 0
if data.bestMatch:
for ele in data.bestJobs:
name, match = ele
name = name[:name.index("cover") - 1]
if match >= 100:
canvas.create_text(data.width // 2, (8 + i) * margin,
text=str(name) + " : " + str(match),
fill="red",
font="Times 15 ")
else:
canvas.create_text(data.width // 2, (8 + i) * margin,
text=str(name) + " : " + str(match),
fill="black",
font="Times 15 ")
i += 1
def predict(f_hr=8, b_hr=3, thr=40):
p = DATA_PATH
# Set logger
logger = generateLogger(p + "log.log")
log(
"--------------------------------------------------------------------------",
logger)
log(
"-------------------------------- Predict -------------------------------",
logger)
# Get data for previous b_hr hours
end_dt = datetime.now()
start_dt = end_dt - timedelta(hours=b_hr + 1)
log("Get data from " + str(start_dt) + " to " + str(end_dt), logger)
df_esdr_array_raw, df_smell_raw = getData(start_dt=start_dt,
end_dt=end_dt,
logger=logger)
df_esdr, df_smell = preprocessData(df_esdr_array_raw=df_esdr_array_raw,
df_smell_raw=df_smell_raw,
logger=logger)
if len(df_esdr) < b_hr + 1:
log("ERROR: Length of esdr is less than " + str(b_hr + 1) + " hours",
logger)
log("Length of esdr = " + str(len(df_esdr)), logger)
return
# Compute features
df_X, _, df_C = computeFeatures(df_esdr=df_esdr,
df_smell=df_smell,
f_hr=f_hr,
b_hr=b_hr,
thr=thr,
is_regr=False,
add_inter=False,
add_roll=False,
add_diff=False,
logger=logger,
in_p_mean=p + "mean.csv",
in_p_std=p + "std.csv")
if len(df_X) != 1:
log("ERROR: Length of X is not 1", logger)
log("Length of X = " + str(len(df_X)), logger)
return
# Select features
# NOTE: currently, the best model uses all the features
#df_feat_selected = pd.read_csv(p+"feat_selected.csv")
#df_X = df_X[df_feat_selected.columns]
# Load model
log("Load model...", logger)
model = joblib.load(p + "model.pkl")
# Predict result
# For the hybrid crowd classifier
# if pred==0, no event
# if pred==1, event predicted by the base estimator
# if pred==2, event detected by the crowd
# if pred==3, event both predicted by the base estimator and detected by the crowd
y_pred = model.predict(df_X, df_C)[0]
log("Prediction for " + str(end_dt) + " is " + str(y_pred), logger)
if y_pred == 1 or y_pred == 3: pushType1(end_dt, logger)
if y_pred == 2 or y_pred == 3: pushType2(end_dt, logger)
#!/usr/bin/python
# -*- coding: utf-8 -*-
#
#Author: Pengfei Shi <*****@*****.**>
#
from getData import *
from usingNeo4j import *
from _ast import List
#import getData
if __name__=="__main__":
url1="http://www.sjtu.edu.cn/xbdh/yjdh/gk/jgsz/jgbc.htm"
s=getData(url1)
s.getPages()
s.getContext()
department=["党务部门","行政部门","直属单位","附属医院","附属学校","学院（系）","研究院"]
# list=s.re(r'(?<=#8a9046">).*(?=</font>)')
list0=s.re(r'(?<="c51084">).*(?=</a>)')
list1=s.re(r'(?<="c51085">).*(?=</a>)')
list2=s.re(r'(?<="c51273">).*(?=</a>)')
list3=s.re(r'(?<="c51244">).*(?=</a>)')
list4=s.re(r'(?<="c51245">).*(?=</a>)')
from sklearn import datasets
# Q1
# [x,y]=getData("https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data")
# x = x[0:100,0]
# y = y[0:100]
# clf = LDA()
# clf.fit(x,y)
# LDA(n_components=None, priors=None, shrinkage=None, solver='svd',
# store_covariance=False, tol=0.0001)
# for i in range(100):
# y_pre = clf.predict([x[i]])
# print(y_pre)
# Q2
[x,y]=getData("https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data")
x = x[0:100]
y = y[0:100]
sum = np.zeros((4,1))
for i in range(4):
for j in range(50):
