def insult():
adjective1 = choice(phrase1)
adjective2 = choice(phrase2)
noun1 = choice(phrase3)
cp("thou art a " + adjective1 + adjective2 + noun1 + "!",
"cyan",
attrs=["bold"])
return "you have been insulted!"
def usernamename():
wantname = input("view username [y/n]: ")
if wantname.lower() == "y":
cp("Your username: "******"cyan", attrs=["bold"])
elif wantname.lower() == "yes":
cp("Your username: "******"cyan", attrs=["bold"])
elif wantname.lower() == "n":
return "USERNAME VIEWER CANCELED!"
elif wantname.lower() == "no":
return "USERNAME VIEWER CANCELED!"
return "Command Complete"
def sessionname():
wantname = input("view session name [y/n]: ")
if wantname.lower() == "y":
cp("Your session name: " + session, "cyan", attrs=["bold"])
elif wantname.lower() == "yes":
cp("Your session name: " + session, "cyan", attrs=["bold"])
elif wantname.lower() == "n":
return "SESSION VIEWER CANCELED!"
elif wantname.lower() == "no":
return "SESSION VIEWER CANCELED!"
return "Command Complete"
def _min_max_scaling(train, test):
"""
using MinMax function for scaling array in range(-1,1)
saving scaler in order to reuse it for precdictions
param:: train:numpy.array, test:numpy.array
return:: train_scaled:numpy.array, test_scaled:numpy.array
s"""
scaler = MinMaxScaler(feature_range=(-1, 1))
scaler = scaler.fit(train)
# reshaping the array so they fit in the scaler
train = train.reshape(train.shape[0], train.shape[1])
test = test.reshape(test.shape[0], test.shape[1])
train_scaled = scaler.transform(train)
print(f'Scaler: {test.shape}')
test_scaled = scaler.transform(test)
# svaing the scaler
try:
save_data_to_(file_name='min-max-scaler.save',
folder_name='saved_model_data',
data=scaler,
is_model=False)
except Exception as save_error:
print(
cp(
f'log[error]:Failed to save data for \'min-max-scaler.save\'. See log file',
'red'))
logging.error('Failed to save data for', exc_info=True)
return train_scaled, test_scaled
def add_color():
cp("pybot: ", "magenta", end="")
if color == attrs == "":
print(bot_response)
elif color == "":
cp(bot_response, "white", attrs=[attrs])
elif attrs == "":
cp(bot_response, color)
else:
cp(bot_response, color, attrs=[attrs])
def _get_key_value_config(key, config):
"""
function returning the value of passed key
param:: key:string
return:: value:string
"""
try:
return config['lstm-script'][key]
except KeyError as key_error:
print(
cp(f'log[error]:Key not in config dictornay. See log file', 'red'))
logging.error('Key not in config dictorna', exc_info=True)
def _load_config_json():
"""
loading the config file in order to get target, feature, date, and groupby value
assigns the constant CONFIG the json as dict{}
"""
try:
with open(CONFIG_PATH, 'r') as conf:
config = json.load(conf)
return config
except Exception as json_error:
print(
cp(
f'log[error]:unable to load preprocessing.config.json file. See log file',
'red'))
logging.error('unable to load preprocessing.config.json file',
exc_info=True)
def stupid():
while True:
stupid_bot_response = choice(stupidanswer)
if color == attrs == "":
print(stupid_bot_response)
elif color == "":
cp(stupid_bot_response, "white", attrs=[attrs])
elif attrs == "":
cp(stupid_bot_response, color)
else:
cp(stupid_bot_response, color, attrs=[attrs])
stupidprompt = input(cd(loginuser + ": ",
"yellow")) # so they get a prompt
if stupidprompt.lower() == "!exitstupid":
return "EXITING STUPID MODE!"
def fake():
while True:
fake_bot_response = choice(fakeanswer)
if color == attrs == "":
print(fake_bot_response)
elif color == "":
cp(fake_bot_response, "white", attrs=[attrs])
elif attrs == "":
cp(fake_bot_response, color)
else:
cp(fake_bot_response, color, attrs=[attrs])
fakeprompt = input(cd(loginuser + ": ",
"yellow")) # so they get a prompt
if fakeprompt.lower() == "!exitfake":
return "EXITING FAKE MODE!"
def save_data_to_(file_name, folder_name, data, is_model):
"""
function to save model and scaler.
