打包exe后，启动速度慢的问题

rsj0315 · 发表于 2021-1-21 15:50:13

马上注册，结交更多好友，享用更多功能^_^

您需要登录才可以下载或查看，没有账号？立即注册

x

废话不多说上源码，两个py文件，第一个叫database。如下：

import pyodbc
import pandas as pd
import os
import tempfile
os.environ['MPLCONFIGDIR']=tempfile.mkdtemp()
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
import warnings
warnings.filterwarnings('ignore')
import datetime
import concurrent.futures
import time
now = datetime.datetime.now().strftime('%Y-%m-%d')
now_15ago=(datetime.datetime.now()-datetime.timedelta(days=14)).strftime('%Y-%m-%d')
now_1ago=(datetime.datetime.now()-datetime.timedelta(days=1)).strftime('%Y-%m-%d')
now_2ago=(datetime.datetime.now()-datetime.timedelta(days=2)).strftime('%Y-%m-%d')
now_30ago = (datetime.datetime.now()-datetime.timedelta(days=30)).strftime('%Y-%m-%d')
plt.rcParams['font.sans-serif'] = ['STZhongsong'] # 中文字体设置-黑体
plt.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
#显示所有列
pd.set_option('display.max_columns', None)
#显示所有行
pd.set_option('display.max_rows', None)
#下边是5条产线，15个access数据库表格
a_fg = r'A Rubber Line POs List.mdb'
a_per = r'A Rubber Line Performance List.mdb'
a_stop = r'A Rubber Line Stops List.mdb'
b_fg = r'B Rubber Line POs List.mdb'
b_per = r'B Rubber Line Performance List.mdb'
b_stop = r'B Rubber Line Stops List.mdb'
c_fg = r'C Rubber Line POs List.mdb'
c_per = r'C Rubber Line Performance List.mdb'
c_stop = r'C Rubber Line Stops List.mdb'
cb_fg = r'CB Rubber Line POs List.mdb'
cb_per = r'CB Rubber Line Performance List.mdb'
cb_stop = r‘CB Rubber Line Stops List.mdb'
ccv_fg = r'CCV XLPE Line POs List.mdb'
ccv_per = r'CCV XLPE Line Performance List.mdb'
ccv_stop = r'CCV XLPE Line Stops List.mdb'
#下边是后边要用的sql语句
fg_a = "SELECT * FROM A_Rubber_Line_POLIST"
per_a = "SELECT * FROM A_Rubber_Line_PerformList"
stop_a = "SELECT * FROM A_Rubber_Line_StopList"
fg_b = "SELECT * FROM B_Rubber_Line_POLIST"
per_b = "SELECT * FROM B_Rubber_Line_PerformList"
stop_b = "SELECT * FROM B_Rubber_Line_StopList"
fg_c = "SELECT * FROM C_Rubber_Line_POLIST"
per_c = "SELECT * FROM C_Rubber_Line_PerformList"
stop_c = "SELECT * FROM C_Rubber_Line_StopList"
fg_cb = "SELECT * FROM CB_Rubber_Line_POLIST"
per_cb = "SELECT * FROM CB_Rubber_Line_PerformList"
stop_cb = "SELECT * FROM CB_Rubber_Line_StopList"
fg_ccv = "SELECT * FROM CCV_XLPE_Line_POLIST"
per_ccv = "SELECT * FROM CCV_XLPE_Line_PerformList"
stop_ccv = "SELECT * FROM CCV_XLPE_Line_StopList"
savepath = r'C:\DATA_FILE'
#下边这个函数是为了让饼图显示数量和百分比的
def my_label(pct, allvals):
try:
absolute = int(pct/100.*np.sum(allvals))
return "{:.1f}%\n({:d})".format(pct, absolute)
except Exception as e:
print(e)
#下边这个函数是读取5个fg数据库表格的通用函数，5个表格字段一样的
def get_data_fg(mdb_file,sql,line):
global savepath
driver = '{Microsoft Access Driver (*.mdb, *.accdb)}'
con = pyodbc.connect(f'Driver={driver};DBQ={mdb_file};Pwd=FSL24680')
df = pd.read_sql(sql,con)
df['PRODUCTION_DATE'] = df['COIL_NO'].str[:8]
df['PRODUCTION_DATE'] = pd.to_datetime(df['PRODUCTION_DATE'], format='%Y-%m-%d')
df['date'] = df['PRODUCTION_DATE']
# df.set_index('date', inplace=True)
df['line'] = line
df['itp_'] = df['QUALITY_DECISION'].str.cat(df['PRODUCTION_DECISION'], na_rep='_') # 连接合并，空的写下划线
df['itp'] = df['itp_'].isin(['Failnan', 'nanFail', 'FailFail', 'FailPass', 'PassFail']) # 只要含有Fail的就认为是不合格轴
df['itp'] = df['itp'].astype('str') # 把判定结果的转换为字符串，False代表合格。
df['段长'] = pd.cut(df['COIL_LENGTH'],
bins=[0, 300, 600, 1000, 10000],
include_lowest=True,
labels=['300m以内', '300到600m', '600到1000m', '1km以上'])
# print(df)
# print(df.info())
df.to_pickle(os.path.join(savepath, f'{line}fg.pkl'))
# df.to_excel(os.path.join(savepath, f'{line}fg.xlsx'))
print('***********',sql,"****************")
return df
##下边这个函数是读取5个per数据库表格的通用函数，5个表格字段一样的
def get_data_per(mdb_file,sql,line):
global savepath
driver = '{Microsoft Access Driver (*.mdb, *.accdb)}'
con = pyodbc.connect(f'Driver={driver};DBQ={mdb_file};Pwd=FSL24680')
df = pd.read_sql(sql,con)
df['date'] = df['PRODUCTION_DATE']
# df.set_index('date', inplace=True)
df['line'] = line
df.to_pickle(os.path.join(savepath, f'{line}per.pkl'))
# df.to_excel(os.path.join(savepath, f'{line}per.xlsx'))
print('***********',sql,"****************")
return df
#下边这个函数是读取5个stop数据库表格的通用函数，5个表格字段一样的
def get_data_stop(mdb_file,sql,line):
global savepath
driver = '{Microsoft Access Driver (*.mdb, *.accdb)}'
con = pyodbc.connect(f'Driver={driver};DBQ={mdb_file};Pwd=FSL24680')
df = pd.read_sql(sql,con)
# print(df)
df['STOP_REASON'] = np.where((df['STOP_TIME_HOUR'] > 10) & (df['STOP_REASON'] == ''), '歇班',
df['STOP_REASON'])
df['STOP_REASON'] = np.where((df['STOP_TIME_HOUR'] < 10) & (df['STOP_REASON'] == ''), '等待下班',
df['STOP_REASON'])
df_stop = df[~df['STOP_REASON'].astype(str).str.contains('歇班')]
# print(df_stop)
a = df_stop.values.tolist()
# print('---')
# print(a)
res = []
res_=[]
for i in range(0, len(a) - 1):
m = a[i + 1][2]
n = a[i][3]