sum[i] = sum[i] + x[j][i]
sum = sum/50 #calculate mean
std = np.std(x[0:50],axis=0)
x = x[50:100]
sum1 = np.zeros((4,1))
for i in range(4):
for j in range(50):
sum1[i] = sum1[i] + x[j][i]
sum1 = sum1/50 #calculate mean
def analyzeStimulus(location): stimulusData = getData(location) def analyzeNeurons(data,neuronNum1,neuronNum2,delay): neuron1=data[neuronNum1] neuron2=data[neuronNum2] if delay>0: neuron1=neuron1[delay:] neuron2=neuron2[:-delay] elif delay<0: neuron1=neuron1[:delay] neuron2=neuron2[-delay:] neuron1 = map(float, neuron1) neuron2 = map(float, neuron2) def findSimilarityCorrelation(neuron1,neuron2): sims=0. total=0. for x in range(len(neuron1)): if neuron1[x]==neuron2[x]: sims+=1. total+=1. return sims/total correlationVal=findSimilarityCorrelation(neuron1,neuron2) tempCorVals=0 tempNeuron2=neuron2[:] lessThanCounter=0.0 if math.isnan(correlationVal): percentile=correlationVal significance=0 else: for x in range(number_of_shuffles): random.shuffle(tempNeuron2) tempCorVal=findSimilarityCorrelation(neuron1,tempNeuron2) if tempCorVal<=correlationVal: lessThanCounter+=1.0 percentile=lessThanCounter/number_of_shuffles if percentile>.95: significance=1 else: significance=0 return [neuronNum1+1, neuronNum2+1, delay, correlationVal, percentile, significance] analyzedNeurons=[] analyzedNeurons.append(['neuron1','neuron2','delay','correlation coefficient','percentile','significance']) for combo in neuronCombinations: for delay in range(-3,4): analyzedNeurons.append(analyzeNeurons(stimulusData,combo[0],combo[1],delay)) return analyzedNeurons
def linearRegreMul(url):
[a,b] = getData(url)
theta = np.dot(np.dot(np.linalg.inv(np.dot(a.T,a)),a.T),b)
y=np.dot(a,theta)
print(pow(sum((y-b)**2),1/2)) #print MSE
rho = rho * dens
return(rho)
radius=10
dist='40.240'
path="two_colloid_dist_" + str(dist)
filename="two_colloid_dist_" + str(dist) + "/ipbs_solution.vtu"
dist=float(dist)
xmin=-dist
xmax=0
ymin=0
ymax=dist/2.
resolution=.125
x,y,sol=getData(filename, xmin, xmax, ymin, ymax, resolution, resolution)
print x.shape, y.shape, sol.shape
phi = scipy.interpolate.RectBivariateSpline(x,y,sol, kx=3, ky=3)
print phi(2,3)
numpy.savetxt("sol.txt", sol.transpose())
#rho=scipy.interpolate.interp2d(x,y,dens(sol), kind='cubic')
#rho=scipy.interpolate.RectBivariateSpline(x,y,sol)
#x = numpy.linspace(xmin,xmax,100)# x coordinates
#y = numpy.linspace(ymin,ymax,100)# y coordinates
#z = solutiondata(x,y)
#print z
print phi(x,y).shape
from scipy import sparse
from scipy.sparse import find
from sklearn.cluster import KMeans
from matplotlib import pyplot as plt
from getData import *
from filterIsolatedDate import *
from getWebName import *
from calMisrate import *
n=1490
largest_K =101
edgeData = getData('../data/edges.txt')
p1 = edgeData[:, 0] - 1 # 矩阵是从0开始的
p2 = edgeData[:, 1] - 1
v = np.ones((edgeData.shape[0], 1)).flatten() # 生成一堆1，准备存放
#A = sparse.csc_matrix((v, (p1, p2)), shape=(n, n))
temp = np.zeros((n, n)).astype(int)
temp[p1,p2] = 1
A = temp
A = (A + np.transpose(A)) / 2 # 连接的地方都是0.5
webName,orientation = getwebname('../data/nodes.txt')
A, zeroIndex,webName,orientation = filterData(A,webName,orientation)
n=A.shape[0]
temp = np.sum(A, axis=1)
D = np.diag(1 / np.sqrt(np.sum(A, axis=1))) # 不用这个.A1，就生成不了矩阵
N = n[:] + 1j * k[:]
eps = N[:] * N[:]
matEps = [lamb, eps.real, eps.imag]
return matEps
import matplotlib.pyplot as plt
