def aggregate(func, columnName, tableName, dictionary):
# print('[FROM AGGREGATE]', dictionary)
if columnName == '*':
sys.exit("[ERROR]: sql query syntax error, aggregate functions not applicable on *")
if columnName not in dictionary[tableName]:
error = "[ERROR]: no column named " + columnName + " found in " + tableName
sys.exit(error)
tName = tableName + '.csv'
fileData = []
readFile(tName, fileData)
colList = []
for data in fileData:
colList.append(int(data[dictionary[tableName].index(columnName)]))
if func.lower() == 'max':
print (max(colList))
elif func.lower() == 'min':
print (min(colList))
elif func.lower() == 'sum':
print (sum(colList))
elif func.lower() == 'avg':
print (sum(colList)/len(colList))
elif func.lower() == 'distinct':
distinct(colList, columnName, tableName, dictionary)
else:
print ("[ERROR]: unknown function :", '"' + func + '"')
def processWhereJoin(whereStr, columnNames, tableNames, dictionary):
tableNames.reverse()
l1 = []
l2 = []
readFile(tableNames[0] + '.csv', l1)
readFile(tableNames[1] + '.csv', l2)
fileData = []
for item1 in l1:
for item2 in l2:
fileData.append(item2 + item1)
dictionary["sample"] = []
for i in dictionary[tableNames[1]]:
dictionary["sample"].append(tableNames[1] + '.' + i)
for i in dictionary[tableNames[0]]:
dictionary["sample"].append(tableNames[0] + '.' + i)
dictionary["test"] = dictionary[tableNames[1]] + dictionary[tableNames[0]]
print (dictionary)
tableNames.remove(tableNames[0])
tableNames.remove(tableNames[0])
tableNames.insert(0, "sample")
if(len(columnNames) == 1 and columnNames[0] == '*'):
columnNames = dictionary[tableNames[0]]
for i in columnNames:
print (i, end='\t')
print ('\n')
a = whereStr.split(" ")
print (a)
check = 0
evalCheck = 0
for data in fileData:
string = evaluate(a, tableNames, dictionary, data)
# print (string)
for col in columnNames:
try:
evalCheck = eval(string)
except:
sys.exit('error')
if evalCheck:
check = 1
if '.' in col:
print (data[dictionary[tableNames[0]].index(col)], end='\t\t')
else:
print (data[dictionary["test"].index(col)], end='\t\t')
if check == 1:
check = 0
print ('\n')
del dictionary['sample']
def downloadAllTweets(start, end, tweetTargetSource):
numRequests = 0
tweetIdSource = 'data/tweets.csv'
dataLines = readFile(tweetIdSource).splitlines()[start:end]
reducedDataLines = sample(dataLines, 400000) #Choosing random sample of 400000\
fieldnames = ['id_str', 'created_at', 'coordinates', 'hashtags', 'text']
setHeaders(tweetTargetSource, fieldnames)
requestStartRow = 0
requestEndRow = 99
while(requestEndRow <= len(reducedDataLines)):
print "Parsing rows " + str(requestStartRow) + '-' + str(requestStartRow + 6000) + ' out of ' + str(len(reducedDataLines))
print str(float(requestStartRow)/float(len(reducedDataLines)) * 100) + "% complete"
#Inner while loop handles API Rate limit logic
while (numRequests < 60):
if (getRateLimit('statuses')['resources']['statuses']['/statuses/lookup']['remaining'] == 0):
break
tweetIdList = parseTweetIds(reducedDataLines, requestStartRow, requestEndRow)
tweetData = getTweetsFromId(tweetIdList)
print "Request: " + str(numRequests)
if (tweetData != None): writeToFile(tweetData, tweetTargetSource, fieldnames)
requestStartRow += 100
requestEndRow += 100
numRequests += 1
print "Rate Limit Exceeded. Waiting...\n"
while (getRateLimit('statuses')['resources']['statuses']['/statuses/lookup']['remaining'] == 0):
sleep(120) #Suspends execution until rate limit refreshed
numRequests = 0
print "Done!"