b = (m - n).total_seconds() / 3600
if b >= 0.2:
res.append(a[i])
else:
res.append([a[i][0], a[i][1], a[i][2], a[i + 1][3], a[i + 1][4], a[i + 1][5],a[i][6],a[i][7],a[i][8]])
res_.append(
[a[i][0], a[i][1], a[i][2], a[i + 1][3], a[i + 1][4], a[i + 1][5] + '+跨班', a[i][6], a[i][7], a[i][8]])
df_stop2 = pd.DataFrame(res)
print(df_stop2)
df_stop2.columns = ['date', 'INDEX SUB_INDEX', 'STOP_FROM', 'STOP_TO', 'STOP_TIME_HOUR', 'STOP_REASON',
'PRODUCTION_DATE', 'SHIFT', 'OPERATOR','LINE']
df_stop2.drop_duplicates(subset=['STOP_TO'], keep='first', inplace=True)
df_stop2['date']=df_stop2['PRODUCTION_DATE']
# df_stop2.set_index('date', inplace=True)
df_stop2['line'] = line
# print(df_stop2)
df_stop2.to_pickle(os.path.join(savepath, f'{line}stop.pkl'))
# df_stop2.to_excel(os.path.join(savepath, f'{line}stop.xlsx'))
if len(res_)==9:
df_stop3 = pd.DataFrame(res_)
df_stop3.columns = ['date', 'INDEX SUB_INDEX', 'STOP_FROM', 'STOP_TO', 'STOP_TIME_HOUR', 'STOP_REASON',
'PRODUCTION_DATE', 'SHIFT', 'OPERATOR']
df_stop3.drop_duplicates(subset=['STOP_TO'], keep='first', inplace=True)
df_stop3['date'] = df_stop3['PRODUCTION_DATE']
df_stop3['line'] = line
# df_stop3.to_excel(os.path.join(savepath, f'{line}stop_跨班.xlsx'))
print('***********',sql,"****************")
return df_stop2
#把5个fg表格合并成一个并输出pkl文件
def concat_file_fg():
start = time.time()
with concurrent.futures.ThreadPoolExecutor() as executor:
fga = executor.submit(get_data_fg, a_fg, fg_a, 'A')
fgb = executor.submit(get_data_fg, b_fg, fg_b, 'B')
fgc = executor.submit(get_data_fg, c_fg, fg_c, 'C')
fgcb = executor.submit(get_data_fg, cb_fg, fg_cb, 'CB')
fgccv = executor.submit(get_data_fg,ccv_fg, fg_ccv, 'CCV')
dfga = fga.result()
dfgb = fgb.result()
dfgc = fgc.result()
dfgcb = fgcb.result()
dfgccv = fgccv.result()
df_fg = pd.concat([dfga, dfgb, dfgc, dfgcb, dfgccv])
df_fg.to_excel(r'C:\PTCC_MES_DATA_FILE\fg2021_all.xlsx', index=False)
df_fg.to_pickle(r'C:\PTCC_MES_DATA_FILE\fg2021_all.pkl')
end = time.time()
print(end-start)
print('fg载入数据完成------------------')
return df_fg
#把5个per表格合并成一个并输出pkl文件
def concat_file_per():
start = time.time()
with concurrent.futures.ThreadPoolExecutor() as executor:
pera = executor.submit(get_data_per, a_per, per_a, 'A')
perb = executor.submit(get_data_per, b_per, per_b, 'B')
perc = executor.submit(get_data_per, c_per, per_c, 'C')
percb = executor.submit(get_data_per, cb_per, per_cb, 'CB')
perccv = executor.submit(get_data_per, ccv_per, per_ccv, 'CCV')
dfpa = pera.result()
dfpb = perb.result()
dfpc = perc.result()
dfpcb = percb.result()
dfpccv = perccv.result()
df_p = pd.concat([dfpa, dfpb, dfpc, dfpcb, dfpccv])
df_p.to_excel(r'C:\PTCC_MES_DATA_FILE\per2021_all.xlsx', index=False)
df_p.to_pickle(r'C:\PTCC_MES_DATA_FILE\per2021_all.pkl')
end = time.time()
print(end - start)
print('per载入数据完成------------------')
return df_p
#把5个stop表格合并成一个并输出pkl文件
def concat_file_stop():
start = time.time()
with concurrent.futures.ThreadPoolExecutor() as executor:
stopa = executor.submit(get_data_stop, a_stop, stop_a, 'A')
stopb = executor.submit(get_data_stop, b_stop, stop_b, 'B')
stopc = executor.submit(get_data_stop, c_stop, stop_c, 'C')
stopcb = executor.submit(get_data_stop, cb_stop, stop_cb, 'CB')
stopccv = executor.submit(get_data_stop, ccv_stop, stop_ccv, 'CCV')
dfsa = stopa.result()
dfsb = stopb.result()
dfsc = stopc.result()
dfscb = stopcb.result()
dfsccv = stopccv.result()
df_s = pd.concat([dfsa, dfsb, dfsc, dfscb, dfsccv])
df_s.to_excel(r'C:\PTCC_MES_DATA_FILE\stop2021_all.xlsx', index=False)
df_s.to_pickle(r'C:\PTCC_MES_DATA_FILE\stop2021_all.pkl')
end = time.time()
print(end - start)
print('stop载入数据完成------------------')
return df_s
#画图函数1，ui上点击按钮对应画图
def out1_data(line,start=now_15ago,end=now,now_1 = now_1ago):
global savepath
df_fg = pd.read_pickle(os.path.join(savepath, f'{line}fg.pkl'))
df_fg.set_index('date', inplace=True)
df_fg = df_fg.loc[start:end]
df_fg['PRODUCTION_DATE'] = df_fg['PRODUCTION_DATE'].astype(str).str[:10]
# print(df_fg)
# df_per = pd.read_excel(os.path.join(savepath, f'{line}per.xlsx'))
df_per = pd.read_pickle(os.path.join(savepath, f'{line}per.pkl'))
df_per.set_index('date', inplace=True)
df_per = df_per.loc[start:end]
df_per['PRODUCTION_DATE']=df_per['PRODUCTION_DATE'].astype(str).str[:10]
# print(df_per)
df_stop = pd.read_pickle((os.path.join(savepath, f'{line}stop.pkl')))
df_stop['PRODUCTION_DATE'] = df_stop['PRODUCTION_DATE'].astype(str).str[:10]
df_stop.set_index('date', inplace=True)
#----------------------------------------------------
plt.figure(facecolor='white', figsize=(16, 9))
sns.set(style="darkgrid")
plt.rcParams['font.sans-serif'] = ['STZhongsong'] # 中文字体设置-黑体
plt.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
try:
ax1 = plt.subplot(221)
sns.lineplot(x='PRODUCTION_DATE',
y='SPEED_EFFICIENCY',
data=df_per,
hue='SHIFT',
marker='*',
ax=ax1)
for x,y in zip(df_per['PRODUCTION_DATE'],df_per['SPEED_EFFICIENCY']):
