def main(argv):
inputPath = argv[0]
gamma, Lambda = readResults(inputPath)
tf, labels_1, labels_2, topic_texts, dictionary, data = em.loadData(
'reutersdata', 'earn', 'grain', maxDocs=len(gamma), returnData=True)
samples = 0
minWordsFromTopic = 5
numTopics = 4
colors = ['red', 'orange', 'green', 'blue']
gamma = gamma - np.min(gamma)
for docIdx, row in enumerate(gamma):
if np.sum(row >= minWordsFromTopic) == numTopics:
topics = np.where(row >= minWordsFromTopic)
if len(data[docIdx].split()) < 100:
print(data[docIdx])
wd = em.wordCountToFlatArray(tf[docIdx, :])
phiAssigned = np.where(getPhi(inputPath, docIdx) > 0.9)
print("")
topicToColor = dict()
for topicIdx, topic in enumerate(topics[0]):
dictionaryIdxs = Lambda[topic, :].argsort()[-10:]
print("Num words from this topic: %f" % row[topic])
for didx in dictionaryIdxs:
print(dictionary[didx])
print("")
topicToColor[topic] = colors[topicIdx]
wordToColoredWord = dict()
for wordIdx, topicIdx in zip(*phiAssigned):
if topicIdx in topicToColor:
wordToColoredWord[dictionary[
wd[wordIdx]]] = topicToColor[topicIdx]
newText = data[docIdx]
for word, color in wordToColoredWord.iteritems():
newText = re.sub(r'\s(%s)\s' % word,
r' \\textcolor{%s}{\1} ' % color,
newText,
flags=re.I)
print(newText)
print("")
samples += 1
if samples > 10:
break
def loadDataSet():
if request.method == 'POST':
#f = request.files['folder']
f = "dataset"
functions.save_features(f)
dataset = functions.loadData("datasets/dataset.csv")
vals = functions.crossValidation(dataset)
return render_template('precision.html',
nr=int(vals[0][0] * 100),
reg=int(vals[2][0] * 100),
tr=int(vals[1][0] * 100),
data=vals)
else:
return "err"
def savem():
nom = str(request.form["n"])
model = str(request.form["nom"])
dataset = functions.loadData("datasets/dataset.csv")
if (model == "log"):
clasif = functions.trainingLogReg(dataset)
functions.savemodel(clasif, nom)
elif (model == "nr"):
clasif = functions.trainingNeuralNetwork(dataset)
functions.savemodel(clasif, nom)
else:
clasif = functions.trainingDecTrees(dataset)
functions.savemodel(clasif, nom)
return 'v'
def main(argv):
gamma = readResults(argv[0])
tf, labels_1, labels_2, topic_texts, dictionary = em.loadData(
'reutersdata', 'earn', 'grain', maxDocs=8000)
mean_stats_low_dim, sd_stats_low_dim, mean_stats_high_dim, sd_stats_high_dim = DocumentClassification(
gamma, tf, labels_1)
plt.figure()
plotSVMModelAccuracy(mean_stats_low_dim, sd_stats_low_dim,
mean_stats_high_dim, sd_stats_high_dim)
mean_stats_low_dim, sd_stats_low_dim, mean_stats_high_dim, sd_stats_high_dim = DocumentClassification(
gamma, tf, labels_2)
plt.figure()
plotSVMModelAccuracy(mean_stats_low_dim, sd_stats_low_dim,
mean_stats_high_dim, sd_stats_high_dim)
plt.show()
from imblearn.combine import SMOTETomek
from imblearn.under_sampling import RandomUnderSampler
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import confusion_matrix, precision_recall_curve, auc, roc_auc_score, roc_curve, recall_score, classification_report
from sklearn.neighbors import KNeighborsClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import KFold
from sklearn.metrics import precision_recall_curve
from sklearn import decomposition, tree
import functions
#Load dataset
data_set = functions.loadData()
print('Dataset is loaded')
#Select specific features from the entire dataset
selected_features = [
"issuercountrycode", "txvariantcode", "EuroAmount", "currencycode",
