def on_actRun_triggered(self):
"""
运行按钮
:return: 无返回值
"""
self.ui.actRun.setEnabled(False) # 这里需要注意的是点击一次run 控件之后，应当设置未为不可选，
self.key_para["SaveData_Statue"] = False
self.key_para.update(self.get_panel_para()) # 运行前应该得到面板参数，返回dict，并更新self.key_para
print(self.key_para)
self._data_thread = QThread()
self.data_analysis = DataAnalysis(self.key_para)
self.data_analysis.pbar.connect(self.set_progressBar_int)
self.data_analysis.tbw.connect(self.add_textBrowser_str)
self.data_analysis.sbar.connect(self.add_statusBar_str)
self.data_analysis.run_end.connect(lambda: self.stop_thread(self._data_thread))
self.data_analysis.moveToThread(self._data_thread)
self._data_thread.started.connect(self.data_analysis.run)
self._data_thread.finished.connect(self.draw_prepare)
show_info = "数据计算中..."
self.add_statusBar_str(show_info)
self.add_textBrowser_str(show_info)
self._data_thread.start()
print(f'开启{self._data_thread.currentThread()}线程，现在线程状态 ：', self._data_thread.isRunning())
def main(file_name):
data = DataAnalysis(file_name)
data.read_data(file_name)
print("Languages:")
print_output(data.top_n_lang_freqs(10))
print("Top level country domains:")
print_output(data.top_n_country_tlds_freqs(10))
def analyze_and_train():
# to view the different metrics obtained from the data run perform data analysis
eda = DataAnalysis(input)
eda.perform_data_analysis()
# to train a model run the trainer function
tr = Trainer(data_path=input, model_name=model_name)
tr.train()
tr.eval()
tr.save()
def main(file_name):
data = DataAnalysis(file_name)
# Report top ten languages by frequency
print("Languages:")
print_output(data.top_n_lang_freqs(10))
# Report top ten country (2 letter) top
# level domains by frequency
print("Top level country domains:")
print_output(data.top_n_country_tlds_freqs(10))
def main(file_name):
'''analyze the users file for top usage of languages
and top level domain'''
data = DataAnalysis(file_name)
data.read_data(file_name)
# Report top ten languages by frequency
print("Languages:")
print_output(data.top_n_lang_freqs(10))
# Report top ten country (2 letter) top
# level domains by frequency
print("Top level country domains:")
print_output(data.top_n_country_tlds_freqs(10))
def splitDataFirstMode():
# use all available data
sampleData = dp.returnData(1, randomSeed=10)
saveDataset(sampleData, 'AD_dataset')
# split data into X and y
da = DataAnalysis(dirName=dirName)
X, y = da.splitXY(sampleData)
# split data into training (80%) and testing (20%) set
xTrain, xTest, yTrain, yTest = da.splitTrainTest(X, y, trainSize=0.8)
saveDataset(xTrain.assign(normality=yTrain.values), 'AD_set_train')
saveDataset(xTest.assign(normality=yTest.values), 'AD_set_test')
# get data characteristics for current dataset size
dataInfo[1] = da.getDataCharacteristics(yTrain, yTest)
# Run predictions
for datasetSize in (selectedSizes or [0.2, 0.4, 0.6, 0.8, 1]):
# get no. of samples and prediction accuracy
noOfSamples, predictions = da.getScores(
xTrain,
xTest,
yTrain,
yTest,
trainSize=datasetSize,
randomSeeds=randomSeeds,
selectedAlgorithms=selectedAlgorithms
) if selectedAlgorithms else da.getScores(xTrain,
yTrain,
xTest,
yTest,
trainSize=datasetSize,
randomSeeds=randomSeeds)
# save results for graphs and .csv files
savePredictionScores(noOfSamples, predictions, datasetSize)
def normalMode():
da = DataAnalysis(dirName=dirName)
# Run predictions
for datasetSize in (selectedSizes or [0.2, 0.4, 0.6, 0.8, 1]):
# get the percentage of all data
sampleData = dp.returnData(datasetSize, randomSeed=10)
saveDataset(sampleData, 'AD_dataset')
# split data into X and y
X, y = da.splitXY(sampleData)
# split data into training (80%) and testing (20%) set
xTrain, xTest, yTrain, yTest = da.splitTrainTest(X, y, trainSize=0.8)
# get data characteristics for current dataset size
dataInfo[datasetSize] = da.getDataCharacteristics(yTrain, yTest)
# get no. of samples and prediction accuracy (trainSize is set to 1, so no further data splitting is done)
# multiple runs make no sense here, since we always take whole set
noOfSamples, predictions = da.getScores(
xTrain,
xTest,
yTrain,
yTest,
trainSize=1,
randomSeeds=[10],
selectedAlgorithms=selectedAlgorithms
) if selectedAlgorithms else da.getScores(xTrain,
yTrain,
xTest,
yTest,
trainSize=datasetSize,
randomSeeds=randomSeeds)
# save results for graphs and .csv files
savePredictionScores(noOfSamples, predictions, datasetSize)
def _main_thread_loop():
global thread_DH
global thread_DA1
global thread_DBStrip
global thread_J
global packet_queue
# Поднимаем потоки
thread_DH.start()
thread_DA1.start()
thread_DBStrip.start()
thread_J.start()
# Перезапускаем потоки по мере их падения
while True:
if thread_DH.isAlive() == False:
thread_DH = DataHarvest(packet_queue)
thread_DH.start()
logger.warning('Поток сборщика данных перезапущен.')