if y<0.95:
plt.text(x=x,y=y+0.02,s='%.01f%%' % (y * 100),color='red',size=14)
plt.xticks(ha='right', rotation=45)
plt.ylim(0, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
# plt.grid(axis='y',linestyle='--',color='red',alpha=0.3)
plt.ylabel('速度效率 / %', fontsize=14, fontweight='bold')
ax1.set_title(f'{line} Line 15天速度效率 %\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天速度效率')
ax2 = plt.subplot(222)
try:
zongshu = df_fg.groupby(['PRODUCTION_DATE', 'PRODUCTION_SHIFT'], as_index=False)['itp'].agg('count')
df_fg_hege = df_fg[df_fg['itp'] == 'False']
hegeshu = df_fg_hege.groupby(['PRODUCTION_DATE','PRODUCTION_SHIFT'],as_index=False)['itp'].agg('count')
df_ftq = pd.concat([zongshu,hegeshu],axis=1)
df_ftq.columns = ['PRODUCTION_DATE','PRODUCTION_SHIFT','zongshu','date','SHIFT','hegeshu']
df_ftq['FTQ']=df_ftq['hegeshu']/df_ftq['zongshu']
# print(df_ftq)
sns.lineplot(x='PRODUCTION_DATE',
y='FTQ',
data=df_ftq,
hue='SHIFT',
marker='*',
ax=ax2)
for x,y in zip(df_ftq['PRODUCTION_DATE'],df_ftq['FTQ']):
if y<0.1:
plt.text(x=x,y=y+0.02,s='%.01f%%' % (y * 100),color='red',size=14)
plt.xticks(ha='right', rotation=45)
plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('FTQ / %', fontsize=14, fontweight='bold')
ax2.set_title(f'{line} Line 15天 FTQ %\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天 FTQ')
ax3 = plt.subplot(223)
try:
sns.lineplot(x='PRODUCTION_DATE',
y='RUN_TIME_HOUR',
data=df_per,
hue='SHIFT',
marker='*',
ax=ax3)
for x,y in zip(df_per['PRODUCTION_DATE'],df_per['RUN_TIME_HOUR']):
plt.text(x=x,y=y,s=round(y,1),color='red',size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('run_time / h', fontsize=14, fontweight='bold')
ax3.set_title(f'{line} Line 15天 Run_Time h\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天 Run_Time')
ax4 = plt.subplot(224)
try:
sns.lineplot(x='PRODUCTION_DATE',
y='OEE',
data=df_per,
hue='SHIFT',
marker='*',
ax=ax4)
for x,y in zip(df_per['PRODUCTION_DATE'],df_per['OEE']):
plt.text(x=x,y=y,s='%.01f%%' % (y * 100),color='red',size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('OEE / %', fontsize=14, fontweight='bold')
ax4.set_title(f'{line} Line 15天 OEE %\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天 OEE')
plt.tight_layout()
plt.show()
###################################################
###################################################
plt.figure(facecolor='white', figsize=(16, 9))
ax1 = plt.subplot(221)
try:
ax = sns.barplot(x='PRODUCTION_DATE',
y='TOTAL_LENGTH',
data=df_per,
hue='SHIFT',
ci=0,
ax=ax1)
for p in ax.patches:
height = p.get_height()
ax.text(x=p.get_x() ,
y=height + 2,
s='{: .0f}m'.format(height), color='red', size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('length / m', fontsize=14, fontweight='bold')
ax1.set_title(f'{line} Line 15天产量/ m\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天产量')
#############################################
ax2 = plt.subplot(222)
try:
df_fg1 = df_fg.loc[now_1:now]
axx = sns.barplot(x='COIL_NO',
y='COIL_LENGTH',
data=df_fg1,
ci=0,
color='blue',
ax=ax2)
for p in axx.patches:
height = p.get_height()
axx.text(x=p.get_x() + (p.get_width() / 2),
y=height,
s='{: .0f}m'.format(height), color='red', size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('length / m', fontsize=14, fontweight='bold')
ax2.set_title(f'{line} Line 昨天产量/ m\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'昨天产量')
##################################
ax3 = plt.subplot(223)
try:
df_stop = df_stop.loc[now_1:now]
axxx = sns.barplot(x='STOP_REASON',
y='STOP_TIME_HOUR',
data=df_stop,
hue='SHIFT',
estimator=sum,
ci=0,
ax=ax3)
for p in axxx.patches:
height = p.get_height()
axxx.text(x=p.get_x() + (p.get_width() / 2),
y=height + 0.1,
s=f'{round(height,1)}h', color='red', size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('小时 / m', fontsize=14, fontweight='bold')
ax3.set_title(f'{line} Line 昨天停车 / h\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'昨天停车')
ax4 = plt.subplot(224)
try:
bing = df_per.groupby('SHIFT')['RUN_TIME_HOUR'].agg('sum').sort_values(ascending=True)
ax4 = bing.plot(kind='pie',
subplots=True,
# labels=labels,
# explode=(0,0.1,0,0),#突显
autopct=lambda x: my_label(x, bing),
# autopct='%.1f%%', # 设置百分比格式，保留一位
pctdistance=0.8, # 设置百分比与圆心的距离
startangle=90, # 设置饼图的开始角度
radius=1, # 设置饼图的半径
counterclock=True, # 是否逆时针，false是顺时针
shadow=False, # 无阴影设置
labeldistance=1.1, # 默认1.1
# rotatelabels=True# True,标签放线是沿半径方向
textprops={'fontsize': 13}
)
plt.axis('scaled') # 'tight''scaled'
plt.ylabel('')
plt.xlabel('')
plt.title('\n15天白夜班开车时间占比\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天白夜班开车时间占比')
plt.tight_layout()
plt.show()
return
#画图函数2，ui上点击按钮对应画图
def out2_data(line,start=now_15ago,end=now):
'''
近1个月
停车分析，
1.次数柏拉图
2.时长sum柏拉图
3.时长mean柱状图图
top3箱图
'''
global savepath
df_s = pd.read_pickle((os.path.join(savepath, f'{line}stop.pkl')))