# Plotting Ag
# hag=getData("database/main/Ag/Hagemann.yml");
range = (0.280, 0.790)
lambdasAg = np.linspace(range[0], range[1], 100)
ag = getData("database/main/Ag/Johnson.yml", lambdasAg)
# print ag
lambdasTiO2 = np.linspace(0.280, 0.790)
# tio2=getData("database/main/TiO2/Devore-o.yml",lambdasTiO2);
tio2 = getData("database/main/SiO2/Malitson.yml", lambdasTiO2)
fig = plt.figure()
plt.plot(lambdasTiO2 * 1e3, [x.real for x in tio2], "g-", label=r"Real($n_{SiO_{2}}$)")
plt.legend()
plt.title(r"Współczynnik załamania $SiO_2$ ($n$)")
plt.ylabel("")
plt.xlabel("wavelength [nm]")
plt.xlim([280, 750])
#!/usr/bin/python
# -*- coding: utf-8 -*-
from __future__ import unicode_literals
import numpy as np
import pylab as plt
from getData import *
import matplotlib
matplotlib.rcParams.update({'font.size': 30})
range=(1,17)
lambdas=np.linspace(range[0],range[1],100)
ag=getData("database/main/GaAs/Skauli.yml",lambdas);
fig=plt.figure()
plt.plot(lambdas,[ (x*x).real for x in ag ],'r-',label=r'Re($\varepsilon_{GaAs}$)',lw=4)
#plt.plot(lambdas,[ (x*x).imag for x in ag] ,'b-',label=r'Imag($\varepsilon_{Ag}$)')
#plt.plot(lambdasTiO2*1e3,[ (x*x).real for x in tio2],'g-',label=r'Real($\varepsilon_{TiO_{2}}$)')
c=299792458
lambdas=lambdas*10e-6
freq=c/lambdas
freq=freq/10e12
#plt.plot(freq,[ (x*x).real for x in ag ],'r-',label=r'Real($\varepsilon_{Ag}$)')
plt.legend()
#plt.title(r"Współczynnik przenikalnośći elektrycznej ($\varepsilon$)");
print "\t\t "+sp.join(mailto[3*n+2])+"\n\n\n"
found=1; break
if found==0:
print "Mail to:"
print "\t\tUnable to locate address for "+cLoc+"\n\n\n"
print "-------------------------------------------------------------------"
print " 'exit' or 'quit' to escape"
#%%%%
#MAIN
#%%%%%
lu=""
mailto = getData("./MOD/prison_add.dict",1)
while lu != "exit" and lu != "Exit" and lu != "quit" and lu != "Quit" and lu !="q" and lu != "Q":
i_no=raw_input("\n\n\n\n\nPrisoner ID:")
run_query(i_no)
lu=raw_input()
#print "\n\n\n\n\nPrisoner ID:"
#!/usr/bin/python
# -*- coding: utf-8 -*-
from __future__ import unicode_literals
import numpy as np
import pylab as plt
from getData import *
import matplotlib
matplotlib.rcParams.update({'font.size': 30})
range=(1,280)
lambdas=np.linspace(range[0],range[1],100)
ag=getData("database/main/Au/Ordal.yml",lambdas);
fig=plt.figure()
plt.plot(lambdas,[ -(x*x).real for x in ag ],'r-',label=r'-Re($\varepsilon_{Au}$)',lw=4)
plt.plot(lambdas,[ (x*x).imag for x in ag] ,'b-',label=r'Im($\varepsilon_{Au}$)',lw=4)
#plt.plot(lambdasTiO2*1e3,[ (x*x).real for x in tio2],'g-',label=r'Real($\varepsilon_{TiO_{2}}$)')
c=299792458
lambdas=lambdas*10e-6
freq=c/lambdas
freq=freq/10e12
#plt.plot(freq,[ (x*x).real for x in ag ],'r-',label=r'Real($\varepsilon_{Ag}$)')
plt.legend(loc=4)
#plt.title(r"Współczynnik przenikalnośći elektrycznej ($\varepsilon$)");
plt.xlabel('wavelength [um]')
def run():
print('available countries are: ' + displayCountries(getCountries()) )
country = (input('Select a country: '))
status = input('Enter Status number (1 : (confirmed) , 2 : (recovered) , 3 : (deaths): ')
possibleProvinceCity = (searchProvince(getData(country)))# searches for possible provinces and/or cities tied to the country
possibleProvinces = ', '.join(list(possibleProvinceCity.keys()))
if possibleProvinces:# makes additional inputs for province and/or cities if needed
print('Available provinces are: ' + possibleProvinces)
province = (input('Enter a province: ').title())