"shoppercountrycode", "shopperinteraction", "cardverificationcodesupplied",
"cvcresponsecode", "accountcode"
]
new_data = data_set[selected_features]
#Create dummies dataset
new_data = pd.get_dummies(new_data)
#Split dataset into train and test
X_train, X_test, y_train, y_test = train_test_split(new_data,
data_set['label'],
def main():
user = sys.argv[-1]
for frame in sys.argv[1:-1]:
# Loop over all csv files
os.system("cls")
print("User: "******"File Path: " + str(frame))
csv_frames = fnc.handlefileinput(frame)
match_overview = csv_frames[0]
match_performance = csv_frames[1]
# sixth_pick_overview = csv_frames[2]
player_round_data = csv_frames[3]
# round_event_breakdown = csv_frames[4]
all_players, all_teams = fnc.loadData()
knowndata = fnc.getknowndata(all_players, all_teams)
all_playernames = knowndata[0]
all_teamnames = knowndata[1]
used_gamemode = knowndata[2]
used_match_info = knowndata[3]
input_correct = False
user_input = []
while not input_correct:
user_input = fnc.handleuserinput(knowndata)
print("\n\nMatchdata: " + str(user_input[0]) + " | Remarks:\t" + str(user_input[1]) + "\n")
map_string = ""
for data in user_input[3]:
map_string += "\t\t\t\t" + str(data) + "\n"
op_string = ""
for data in user_input[4]:
op_string += "\t\t\t\t" + str(data) + "\n"
print(
"Team 1: " + str(user_input[2]) + "\n\tBanned:\n\t\tMaps:\n" + map_string + "\n\t\tOps:\n" + op_string)
map_string = ""
for data in user_input[6]:
map_string += "\t\t\t\t" + str(data) + "\n"
op_string = ""
for data in user_input[7]:
op_string += "\t\t\t\t" + str(data) + "\n"
print(
"Team 2: " + str(user_input[5]) + "\n\tBanned:\n\t\tMaps:\n" + map_string + "\n\t\tOps:\n" + op_string)
confirm = input("Input correct?\t Press y to confirm...\n")
if confirm == "y" or confirm == "Y":
input_correct = True
blue_team = user_input[2]
orange_team = user_input[5]
if not (blue_team in all_teamnames):
print("Add new team...")
all_teams.append(ds.Team(blue_team))
if not (orange_team in all_teamnames):
print("Add new team...")
all_teams.append(ds.Team(orange_team))
for team in all_teams:
print("Add team round data...")
if team.name in [blue_team, orange_team]:
try:
team.addMatch(match_overview, player_round_data, user_input, roundoverview)
except NameError:
roundoverview = team.addMatch(match_overview, player_round_data, user_input)
for player_name in match_performance.Player.values:
if not (player_name in all_playernames):
print("Add new player match data...")
all_players.append(ds.Player(player_name))
all_players[len(all_players) - 1].addMatch(player_name, match_overview, match_performance, roundoverview)
else:
for player in all_players:
if player.name == player_name:
print("Add player match data...")
player.addMatch(player_name, match_overview, match_performance, roundoverview)
fnc.saveData(all_players, all_teams)
if random.randint(1, 1001) > 999:
# this ends the progress bar
toolbar_width = 100
# setup toolbar
sys.stdout.write("[%s]" % (" " * toolbar_width))
sys.stdout.flush()
sys.stdout.write("\b" * (toolbar_width + 1)) # return to start of line, after '['
for i in range(toolbar_width):
time.sleep(1 / random.randint(25, 50)) # do real work here
# update the bar
if i == (toolbar_width / 4):
sys.stdout.write("TR")
sys.stdout.flush()
elif (3 * toolbar_width / 4) > i > (toolbar_width / 4):
sys.stdout.write("OL")
sys.stdout.flush()
else:
sys.stdout.write("-")
sys.stdout.flush()
sys.stdout.write("]\n")
print("File Path: " + str(frame))
print("Finished")