df_s['PRODUCTION_DATE'] = df_s['PRODUCTION_DATE'].astype(str).str[:10]
df_s.set_index('date', inplace=True)
# 筛选出近一个月的数据
df_s = df_s.loc[start:end]
# df_s['shichang'].dropna(inplace=True)
try:
# 筛选停车原因次数中的top3
top3list = df_s['STOP_REASON'].value_counts().head(3).index.tolist()
df_s_top1 = df_s[df_s['STOP_REASON'].isin([top3list[0]])]
title1 = top3list[0]
df_s_top1_count = df_s_top1.groupby('OPERATOR')['STOP_REASON'].agg('count').sort_values(ascending=False)
# top2的df
df_s_top2 = df_s[df_s['STOP_REASON'].isin([top3list[1]])]
title2 = top3list[1]
df_s_top2_count = df_s_top2.groupby('OPERATOR')['STOP_REASON'].agg('count').sort_values(ascending=False)
# top3的df
df_s_top3 = df_s[df_s['STOP_REASON'].isin([top3list[2]])]
title3 = top3list[2]
df_s_top3_count = df_s_top3.groupby('OPERATOR')['STOP_REASON'].agg('count').sort_values(ascending=False)
except Exception as e:
print(e,'停车种类不够三种')
plt.figure(facecolor='white', figsize=(16, 9))
sns.set(style="darkgrid")
plt.rcParams['font.sans-serif'] = ['STZhongsong'] # 中文字体设置-黑体
plt.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
ax1 = plt.subplot(231)
try:
df1 = df_s['STOP_REASON'].value_counts()
p1 = 1.0 * df1.cumsum() / df1.sum()
df1.plot(kind='bar', ax=ax1)
for i, v in enumerate(df1.values):
a = v.__str__().strip('[,]')
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
plt.ylabel(u"stop_count")
p1.plot(color='r', secondary_y=True, style='-o', linewidth=0.5, ax=ax1)
plt.annotate(format(p1[2], '.2%'), xy=(2, p1[2]), xytext=(2 * 0.9, p1[2] * 0.9),
arrowprops=dict(arrowstyle='->', connectionstyle='arc3,rad=.2'))
plt.ylabel(u'count_rate')
for tick in ax1.get_xticklabels():
tick.set_rotation(45)
tick.set_horizontalalignment('right')
ax1.set_title(f'{line} Line stop_count\n', fontsize=18, fontweight='bold')
except Exception as e:
print('停车次数', e)
# ---------------------------
ax2 = plt.subplot(232)
try:
df2 = df_s.groupby('STOP_REASON')['STOP_TIME_HOUR'].agg('sum').sort_values(ascending=False)
# print(df2)
p2 = 1.0 * df2.cumsum() / df2.sum()
df2.plot(kind='bar', ax=ax2)
for i, v in enumerate(df2.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
plt.ylabel(u"stop time_sum")
p2.plot(color='r', secondary_y=True, style='-o', linewidth=0.5, ax=ax2)
plt.annotate(format(p2[2], '.2%'), xy=(2, p2[2]), xytext=(2 * 0.9, p2[2] * 0.9),
arrowprops=dict(arrowstyle='->', connectionstyle='arc3,rad=.2'))
plt.ylabel(u'time_rate')
ax2.set_xlabel('')
for tick in ax2.get_xticklabels():
tick.set_rotation(45)
tick.set_horizontalalignment('right')
ax2.set_title(f'{line} Line stop_time/h\n', fontsize=18, fontweight='bold')
except Exception as e:
print('停车总时长', e)
ax3 = plt.subplot(233)
try:
df3 = df_s.groupby('STOP_REASON')['STOP_TIME_HOUR'].agg('mean').sort_values(ascending=False)
df3.plot(kind='bar', ax=ax3)
for i, v in enumerate(df3.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
ax3.set_xlabel('')
plt.ylabel(u'stop time_ave/h')
for tick in ax3.get_xticklabels():
tick.set_rotation(45)
tick.set_horizontalalignment('right')
ax3.set_title(f'{line} Line stop time_ave/h\n', fontsize=18, fontweight='bold')
except Exception as e:
print('停车平均时长', e)
# ------------------------
ax4 = plt.subplot(234)
try:
df_s_top1_median = df_s_top1.groupby('OPERATOR')['STOP_TIME_HOUR'].median()
list1 = df_s_top1_median.index.tolist()
ax4 = sns.boxplot(
x='OPERATOR',
y='STOP_TIME_HOUR',
hue='OPERATOR',
data=df_s_top1,
order=list1
)
df_s_top1_25 = df_s_top1.groupby('OPERATOR')['STOP_TIME_HOUR'].describe()['25%']
df_s_top1_75 = df_s_top1.groupby('OPERATOR')['STOP_TIME_HOUR'].describe()['75%']
for i, v in enumerate(df_s_top1_median.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(df_s_top1_25.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(df_s_top1_75.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
ax4.tick_params(axis='x', labelsize=15)
plt.title(f"TOP1：《{title1}》\n", fontsize=18, fontweight='bold')
plt.ylabel('')
plt.xlabel('')
plt.grid(axis='y', color='red', linestyle=':')
except Exception as e:
print(e)
# -------------------------------------------------------------
ax5 = plt.subplot(235)
try:
df_s_top2_median = df_s_top2.groupby('OPERATOR')['STOP_TIME_HOUR'].median()
list2 = df_s_top2_median.index.tolist()
ax5 = sns.boxplot(
x='OPERATOR',
y='STOP_TIME_HOUR',
hue='OPERATOR',
data=df_s_top2,
order=list2
)
df_s_top2_25 = df_s_top2.groupby('OPERATOR')['STOP_TIME_HOUR'].describe()['25%']
df_s_top2_75 = df_s_top2.groupby('OPERATOR')['STOP_TIME_HOUR'].describe()['75%']
for i, v in enumerate(df_s_top2_median.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(df_s_top2_25.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(df_s_top2_75.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
ax5.tick_params(axis='x', labelsize=15)
plt.title(f"TOP2：《{title2}》\n", fontsize=18, fontweight='bold')
plt.ylabel('')
plt.xlabel('')
plt.grid(axis='y', color='red', linestyle=':')
