def button_event_6():
#防止未导入信息而报错
if di.global_class_names == 'null':
roll_text_1.insert('insert', "\n请先导入信息，再进行标签预测！")
else:
#获取prediction数据的地址
prediction.predict_txt_dir = entry_2_8.get()
#将myGUI传给dpp的数据同样传给precision
#直接由myGUI传比较好，直接得到模型就能预测
#传入channel
channel = entry_2_6.get()
prediction.channel_s = int(channel.split(',')[0])
prediction.channel_e = int(channel.split(',')[1])
#传入botton_state
#判断复选框状态，确定dpp.main()中是否全部运行
#裁剪
if cb_v1.get() == 0:
prediction.botton_state[0] = 0
else:
prediction.botton_state[0] = 1
#均值
if cb_v2.get() == 0:
prediction.botton_state[1] = 0
else:
prediction.botton_state[1] = 1
#去背景
if cb_v3.get() == 0:
prediction.botton_state[2] = 0
else:
prediction.botton_state[2] = 1
#数值缩放
if cb_v4.get() == 0:
prediction.botton_state[3] = 0
else:
prediction.botton_state[3] = 1
#降维
if cb_v5.get() == 0:
prediction.botton_state[4] = 0
else:
prediction.botton_state[4] = 1
#生成标签
if cb_v6.get() == 0:
prediction.botton_state[5] = 0
else:
prediction.botton_state[5] = 1
#传入其他三项
prediction.com_2_scale_state = combobox_2.get()
prediction.com_3_dre_state = combobox_3.get()
prediction.dred_num = int(entry_2_11.get())
#开始预测
roll_text_1.insert('end', "\n开始预测...")
predict_results = prediction.main()
for predict_result in predict_results:
roll_text_1.insert(
'end', "\n 预测该条数据对应样本类型为：" +
str(di.global_class_names[predict_result]))
roll_text_1.insert('end', "\n预测完成！\n")
return 0
def final_predict():
file_name = request.get_json()['file_name']
algorithm_name = request.get_json()['algorithm_name']
pred_type = request.get_json()['pred_type']
features = request.get_json()['features']
target = request.get_json()['target']
inputType = request.get_json()['inputType']
feature_values = request.get_json()['feature_values']
result = prediction.main(file_name, algorithm_name, pred_type, features,
target, inputType, feature_values)
return jsonify({"result": result.tolist()})
def start1():
#click.echo("HELLO")
df=pd.DataFrame(columns=["S&P 50","Nasdaq","Dow-Jones","Company","DateTime","Price","Price_Difference","Moving Average"])
dj=[]
nasdaq=[]
sp=[]
p_diff=[]
u=scrape.url(df,dj,nasdaq,sp,p_diff)
#output_file=df.to_csv('Stock2.csv')
pred=prediction.main()
return u,pred
def start1():
#click.echo("HELLO")
df = pd.DataFrame(columns=[
"S&P 50", "Nasdaq", "Dow-Jones", "Company", "DateTime", "Price"
])
dj = []
nasdaq = []
sp = []
p_diff = []
u = scrape.url(df, dj, nasdaq, sp, p_diff)
pred = prediction.main()
return u, pred
def post(self):
try:
files = self.request.files
file = files['file'][0]['body']
mime = files['file'][0]['content_type']
pictureid = str(uuid.uuid4())
picturefullpath = "static/uploaddata/" + pictureid + ".jpeg"
with open(picturefullpath, 'wb') as f:
f.write(file)
# クラス判定
result = prediction.main(picturefullpath)
# 結果を保存
with open('database.tsv', 'a') as f:
f.write(pictureid + "\t" + str(result) + "\n")
self.render("imageclass.html", pictureid=pictureid, result=result)
# self.write("ファイルを保存しました")
except KeyError:
self.render("imageclass.html", pictureid="", result="")
def myCustomCalculation(cID, rID, cOR, y, m, d):
dat = dt.datetime(year=y, month=m, day=d)
print("\nCOMPANY:", cID, "RESTAURANT:", rID, "Delivery Date:", dat,
'\n========================================================')
PREDICTION = prediction.main(cID, rID, cOR, dat)
return {'cID': cID, 'rID': rID, 'prediction': PREDICTION}
def predictions_function():
prediction.main()
def handle_pre(self, filedir, filename, index):
result = main(filedir + "/", filename, index)
value = result.get(index)
return value
def main():
prediction.main()