if thread_DA1.isAlive() == False:
thread_DA1 = DataAnalysis(packet_queue)
thread_DA1.start()
logger.warning('Поток анализа данных канала перезапущен.')
if thread_DBStrip.isAlive() == False:
thread_DBStrip = DBStriper()
thread_DBStrip.start()
logger.warning('Поток удаления старых данных перезапущен.')
if thread_J.isAlive() == False:
thread_J = EventJournal()
thread_J.start()
logger.warning('Поток журналирования перезапущен.')
sleep(THREAD_CHECK_TIMEOUT)
# Load config
config = json.load(open("../config.json"))
# Load markets/articles
articles_csv = csv.reader(open("articles.csv"))
# Parse markets/articles
articles = []
for row in articles_csv:
if row[0] == "Article Title":
# Skip header
continue
article_obj = {}
article_obj["article"] = {"title": row[0]}
article_obj["market"] = {
"symbol": "TEST_MARKET",
"entities": row[1].split(","),
"wikipedia_urls": row[2].split(","),
"target_words": row[3].split(","),
"anti_target_words": row[4].split(",")
}
article_obj["label"] = int(row[5])
articles.append(article_obj)
da = DataAnalysis(config, load_model=False)
da.create_model(articles)
da.load_model()
import sys import matplotlib.pyplot as plt import intrinio from data_analysis import DataAnalysis # USERNAME = '******' # PASSWORD = '******' USERNAME = '******' PASSWORD = '******' data_collection = DataCollection(username=USERNAME, password=PASSWORD) ticker_symbol = "AAPL" dates, prices = data_collection.retrieve_data(ticker_symbol) da = DataAnalysis() # data = da.get_yearly_trends(dates, prices, [3]) bounds, sell_dates, profits = da.get_optimal_bounds(dates, prices, low_bound_min=0.3, top_bound_max=0.5, interval=0.01, investment=100, purchase_strategy="immediate") da.plot_data(dates, prices, ticker_symbol, sell_dates=sell_dates) # if sys.argv[1]: # ticker = sys.argv[1] # start_date = sys.argv[2] # results = data_collection.get_prices(ticker, start_date) # data_collection.plot_data(ticker) # data_collection.save_data(ticker)
import threading
import logging
from logging.handlers import RotatingFileHandler
from time import sleep
from sqlalchemy import create_engine
from sqlalchemy_utils import database_exists, create_database
import os
from param import extractParams
appdir = os.path.abspath(os.path.dirname(__file__))
logger = logging.getLogger("mitter")
packet_queue = Queue()
thread_DH = DataHarvest(packet_queue)
thread_DA1 = DataAnalysis(packet_queue)
thread_DBStrip = DBStriper()
thread_J = EventJournal()
THREAD_CHECK_TIMEOUT = 1 # Интервал контроля работоспособности потоков в сек.
def MainRoutine():
_logger_config()
_postgres_db_check()
extractParams()
try:
_main_thread_loop()
except Exception:
def AnalysisTrainModels():
if request.method == 'GET':
dataset_name = 'data.csv'
out = DataAnalysis(dataset_name)
train = MainML(dataset_name)
return train
class Signal(QtWidgets.QWidget):
signal = QtCore.pyqtSignal()
class COMThreadManual(QtCore.QThread):
def __init__(self):
QtCore.QThread.__init__(self)
def __del__(self):
self.wait()
def run(self):
ui.refresh_com()
if __name__ == "__main__":
import sys
app = QtWidgets.QApplication(sys.argv)
app.setStyle('Fusion')
pinchSensorUI = MainWindow()
ui = Ui_pinchSensorUI()
ui2 = DataAnalysis()
pinchSensorUI.show()
sys.exit(app.exec_())
# TODO bootup loading screen