except Exception as e:
print(e)
# ----------------------------------------------------------
ax6 = plt.subplot(236)
try:
df_s_top3_median = df_s_top3.groupby('OPERATOR')['STOP_TIME_HOUR'].median()
list3 = df_s_top3_median.index.tolist()
ax6 = sns.boxplot(
x='OPERATOR',
y='STOP_TIME_HOUR',
hue='OPERATOR',
data=df_s_top3,
order=list3)
df_s_top3_25 = df_s_top3.groupby('OPERATOR')['STOP_TIME_HOUR'].describe()['25%']
df_s_top3_75 = df_s_top3.groupby('OPERATOR')['STOP_TIME_HOUR'].describe()['75%']
for i, v in enumerate(df_s_top3_median.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(df_s_top3_25.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(df_s_top3_75.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
ax6.tick_params(axis='x', labelsize=15)
plt.title(f"TOP3：《{title3}》\n", fontsize=18, fontweight='bold')
plt.ylabel('')
plt.xlabel('')
plt.grid(axis='y', color='red', linestyle=':')
except Exception as e:
print(e)
plt.tight_layout()
plt.show()
return
#画图函数3，ui上点击按钮对应画图
def out3_data(line,start=now_15ago,end=now):
'''
按钮三种的数据都是默认筛选近15天的数据
1.段长饼图----》fg
2.run time饼图/人之间的关系----》per
3.轴长度直方图----》fg
4.run time直方图----》per
'''
global savepath
df_fg = pd.read_pickle(os.path.join(savepath, f'{line}fg.pkl'))
df_fg.set_index('date', inplace=True)
df_fg = df_fg.loc[start:end]
df_fg['PRODUCTION_DATE'] = df_fg['PRODUCTION_DATE'].astype(str).str[:10]
df_per = pd.read_pickle(os.path.join(savepath, f'{line}per.pkl'))
df_per.set_index('date', inplace=True)
df_per = df_per.loc[start:end]
df_per['PRODUCTION_DATE'] = df_per['PRODUCTION_DATE'].astype(str).str[:10]
# 筛选出近一个月的数据
df_p = df_per.loc[start:end]
df_fg = df_fg.loc[start:end]
plt.figure(facecolor='white', figsize=(16, 9))
sns.set(style="darkgrid")
plt.rcParams['font.sans-serif'] = ['STZhongsong'] # 中文字体设置-黑体
plt.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
ax1 = plt.subplot(221)
try:
tu_1 = df_fg['段长'].value_counts()
ax1 = tu_1.plot(kind='pie',
subplots=True,
# labels=labels,
# explode=(0,0.1,0,0),#突显
autopct=lambda x: my_label(x, tu_1),
# 设置百分比格式，保留一位#autopct=lambda x: my_label(x,data)
# autopct='%.1f%%', # 设置百分比格式，保留一位
pctdistance=0.8, # 设置百分比与圆心的距离
startangle=90, # 设置饼图的开始角度
radius=1, # 设置饼图的半径
counterclock=False, # 是否逆时针，false是顺时针
shadow=False, # 无阴影设置
labeldistance=1.1, # 默认1.1
# rotatelabels=True# True,标签放线是沿半径方向
textprops={'fontsize': 13}
)
plt.axis('scaled') # 'tight''scaled'
plt.ylabel('')
plt.xlabel('')
plt.title(f'{start}~{end}\n{line} Line 段长分布饼图\n', fontsize=18, fontweight='bold')
except Exception as e:
print('段长分布饼图',e)
ax2 = plt.subplot(222)
try:
tu_2 = df_p.groupby('OPERATOR')['RUN_TIME_HOUR'].agg('sum').sort_values(ascending=True)
ax2=tu_2.plot(kind='pie',
subplots=True,
# labels=labels,
# explode=(0,0.1,0,0),#突显
autopct=lambda x: my_label(x, tu_2), # 设置百分比格式，保留一位#autopct=lambda x: my_label(x,data)
# autopct='%.1f%%', # 设置百分比格式，保留一位
pctdistance=0.8, # 设置百分比与圆心的距离
startangle=90, # 设置饼图的开始角度
radius=1, # 设置饼图的半径
counterclock=True, # 是否逆时针，false是顺时针
shadow=False, # 无阴影设置
labeldistance=1.1, # 默认1.1
# rotatelabels=True# True,标签放线是沿半径方向
textprops={'fontsize': 13}
)
plt.axis('scaled') # 'tight''scaled'
plt.ylabel('')
plt.xlabel('')
plt.title('\n Run_Time VS 人员分布饼图\n', fontsize=18, fontweight='bold')
except Exception as e:
print('runtime饼图',e)
ax3 = plt.subplot(223)
try:
tu_3 = df_fg
ax3 = sns.distplot(
tu_3['COIL_LENGTH'],
bins='auto', ##auto,fd,doane,scott,stone,rice,sturges,sqrt
hist=True,
rug=False)
ax3.tick_params(axis='x', labelsize=13)
ax3.tick_params(axis='y', labelsize=13)
plt.xlabel('轴长度 / m',fontsize=15)
plt.title( '\n轴长度直方图\n',fontsize=18, fontweight='bold')
except Exception as e:
print('段长分布直方图',e)
ax4 = plt.subplot(224)
try:
tu_8 = df_p
ax4=sns.distplot(
tu_8['RUN_TIME_HOUR'],
bins='auto',
hist=True,
rug=False)
ax4.tick_params(axis='x', labelsize=13)
ax4.tick_params(axis='x', labelsize=13)
plt.xlabel('run_time / h',fontsize=15)
plt.title('\n Run_Time 直方图\n',fontsize=18, fontweight='bold')
except Exception as e:
print('runtime直方图',e)
plt.tight_layout()
plt.show()
return
#画图函数4，ui上点击按钮对应画图
def out4_data(start=now_30ago,end=now):
'''
1.每条产线段长箱图
2.每条产线的runtime箱图
3.每条产线的oee箱图
4. 每条产线的top3停车时长之和占总的监控运行时间占比饼图
5.每个人OEE箱图
'''
df_p = pd.read_pickle(os.path.join(savepath, 'per2021_all.pkl'))
df_p.set_index('date',inplace=True)
#计算运行时间
df_p = df_p.loc[start:end]
yunxingshijian = df_p.groupby('line')['TOTAL_TIME_HOUR'].agg('sum')
# print(yunxingshijian)
df_fg = pd.read_pickle(os.path.join(savepath, 'fg2021_all.pkl'))
df_fg.set_index('date', inplace=True)
df_fg = df_fg.loc[start:end]
df_s = pd.read_pickle(os.path.join(savepath, 'stop2021_all.pkl'))
df_s.set_index('date',inplace=True)
df_s = df_s.loc[start:end][['line','STOP_REASON','STOP_TIME_HOUR']]
# print(df_s.info())
df_s['计数'] = df_s.groupby(['line','STOP_REASON']).transform('count')
#按照产线和原因分组，按照次数的top3进行时长求和