Needed to change the current directory
The is_model defindes which saver to use. for scaler and model two different needed
param:: file_name:string, folder_name:string, data:obj, is_model:boolean
return:: -
raise: Execption
"""
try:
os.chdir('infrastructure/ml-models/'.join(folder_name))
if is_model:
data.save(file_name)
else:
joblib.dump(data, file_name)
except Exception as saving_error:
print(
cp(f'log[error]:Failed to save data for {file_name}. See log file',
'red'))
logging.error('Failed to save data for', exc_info=True)
finally:
os.path.dirname(os.path.abspath(__file__))
def prize():
global prize_given
if prize_given:
cp("YOU ALREADY UNBOXED A PRIZE!", "magenta")
else:
prize_response = choice(prizeunbox)
cp(prize_response, "cyan")
cp("PRIZE UNBOXED!", "magenta")
prize_given = True
return "Command Complete"
"haya",
"timchen",
"lmntl dj",
"jimmymcjimface",
"jbrown",
]
staffpassword = [
"7ebf013ae25747d5f9646b5380adf0d1add777f912ebd88f64cd2ca95faeddb81c33107a7e1231a754f40bdc0289135d11cbb17573d2f2724afb1692ac15adb5"
]
semistaffpassword = [
"0913de549bbebb374c0a55eb695a0e6cc556c7db15b2d5c759c7abb06320e6eb3addd75052b9a0d65a43ccf6658e69e28dd60964c36f5e42caa7508744e7b858"
]
cp(ff("""PyBot\n v0.9.4""", font="starwars"), "yellow", attrs=["bold"])
print("\n" * 10)
nickname = input("What Should I Call You: ")
clear()
cp(ff("PyBot\n v0.9.4", font="starwars"), "yellow", attrs=["bold"])
print("\n" * 10)
lang = input("what's your spoken language: ")
if lang.lower() == "":
lang = "english"
def start_processor():
"""
this is the main function with orchastrats the programm
first the data (csv) will be load - as well as the config file. The data frame will be reduced to the data needed.
next the differences of sales and prev-sales are computed followed by the creation of lags needed for the LSTM model
as last step the LSTM model will be saved (both architecture and weights) in a .h5 file for later usage.
In addition the fitted MinMaxScaler will be saved so it can be imported with the model.h5 for making proper predictions
param:: -
return:: -
"""
__init_logging()
# fetch config file, extract values
config = _load_config_json()
df_path = _get_key_value_config('dataFrame-path', config)
print(df_path)
date_col = _get_key_value_config('date-col', config)
target_col = _get_key_value_config('target-col', config)
#loading data frame
data = pd.read_csv(df_path, encoding='ISO-8859-1', sep=",")
data = data.rename(columns={date_col: 'date', target_col: 'sales'})
date_col = 'date'
target_col = 'sales'
# data frame with Sales and Date only
print(data.head())
df_lstm = data.loc[:, (date_col, target_col)].copy()
# convert date field to pandas.Datetime obj and group by month
df_lstm[date_col] = pd.to_datetime(df_lstm[date_col])
df_lstm[date_col] = df_lstm[date_col].dt.year.astype(
'str') + '-' + df_lstm[date_col].dt.month.astype('str') + '-01'
df_lstm[date_col] = pd.to_datetime(df_lstm[date_col])
df_lstm = df_lstm.groupby(date_col).sales.sum().reset_index()
#############################################################
# comupting the difference between sales and prev sales. Adding the lags to the data frame
df_diff = df_lstm.copy()
df_diff['prev_sales'] = df_diff['sales'].shift(1)
df_diff = df_diff.dropna()
df_diff['difference'] = _compute_diff_of_sales(df_diff['sales'],
df_diff['prev_sales'])