possibleCities = ', '.join(list(possibleProvinceCity[province]))
if possibleCities:
print('Available cities are: ' + possibleCities)
city = input('Enter a City')
else:
city = ''
elif possibleProvinceCity['']:
possibleCities = ', '.join(list(possibleProvinceCity[''].values()))
print('Available cities are: ' + possibleCities)
city = input('Enter a City')
else:
province = ''
city = ''
if province:# confirms input
if city:
print('Processing input of : Country: ' + country + ', Province: ' + province + ', City: ' + city)
elif not city:
print('Processing input of : Country: ' + country + ', Province: ' + province )
else:
print('Processing input of: ' +country)
#sets the data
data = machineLearning(parseData(getData(country,status),province, city))
if data['daysPredicted']and data['type'] == 'confirmed':
if data['i'] != LIMIT:
print('expected resolution on day '+ str(data['daysPredicted'][-1]))
x1 = (data['days'])
x2 = data['daysPredicted']
y1 = (data['totalCases'])
y2 = (data['casesPerDay'])
y11 = (data['perDayPredicted'])
y22 = (data['totalPredicted'])
# plot the data
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
if data['i'] != LIMIT:
ax.plot(x2, y22,'--', color='tab:blue', label = 'predicted total cases')
ax.plot(x2, y11,'--', color='tab:orange', label = 'predicted cases per day')
plt.scatter(x1, y1,s= 1, color='tab:blue', label = 'total cases')
plt.scatter(x1, y2,s = 1, color='tab:orange', label = 'cases per day')
ax.set_xlabel('number of days')
ax.set_ylabel(('cases of status : ' + data['type']))
ax.legend()
ax.set_title('# of cases of status: ' + data['type'] + ', with respect to time, starting from ' + data['day1'] +' in '+ data['province'] + data['city'] + ', ' + data['country'])
# display the plot
return plt.show()
b = np.array([np.sum(weights * y), np.sum(weights * y * x)])
A = np.array([[np.sum(weights), np.sum(weights * x)],
[np.sum(weights * x), np.sum(weights * x * x)]])
beta = linalg.solve(A, b)
yest[i] = beta[0] + beta[1] * x[i]
residuals = y - yest
s = np.median(np.abs(residuals))
delta = np.clip(residuals / (6.0 * s), -1, 1)
delta = (1 - delta ** 2) ** 2
return yest
if __name__ == '__main__':
import math
n = 200
[x, y] = getData("http://www.cs.iit.edu/~agam/cs584/data/regression/svar-set4.dat")
x = x.T[0]
y = y.T[0]
#x = np.linspace(0, 2 * math.pi, n)
#y = np.sin(x) + 0.3 * np.random.randn(n)
f = 0.25
yest = lowess(x, y, f=f, iter=3)
import pylab as pl
pl.clf()
pl.plot(x, y, '.',label='y noisy')
pl.plot(x, yest, '.',label='y pred')
pl.legend()
pl.show()
comRange = np.empty(shape=(2, 1))
comRange[0] = matRange1[0] if matRange1[0] > matRange2[0] else matRange2[0]
comRange[1] = matRange1[1] if matRange1[1] < matRange2[1] else matRange2[1]
sys.stderr.write("comRange=" + str(comRange))
#Sampling definition:
lambdas = np.linspace(comRange[0], comRange[1], 100)
#alternativelly hard specified
lambdas = np.linspace(7, 10, 100)
#Change refractive index to epsilon
matData1 = getData(file1, lambdas)
for i in range(0, len(matData1)):
matData1[i] = matData1[i] * matData1[i]
matData2 = getData(file2, lambdas)
for i in range(0, len(matData2)):
matData2[i] = matData2[i] * matData2[i]
R1 = np.empty(shape=(0, len(lambdas)))
I1 = np.empty(shape=(0, len(lambdas)))
R2 = np.empty(shape=(0, len(lambdas)))
I2 = np.empty(shape=(0, len(lambdas)))
for f in np.linspace(0, 1, 100):
(efw1, efw2) = effEpsilon(matData1, matData2, f)