df_s1 = df_s.groupby(['line', 'STOP_REASON', '计数']).sum().groupby('line', as_index=False).apply(
lambda x: x.sort_values('计数', ascending=False).head(3))
#每条线前三的时长之和
df_s2 = df_s1.groupby('line')['STOP_TIME_HOUR'].agg('sum')
df_s3 = pd.concat([yunxingshijian,df_s2],axis=1)
df_s3.reset_index(inplace=True)
df_s3['tingchezhanbi']=df_s3['STOP_TIME_HOUR']/df_s3['TOTAL_TIME_HOUR']
# print(df_s3)
# df_s3.to_excel('2.xlsx')
# y = df_s2.values/yunxingshijian.values
# x = yunxingshijian.index.tolist()
# print(x)
# print(y)
plt.figure(facecolor='white', figsize=(16, 9))
sns.set(style="darkgrid")
plt.rcParams['font.sans-serif'] = ['STZhongsong'] # 中文字体设置-黑体
plt.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
ax1 = plt.subplot(241)
try:
ax1 = sns.boxplot(
x='line',
y='COIL_LENGTH',
data=df_fg,
order=['A','B', 'C', 'CB','CCV'])
tu_1 = df_fg.groupby('line')['COIL_LENGTH'].median()
# print(tu_13)
tu_1_ = df_fg.groupby('line')['COIL_LENGTH'].describe()['25%']
tu_1__ = df_fg.groupby('line')['COIL_LENGTH'].describe()['75%']
for i, v in enumerate(tu_1.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_1_.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_1__.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
ax1.tick_params(axis='x', labelsize=18)
ax1.tick_params(axis='y', labelsize=16)
plt.xlabel('')
plt.ylabel('coil_lenth/m', fontsize=18, fontweight='bold')
ax1.set_title(f'\n{start}-{end}\nCoil_Length vs Line\n', fontsize=24, fontweight='bold')
except Exception as e:
print('轴段长箱图',e)
ax2 = plt.subplot(242)
try:
ax2 = sns.boxplot(
x='line',
y='RUN_TIME_HOUR',
data=df_p,
order=['A', 'B', 'C', 'CB', 'CCV'])
tu_2 = df_p.groupby('line')['RUN_TIME_HOUR'].median()
# print(tu_14)
tu_2_ = df_p.groupby('line')['RUN_TIME_HOUR'].describe()['25%']
tu_2__ = df_p.groupby('line')['RUN_TIME_HOUR'].describe()['75%']
for i, v in enumerate(tu_2.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_2_.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_2__.values):
a = round(float(v.__str__().strip('[,]')), 1)
plt.text(i, v, a,
ha='center',
va='bottom',
fontsize=12)
ax2.tick_params(axis='x', labelsize=18)
ax2.tick_params(axis='y', labelsize=16)
plt.xlabel('')
plt.ylabel('run_time/h', fontsize=18, fontweight='bold')
ax2.set_title('\nRun_Time vs Line\n', fontsize=24, fontweight='bold')
except Exception as e:
print('runtime箱图',e)
ax3 = plt.subplot(243)
try:
ax3 = sns.boxplot(
x='line',
y='OEE',
data=df_p,
order=['A', 'B', 'C', 'CB', 'CCV'])
tu_3 = df_p.groupby('line')['OEE'].median()
tu_3_ = df_p.groupby('line')['OEE'].describe()['25%']
tu_3__ = df_p.groupby('line')['OEE'].describe()['75%']
for i, v in enumerate(tu_3.values):
a = float(v.__str__().strip('[,]'))
b = '%.01f%%' % (a * 100)
# print(a,type(a))
plt.text(i, v, b,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_3_.values):
a = float(v.__str__().strip('[,]'))
b = '%.01f%%' % (a * 100)
# print(a,type(a))
plt.text(i, v, b,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_3__.values):
a = float(v.__str__().strip('[,]'))
b = '%.01f%%' % (a * 100)
# print(a,type(a))
plt.text(i, v, b,
ha='center',
va='bottom',
fontsize=12)
ax3.tick_params(axis='x', labelsize=18)
ax3.tick_params(axis='y', labelsize=16)
# ax.set_xticklabels(ax.get_xticklabels(), rotation=45)
plt.ylim(0, 1.2)
plt.xlabel('')
plt.ylabel('OEE/%', fontsize=18, fontweight='bold')
ax3.set_title('\nOEE vs Line\n', fontsize=24, fontweight='bold')
except Exception as e:
print('OEE箱图',e)
ax4 = plt.subplot(244)
try:
axx=sns.barplot(x='line',
y='tingchezhanbi',
data=df_s3,
ci=0,
# color='blue',
ax=ax4)
for p in axx.patches:
height = p.get_height()
axx.text(x=p.get_x(),
y=height+0.02,
s='%.01f%%' % (height * 100),
color='black',
size=12)
# plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
ax4.tick_params(axis='x', labelsize=18)
ax4.tick_params(axis='y', labelsize=16)
plt.xlabel('')
plt.ylabel('top3 停车/运行时间', fontsize=16, fontweight='bold')
ax4.set_title('\nTop3停车占系统运行时间\n', fontsize=24, fontweight='bold')
# try:
# plt.bar(x,y,color=["brown", "green", "red", "grey", "purple"],alpha=0.8)
# for a,b in zip(x,y):
# ax4.text(a,b+0.02,'%.01f%%' % (b * 100),
# ha='center',
# va='bottom',
# fontsize=12)
# ax4.tick_params(axis='x', labelsize=18)
# ax4.tick_params(axis='y', labelsize=16)
# plt.ylim(0,1.0)
# plt.xlabel('')
# plt.ylabel('top3 stop/sys total time%', fontsize=18, fontweight='bold')
# plt.title('\ntop3stop/sys total time\n', fontsize=24, fontweight='bold')
except Exception as e:
print('三大停车时间/运行时间占比',e)
ax5 = plt.subplot(212)
try:
tu_5_ = df_p.groupby('OPERATOR')['OEE'].describe()
order_list = tu_5_.index.tolist()
ax = sns.boxplot(x='OPERATOR',
y='OEE',
data=df_p,
order=order_list
)
tu_5 = df_p.groupby('OPERATOR')['OEE'].median()
tu_5_ = df_p.groupby('OPERATOR')['OEE'].describe()['25%']
tu_5__ = df_p.groupby('OPERATOR')['OEE'].describe()['75%']
for i, v in enumerate(tu_5.values):