def selectColumns(columnNames, tableNames, dictionary):
if len(columnNames) == 1 and columnNames[0] == '*':
columnNames = dictionary[tableNames[0]]
for i in columnNames:
if i not in dictionary[tableNames[0]]:
error = "[ERROR]: no column named '" + \
i + "' found in " + tableNames[0]
sys.exit(error)
printHeader(columnNames, tableNames, dictionary)
tName = tableNames[0] + '.csv'
fileData = []
readFile(tName, fileData)
printData(fileData, columnNames, tableNames, dictionary)
def join(columnNames, tableNames, dictionary):
tableNames.reverse()
l1 = []
l2 = []
readFile(tableNames[0] + '.csv', l1)
readFile(tableNames[1] + '.csv', l2)
fileData = []
for item1 in l1:
for item2 in l2:
fileData.append(item2 + item1)
# dictionary["sample"] = dictionary[b] + dictionary[a]
dictionary["sample"] = []
for i in dictionary[tableNames[1]]:
dictionary["sample"].append(tableNames[1] + '.' + i)
for i in dictionary[tableNames[0]]:
dictionary["sample"].append(tableNames[0] + '.' + i)
dictionary["test"] = dictionary[tableNames[1]] + dictionary[tableNames[0]]
# print (dictionary["test"])
tableNames.remove(tableNames[0])
tableNames.remove(tableNames[0])
tableNames.insert(0, "sample")
if (len(columnNames) == 1 and columnNames[0] == '*'):
columnNames = dictionary[tableNames[0]]
# print (header)
for i in columnNames:
print(i, end='\t')
print('\n')
# printData(fileData,columnNames,tableNames,dictionary)
for data in fileData:
for col in columnNames:
if '.' in col:
print(data[dictionary[tableNames[0]].index(col)], end='\t\t')
else:
print(data[dictionary["test"].index(col)], end='\t\t')
print('\n')
def processWhere(whereStr, columnNames, tableNames, dictionary):
a = whereStr.split(" ")
if a[0] not in dictionary[tableNames[0]]:
error = "[ERROR]: unknown column '" + a[0] + "' in where clause"
sys.exit(error)
# print (a)
# print (len(a))
if len(a) > 3 and a[4] not in dictionary[tableNames[0]]:
error = "[ERROR]: unknown column '" + a[4] + "' in where clause"
sys.exit(error)
if(len(columnNames) == 1 and columnNames[0] == '*'):
columnNames = dictionary[tableNames[0]]
printHeader(columnNames, tableNames, dictionary)
tName = tableNames[0] + '.csv'
fileData = []
readFile(tName, fileData)
check = 0
evalCheck = 0
# print (fileData)
for data in fileData:
string = evaluate(a, tableNames, dictionary, data)
for col in columnNames:
# print(col)
try:
evalCheck = eval(string)
except:
sys.exit('error')
if evalCheck:
check = 1
print (data[dictionary[tableNames[0]].index(col)], end="\t\t")
if check == 1:
check = 0
print ('\n')
def main():
"""Prompts user for data set and classifier
Displays percent correct"""
print("Which data would you like to use?")
print("1 - Iris")
print("2 - From file")
option = int(input())
# default to Iris with improper input
if option == 2:
data = rd.readFile()
else:
data = rd.readIris()
data_train, data_test, target_train, target_test = \
train_test_split(data.data, data.target, test_size=.3)
print("Which classifier would you like to use?")
print("1 - GaussianNB")
print("2 - Hard Coded")
print("3 - KNN")
option = int(input())
# default to hard_coded_classifier with improper input
if option == 1:
classifier = GaussianNB()
elif option == 2:
classifier = hc.HardCodedClassifier()
else:
classifier = knn.Classifier()
model = classifier.fit(data_train, target_train)
target_predicted = model.predict(data_test)
# loop through test target and predicted target
# if they are equal, increment number correct
num_correct = 0
for i in range(0, len(target_predicted)):
if target_predicted[i] == target_test[i]:
num_correct += 1
# get percent correct
percent_accurate = round((num_correct / len(target_predicted)) * 100, 2)
print("Classifier is", percent_accurate, "% accurate.")
n1 = eval(input("Enter the size: ",))
import read_data
import ship_file
import string
import random
lst_data = read_data.readFile('data.txt')
lst_of_grid = read_data.read_field(lst_data)
lst_alfabet = [i for i in string.ascii_uppercase]
def is_valid (lst_of_grid):
lst = []
assert isinstance(n1, object)
n1 = 0
n2 = 0
n3 = 0
n4 = 0
if len(lst_of_grid)**(1/2) == n1:
for i in lst_of_grid:
if ship_file.ship_size(lst_of_grid, i[0]) == 1:
lst.append(1)
n1+=1
elif ship_file.ship_size(lst_of_grid, i[0]) == 2:
lst.append(2)
n2+=1
elif ship_file.ship_size(lst_of_grid, i[0]) == 3:
lst.append(3)
n3+=1
elif ship_file.ship_size(lst_of_grid, i[0]) == 4:
lst.append(4)
n4+=1
if lb >= optErrorT:
continue
nodeTrans.ub = ub
nodeTrans.lb = lb
queueTrans.put(nodeTrans)
# print(count)
return optErrorT, T
if __name__ == '__main__':
dir = 'point_cloud_registration'
filenames = ['pointcloud1.fuse', 'pointcloud2.fuse']
names = ['pointcloud1', 'pointcloud2']
pointcloud1 = readFile('{}/{}'.format(dir, filenames[0]), names[0])
pointcloud2 = readFile('{}/{}'.format(dir, filenames[1]), names[1])
print(pointcloud1.shape)
print(pointcloud2.shape)
i1 = None
i2 = None
if pointcloud1.shape[1] == 4:
# store available intensities separately
i1 = pointcloud1[:, 3]
i2 = pointcloud2[:, 3]
else:
# Create homogenous coordinates to redue time of transformations.
pointcloud1 = np.hstack((pointcloud1, np.ones(
(pointcloud1.shape[0], 1))))
pointcloud2 = np.hstack((pointcloud2, np.ones(