def searchRef(yamlFile,bookName,nmin,nmax):
#List for result
result=[]
#Function used only internally to display material parameters values
#ATTENTION!!! comment is not only a comment is like command for this function (yes, it's dirty ;))
# def printParams(paramsArray,paramList,comment):
# for param in paramList:
# sys.stdout.write(str(paramsArray[-1][param])+" ");
# sys.stdout.write(comment+"\n");
# if comment.find("start_range")!=-1:
# printParams.lastStart=paramsArray[-1]["l"]
# elif comment.find("end_range")!=-1:
# result.append((printParams.lastStart,paramsArray[-1]["l"]))
#printParams.lastStart=0;
#Just some boolen, if the firs element is OK it may not be shown
#it's defined only for first material for the rest it's undefined at the begining
print_in_my_range=False;
#This parameter decides additional output about parameters inside range have been met
my_range=False;
yamlStream=open(yamlFile,'r')
allData=yaml.load(yamlStream);
materialData=allData["DATA"][0]
#dataRange - where we can check the data
dataRange=getRange(yamlFile)
if dataRange[0]==0:
return []
#wavelengths to check for this material:
#HERE WE DEFINE THE SAMPLING
lambdas=np.linspace(dataRange[0],dataRange[1],100)
#refractive indx
nList=np.empty((1,0))
# try:
nList=getData(yamlFile,lambdas[:])
if len(nList) == 1:
return []
eps=np.zeros(len(nList),dtype="complex")
for i in range(0,len(nList)):
eps[i]=nList[i]*nList[i]
# except UnsupportedDataType as e:
# sys.stderr.write(e)
if len(nList)<100:
return result
inRange=False
rangeStart=0
for i in range(0,len(lambdas)):
if eps[i].real>nmin and eps[i].real<nmax:
if inRange==False:
inRange=True
rangeStart=lambdas[i]
elif inRange:
result.append((rangeStart,lambdas[i]))
inRange=False
return result
,x.strip().split(','))
,data)
return data
def writeFile(location,data):
f = open(location+".txt", "w")
f.write(str(data)[1:-1])
f.close()
fullData=[]
temp=[]
for stimulus in stimuliNames:
temp=getData(outputDirectory + "exportingN/" + stimulus)
tempPositive = map(lambda x:
map(lambda y: 0 if y==-1 else y,x),
temp)
tempNegative = map(lambda x:
map(lambda y: 0 if y==1 else y,x),
temp)
newDataPositive=map(lambda x: sum(x) / len(x), (np.transpose(tempPositive)).tolist())
newDataNegative=map(lambda x: -1* sum(x) / len(x), (np.transpose(tempNegative)).tolist())
writeFile(outputDirectory + "histogramData/" + stimulus + "Positive", newDataPositive)
writeFile(outputDirectory + "histogramData/" + stimulus + "Negative", newDataNegative)
preDataFrameDict["instrumentalness"].append(
playlistSongFeatures[i]["instrumentalness"])
preDataFrameDict["liveness"].append(
playlistSongFeatures[i]["liveness"])
preDataFrameDict["loudness"].append(
playlistSongFeatures[i]["loudness"])
preDataFrameDict["speechiness"].append(
playlistSongFeatures[i]["speechiness"])
preDataFrameDict["tempo"].append(playlistSongFeatures[i]["tempo"])
preDataFrameDict["valence"].append(playlistSongFeatures[i]["valence"])
df = pd.DataFrame(preDataFrameDict)
return df.set_index("Song")
#getting data to play with
playlistIDs, playlistSongs, playlistSongFeatures = getData(userID, accessToken)
#example of basic visual
#DeathflowFrame = playlistToDataFrame("Deathflow", playlistSongs["Deathflow"], playlistSongFeatures["Deathflow"])
#DeathflowFrame[["acousticness","danceability","energy"]].plot.area()
#plt.show()
SavedTracksFrame = playlistToDataFrame("savedTracks",
playlistSongs["savedTracks"],