a = float(v.__str__().strip('[,]'))
b = '%.01f%%' % (a * 100)
# print(a,type(a))
plt.text(i, v, b,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_5_.values):
a = float(v.__str__().strip('[,]'))
b = '%.01f%%' % (a * 100)
# print(a,type(a))
plt.text(i, v, b,
ha='center',
va='bottom',
fontsize=12)
for i, v in enumerate(tu_5__.values):
a = float(v.__str__().strip('[,]'))
b = '%.01f%%' % (a * 100)
# print(a,type(a))
plt.text(i, v, b,
ha='center',
va='bottom',
fontsize=12)
ax5.tick_params(axis='x', labelsize=18)
ax5.tick_params(axis='y', labelsize=16)
ax.set_xticklabels(ax.get_xticklabels(), rotation=45)
plt.ylim(0, 1.2)
plt.xlabel('')
plt.ylabel('OEE/%', fontsize=16, fontweight='bold')
plt.title('\nOEE vs Stuff\n', fontsize=24, fontweight='bold')
except Exception as e:
print('OEE个人箱图',e)
plt.tight_layout()
plt.show()
return df_s1
#画图函数5，ui上点击按钮对应画图
def out11_data(line,start,end):
'''
1.白班和夜班的速度效率（速度效率---per）
2.白班和夜班FTQ（质量生产判定---fg）
3.白班和夜班RUN_TIME(runtime---per)
4.白班和夜班OEE(oee---per)
5.15天产量(real_length---fg)
6.runtime饼图(runtime---per)
'''
global savepath
df_fg = pd.read_pickle(os.path.join(savepath, f'{line}fg.pkl'))
df_fg.set_index('date', inplace=True)
df_fg = df_fg.loc[start:end]
df_fg['PRODUCTION_DATE'] = df_fg['PRODUCTION_DATE'].astype(str).str[:10]
# print(df_fg)
# df_per = pd.read_excel(os.path.join(savepath, f'{line}per.xlsx'))
df_per = pd.read_pickle(os.path.join(savepath, f'{line}per.pkl'))
df_per.set_index('date', inplace=True)
df_per = df_per.loc[start:end]
df_per['PRODUCTION_DATE']=df_per['PRODUCTION_DATE'].astype(str).str[:10]
# print(df_per)
df_stop = pd.read_pickle((os.path.join(savepath, f'{line}stop.pkl')))
df_stop['PRODUCTION_DATE'] = df_stop['PRODUCTION_DATE'].astype(str).str[:10]
df_stop.set_index('date', inplace=True)
#----------------------------------------------------
plt.figure(facecolor='white', figsize=(16, 9))
sns.set(style="darkgrid")
plt.rcParams['font.sans-serif'] = ['STZhongsong'] # 中文字体设置-黑体
plt.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
try:
ax1 = plt.subplot(231)
sns.lineplot(x='PRODUCTION_DATE',
y='SPEED_EFFICIENCY',
data=df_per,
hue='SHIFT',
marker='*',
ax=ax1)
for x,y in zip(df_per['PRODUCTION_DATE'],df_per['SPEED_EFFICIENCY']):
if y<0.95:
plt.text(x=x,y=y+0.02,s='%.01f%%' % (y * 100),color='red',size=14)
plt.xticks(ha='right', rotation=45)
plt.ylim(0, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('速度效率 / %', fontsize=14, fontweight='bold')
ax1.set_title(f'{line} Line 15天速度效率 %\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天速度效率')
ax2 = plt.subplot(232)
try:
zongshu = df_fg.groupby(['PRODUCTION_DATE', 'PRODUCTION_SHIFT'], as_index=False)['itp'].agg('count')
df_fg_hege = df_fg[df_fg['itp'] == 'False']
hegeshu = df_fg_hege.groupby(['PRODUCTION_DATE','PRODUCTION_SHIFT'],as_index=False)['itp'].agg('count')
df_ftq = pd.concat([zongshu,hegeshu],axis=1)
df_ftq.columns = ['PRODUCTION_DATE','PRODUCTION_SHIFT','zongshu','date','SHIFT','hegeshu']
df_ftq['FTQ']=df_ftq['hegeshu']/df_ftq['zongshu']
# print(df_ftq)
sns.lineplot(x='PRODUCTION_DATE',
y='FTQ',
data=df_ftq,
hue='SHIFT',
marker='*',
ax=ax2)
for x,y in zip(df_ftq['PRODUCTION_DATE'],df_ftq['FTQ']):
if y<0.1:
plt.text(x=x,y=y+0.02,s='%.01f%%' % (y * 100),color='red',size=14)
plt.xticks(ha='right', rotation=45)
plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('FTQ / %', fontsize=14, fontweight='bold')
ax2.set_title(f'{line} Line 15天 FTQ %\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天 FTQ')
ax3 = plt.subplot(233)
try:
sns.lineplot(x='PRODUCTION_DATE',
y='RUN_TIME_HOUR',
data=df_per,
hue='SHIFT',
marker='*',
ax=ax3)
for x,y in zip(df_per['PRODUCTION_DATE'],df_per['RUN_TIME_HOUR']):
plt.text(x=x,y=y,s=round(y,1),color='red',size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('run_time / h', fontsize=14, fontweight='bold')
ax3.set_title(f'{line} Line 15天 Run_Time h\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天 Run_Time')
ax4 = plt.subplot(234)
try:
sns.lineplot(x='PRODUCTION_DATE',
y='OEE',
data=df_per,
hue='SHIFT',
marker='*',
ax=ax4)
for x,y in zip(df_per['PRODUCTION_DATE'],df_per['OEE']):
plt.text(x=x,y=y,s='%.01f%%' % (y * 100),color='red',size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('OEE / %', fontsize=14, fontweight='bold')
ax4.set_title(f'{line} Line 15天 OEE %\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天 OEE')
ax5 = plt.subplot(235)
try:
ax = sns.barplot(x='PRODUCTION_DATE',
y='TOTAL_LENGTH',
data=df_per,
hue='SHIFT',
ci=0,
ax=ax5)
for p in ax.patches:
height = p.get_height()
ax.text(x=p.get_x() ,
y=height + 2,
s='{: .0f}m'.format(height), color='red', size=14)
plt.xticks(ha='right', rotation=45)
# plt.ylim(0.6, 1.2)
plt.xticks(fontsize=13)
plt.yticks(fontsize=13)
plt.xlabel('')
plt.ylabel('length / m', fontsize=14, fontweight='bold')
ax5.set_title(f'{line} Line 15天产量/ m\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天产量')