df_supervised = df_diff.drop(['prev_sales'], axis=1).copy()
df_supervised = _create_lags_for_supervised(df_supervised, 12)
df_features = df_supervised.drop(['sales', 'date'], axis=1)
#############################################################
# split data frame in train test sets and scale both arrays
train_set, test_set = df_features[0:-6].values, df_features[-6:].values
train_scaled, test_scaled = _min_max_scaling(train_set, test_set)
#############################################################
# creating feature and target variables. Model creation / fiting and saving
X_train, y_train = train_scaled[:, 1:], train_scaled[:, 0:1]
X_train = X_train.reshape(X_train.shape[0], 1, X_train.shape[1])
X_test, y_test = test_scaled[:, 1:], test_scaled[:, 0:1]
X_test = X_test.reshape(X_test.shape[0], 1, X_test.shape[1])
lstm_model = _create_LSTM_model(X_train, y_train)
try:
save_data_to_(file_name='lstm-model.h5',
folder_name='saved_model_data',
data=lstm_model,
is_model=True)
except Exception as save_error:
print(
cp(
f'log[error]:Failed to save data for \'lstm-model.h5\'. See log file',
'red'))
logging.error('Failed to save data for', exc_info=True)
def log(mode="None",out="",noLog=False,file=fnameSet):
try:
from termcolor import colored as col
from termcolor import cprint as cp
import colorama
except ImportError:
print("\nMissing dependancy\n")
raise missingContent
from datetime import datetime
colorama.init()
type=""
if out == "":#User used the function incorrectly
raise incorrectUsage#User did an oof
if mode == "None":
raise incorrectUsage
elif mode == 'e':
type = "ERROR"
if out != "": cp("{}".format(type),"white",'on_red',attrs=['bold'],end=""); print(": {}".format(out)) # this shows up in the interactive prompt
elif mode == 'w':
type = "Warning"
if out != "": cp("{}".format(type),"yellow",attrs=['bold'],end=""); print(": {}".format(out)) # this shows up in the interactive prompt
elif mode == 'n':
type = "Notice"
if out != "": print("{}: {}".format(type,out)) # this shows up in the interactive prompt
elif mode == 'i':
type = "Info"
if out != "": cp("{}".format(type),"blue",attrs=['bold'],end=""); print(": {}".format(out)) # this shows up in the interactive prompt
elif mode == 's':
type = "SUCCESS"
if out != "": print("{}: {}".format(col(type,"green",attrs=['bold']),out)) # this shows up in the interactive prompt
elif mode == 'z':
type = "SPAM"
if noLog == True: raise incorrectUsage
if noLog == False:
try:
f = open(file, 'a') #prints now go the the file
f.write("\n{} - {}: {}".format(str(datetime.now().strftime('%Y-%m-%d %H:%M:%S')),type,out)) # nothing appears. it's written to log file instead
f.close
except FileNotFoundError:
cp("consoleTools ERROR ","white",'on_red',attrs=['bold'],end=""); cp("Issue writing to log file - does not exist. Are you running from an IDE? (doesn't always work in an IDE)","red",'on_white',attrs=['bold'],end="\n")
cp("Try specifying a log file when using the log tool if you are not running the script from the same directory as it is located.","red",'on_white',attrs=['bold'],end="\n")
except FileExistsError:
cp("consoleTools ERROR ","white",'on_red',attrs=['bold','blink'],end=""); cp("Issue writing to log file - file exists. Is the file open in the background/by another process (won't be able to write if so)","red",'on_white',attrs=['bold'],end="\n")