playlistSongFeatures["savedTracks"])
SavedTracksFrame[["acousticness", "danceability", "energy"]].plot.area()
plt.show()
# using KNN to determine which playlist a song belongs in
# to do this i will average the song features of each playlist dataframe and construct
# a new dataframe where the index is the playlist and the features are the averaged features of the
from getData import *
import pybrain as pb
from pybrain.tools.shortcuts import buildNetwork
from pybrain.datasets import SupervisedDataSet
from pybrain.supervised.trainers import BackpropTrainer
trainData = getData('data_2.csv');
testData = getData('data_1_2.csv');
#training
fnn = buildNetwork(24, 30, 8, 1, bias=True);
ds = SupervisedDataSet(24, 1);
for i in trainData:
ds.addSample([i[j] for j in Name[1:]], i[Name[0]]);
trainer = BackpropTrainer(fnn, ds);
print "Begin....";
trainer.trainEpochs(epochs=100);
print "Ing...";
out = SupervisedDataSet(24, 1);
for i in testData:
out.addSample([i[j] for j in Name[1:]], [0]);
res = fnn.activateOnDataset(out);
for i, o in out:
print i, "->", o;
print res
comRange = np.empty(shape=(2, 1))
comRange[0] = matRange1[0] if matRange1[0] > matRange2[0] else matRange2[0]
comRange[1] = matRange1[1] if matRange1[1] < matRange2[1] else matRange2[1]
sys.stderr.write("comRange=" + str(comRange))
# Sampling definition:
lambdas = np.linspace(comRange[0], comRange[1], 100)
# alternativelly hard specified
lambdas = np.linspace(7, 10, 100)
# Change refractive index to epsilon
matData1 = getData(file1, lambdas)
for i in range(0, len(matData1)):
matData1[i] = matData1[i] * matData1[i]
matData2 = getData(file2, lambdas)
for i in range(0, len(matData2)):
matData2[i] = matData2[i] * matData2[i]
R1 = np.empty(shape=(0, len(lambdas)))
I1 = np.empty(shape=(0, len(lambdas)))
R2 = np.empty(shape=(0, len(lambdas)))
I2 = np.empty(shape=(0, len(lambdas)))
for f in np.linspace(0, 1, 100):
(efw1, efw2) = effEpsilon(matData1, matData2, f)
def main(argv):
p = "data_main/"
mode = None
if len(argv) >= 2:
mode = argv[1]
# Parameters
is_regr = False # False for classification, True for regression
smell_thr = 40 # threshold to define a smell event
start_dt = datetime(2016, 10, 31, 0, tzinfo=pytz.timezone("US/Eastern"))
end_dt = datetime(2019, 8, 5, 0, tzinfo=pytz.timezone("US/Eastern"))
# Set mode
get_data, preprocess_data, analyze_data, compute_features, cross_validation = False, False, False, False, False
if mode == "pipeline":
get_data = True
preprocess_data = True
compute_features = True
cross_validation = True
elif mode == "data":
get_data = True
elif mode == "preprocess":
preprocess_data = True
elif mode == "feature":
compute_features = True
elif mode == "validation":
cross_validation = True
elif mode == "analyze":
analyze_data = True
else:
get_data = True
preprocess_data = True
compute_features = True
cross_validation = True
# Get data
# OUTPUT: raw esdr and raw smell data
if get_data:
getData(out_p=[p + "esdr_raw/", p + "smell_raw.csv"],
start_dt=start_dt,
end_dt=end_dt)
# Preprocess data
# INPUT: raw esdr and raw smell data
# OUTPUT: preprocessed esdr and smell data
if preprocess_data:
preprocessData(in_p=[p + "esdr_raw/", p + "smell_raw.csv"],
out_p=[p + "esdr.csv", p + "smell.csv"])
# Analyze data
if analyze_data:
analyzeData(in_p=[p + "esdr.csv", p + "smell.csv"],
out_p_root=p,
start_dt=start_dt,
end_dt=end_dt)
# Compute features
# INPUT: preprocessed esdr and smell data
# OUTPUT: features and labels
if compute_features:
computeFeatures(in_p=[p + "esdr.csv", p + "smell.csv"],
out_p=[p + "X.csv", p + "Y.csv", p + "C.csv"],
is_regr=is_regr,
f_hr=8,
b_hr=3,
thr=smell_thr,
add_inter=False,
add_roll=False,
add_diff=False)
# Cross validation
# INPUT: features
# OUTPUT: plots or metrics
if cross_validation:
#methods = ["ET", "RF", "SVM", "RLR", "LR", "LA", "EN", "MLP", "KN", "DMLP"] # regression
#methods = ["ET", "RF", "SVM", "LG", "MLP", "KN", "DMLP", "HCR", "CR", "DT"] # classification
methods = ["RF", "ET"] # default for random forest and extra trees
#methods = genModelSet(is_regr)
p_log = p + "log/"
if is_regr: p_log += "regression/"
else: p_log += "classification/"
checkAndCreateDir(p_log)
num_folds = (end_dt - start_dt).days / 7 # one fold represents a week
for m in methods:
start_time_str = datetime.now().strftime("%Y-%d-%m-%H%M%S")
lg = generateLogger(p_log + m + "-" + start_time_str + ".log",
format=None)
crossValidation(in_p=[p + "X.csv", p + "Y.csv", p + "C.csv"],
out_p_root=p,
event_thr=smell_thr,
method=m,
is_regr=is_regr,
logger=lg,
num_folds=num_folds,
skip_folds=48,
train_size=8000)
from getData import *
from math import *
import matplotlib.pyplot as plt
def d2p(x):
if x < 90:
x = x;
elif x < 180:
x = 180 - x;
elif x < 270:
x = x - 180;
else:
x = 360 - x;
return x/180*pi
data = getData('data_2.csv');
l = len(data);
x = range(l);
y1 = [i['power'] for i in data];
y2 = [ (i['WS30']*cos(d2p(i['DIRECTION30']))+i['WS31']*cos(d2p(i['DIRECTION31']))+ \
i['WS32']*cos(d2p(i['DIRECTION32']))+i['WS10']*cos(d2p(i['DIR10']))+ \
i['WS10S']*cos(d2p(i['DIR10S'])))**3 for i in data];
print y1;
print y2;
plt.figure(figsize=(16,8));
plt.plot(y2, y1, 'o');
#plt.plot(x, y1, 'r', linewidth=2);
#plt.plot(x, y2, 'b', linewidth=2);
plt.legend()
for n in range(len(mailto) / 3):
if mailto[3 * n][0] == parts[1]:
print "Mail to:"
print "\t\t" + pName.upper() + " - " + i_no
print "\t\t" + cLoc
print "\t\t " + sp.join(mailto[3 * n + 1])
print "\t\t " + sp.join(mailto[3 * n + 2]) + "\n\n\n"
found = 1
break
if found == 0:
print "Mail to:"
print "\t\tUnable to locate address for " + cLoc + "\n\n\n"
print "-------------------------------------------------------------------"
print " 'exit' or 'quit' to escape"
#%%%%
#MAIN
#%%%%%
lu = ""
mailto = getData("./MOD/prison_add.dict", 1)
while lu != "exit" and lu != "Exit" and lu != "quit" and lu != "Quit" and lu != "q" and lu != "Q":
i_no = raw_input("\n\n\n\n\nPrisoner ID:")
run_query(i_no)
lu = raw_input()
#print "\n\n\n\n\nPrisoner ID:"
#!/usr/bin/python
# -*- coding: utf-8 -*-
from __future__ import unicode_literals
import numpy as np
import pylab as plt
from getData import *
import matplotlib
matplotlib.rcParams.update({'font.size': 30,'font.weight':'bold'})
range=(7,10)
lambdas=np.linspace(range[0],range[1],100)
ag=getData("database/main/NaCl/Li.yml",lambdas);
fig=plt.figure()
plt.plot(lambdas,[ (x*x).real for x in ag ],'r-',label=r'Re',lw=4)
plt.plot(lambdas,[ (x*x).imag for x in ag] ,'b--',label=r'Im',lw=4)
#plt.plot(lambdasTiO2*1e3,[ (x*x).real for x in tio2],'g-',label=r'Real($\varepsilon_{TiO_{2}}$)')
c=299792458
lambdas=lambdas*10e-6
freq=c/lambdas
freq=freq/10e12
#plt.plot(freq,[ (x*x).real for x in ag ],'r-',label=r'Real($\varepsilon_{Ag}$)')
plt.legend(loc=1,borderpad=0,frameon=False)
#plt.title(r"Współczynnik przenikalnośći elektrycznej ($\varepsilon$)");
plt.xlabel(r'$\lambda$ [$\mu m$]')