#############################################
ax6 = plt.subplot(236)
try:
bing = df_per.groupby('SHIFT')['RUN_TIME_HOUR'].agg('sum').sort_values(ascending=True)
ax6 = bing.plot(kind='pie',
subplots=True,
# labels=labels,
# explode=(0,0.1,0,0),#突显
# autopct='%.1f%%', # 设置百分比格式，保留一位
autopct=lambda x: my_label(x, bing),
pctdistance=0.8, # 设置百分比与圆心的距离
startangle=90, # 设置饼图的开始角度
radius=1, # 设置饼图的半径
counterclock=True, # 是否逆时针，false是顺时针
shadow=False, # 无阴影设置
labeldistance=1.1, # 默认1.1
# rotatelabels=True# True,标签放线是沿半径方向
textprops={'fontsize': 13}
)
plt.axis('scaled') # 'tight''scaled'
plt.ylabel('')
plt.xlabel('')
plt.title('\n15天白夜班开车时间占比\n', fontsize=18, fontweight='bold')
except Exception as e:
print(e,'15天白夜班开车时间占比')
plt.tight_layout()
plt.show()
return

复制代码

下边是ui的py文件
ui上有各数据加载的按钮，点击加载按钮，调用database文件中的3个合并文件的函数
其他的按钮就是对应出图的

from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
from PyQt5.uic import loadUiType
from threading import Thread
import database as db
import datetime
now = datetime.datetime.now().strftime('%Y-%m-%d')
now_15ago=(datetime.datetime.now()-datetime.timedelta(days=14)).strftime('%Y-%m-%d')
now_1ago=(datetime.datetime.now()-datetime.timedelta(days=1)).strftime('%Y-%m-%d')
#UI--Logic分离
ui,_ = loadUiType('main_2021new.ui')
class Signal_progressbar(QObject):
progress_update = pyqtSignal(int)
#实例化信号
so = Signal_progressbar()
class Mainapp(QMainWindow,ui):
#定义构造方法
def __init__(self):
QMainWindow.__init__(self)
self.setupUi(self)
self.handle_ui_change()
self.handle_buttons()
so.progress_update.connect(self.setProgress)
self.hebing_progressBar.setRange(0, 4)
self.ongoing = False
self.jishu_dateEdit.setDate(QDate.currentDate())
self.kaishi_dateEdit.setDate(QDate.fromString(now_15ago, "yyyy-MM-dd"))
# UI变化处理
def handle_ui_change(self):
self.tabWidget.tabBar().setVisible(True)
# 处理所按钮的消息与槽的通信
def handle_buttons(self):
self.queding_pushButton.clicked.connect(self.hebing_click)
self.tuichu_pushButton.clicked.connect(QCoreApplication.quit)
self.a1_pushButton.clicked.connect(lambda: self.out1_click('A'))
self.b1_pushButton.clicked.connect(lambda: self.out1_click('B'))
self.c1_pushButton.clicked.connect(lambda: self.out1_click('C'))
self.cb1_pushButton.clicked.connect(lambda: self.out1_click('CB'))
self.ccv1_pushButton.clicked.connect(lambda: self.out1_click('CCV'))
self.a2_pushButton.clicked.connect(lambda: self.out2_click('A'))
self.b2_pushButton.clicked.connect(lambda: self.out2_click('B'))
self.c2_pushButton.clicked.connect(lambda: self.out2_click('C'))
self.cb2_pushButton.clicked.connect(lambda: self.out2_click('CB'))
self.ccv2_pushButton.clicked.connect(lambda: self.out2_click('CCV'))
self.aa1_pushButton.clicked.connect(lambda: self.out11_click('A'))
self.bb1_pushButton.clicked.connect(lambda: self.out11_click('B'))
self.cc1_pushButton.clicked.connect(lambda: self.out11_click('C'))
self.cbcb1_pushButton.clicked.connect(lambda: self.out11_click('CB'))
self.ccvccv1_pushButton.clicked.connect(lambda: self.out11_click('CCV'))
self.aa2_pushButton.clicked.connect(lambda: self.out2_click('A'))
self.bb2_pushButton.clicked.connect(lambda: self.out2_click('B'))
self.cc2_pushButton.clicked.connect(lambda: self.out2_click('C'))
self.cbcb2_pushButton.clicked.connect(lambda: self.out2_click('CB'))
self.ccvccv2_pushButton.clicked.connect(lambda: self.out2_click('CCV'))
self.aa3_pushButton.clicked.connect(lambda: self.out3_click('A'))
self.bb3_pushButton.clicked.connect(lambda: self.out3_click('B'))
self.cc3_pushButton.clicked.connect(lambda: self.out3_click('C'))
self.cbcb3_pushButton.clicked.connect(lambda: self.out3_click('CB'))
self.ccvccv3_pushButton.clicked.connect(lambda: self.out3_click('CCV'))
self.box_pushButton.clicked.connect(lambda: self.out4_click())
def kaishi_jieshu_click(self):
start = self.kaishi_dateEdit.text()
end = self.jishu_dateEdit.text()
print(start,end)
return (start,end)
def hebing_click(self):
try:
def workerThread():
self.ongoing = True
self.hebing_progressBar.setRange(0, 3)
db.concat_file_per()
so.progress_update.emit(1)
db.concat_file_fg()
so.progress_update.emit(2)
db.concat_file_stop()
so.progress_update.emit(3)
self.ongoing = False
if self.ongoing:
QMessageBox.warning(self.window(), '警告', '任务进行中，请等待完成')
return
worker = Thread(target=workerThread)
worker.start()
except Exception as e:
print(e)
def setProgress(self, value):
self.hebing_progressBar.setValue(value)
def out1_click(self,line):#近15天
db.out1_data(line,start=now_15ago,end=now,now_1=now_1ago)
def out2_click(self,line):
start, end = self.kaishi_jieshu_click()
db.out2_data(line,start,end)
def out11_click(self,line):
start, end = self.kaishi_jieshu_click()
db.out11_data(line,start,end)
def out3_click(self,line):
start, end = self.kaishi_jieshu_click()
db.out3_data(line,start,end)
def out4_click(self):
start, end = self.kaishi_jieshu_click()
db.out4_data(start,end)
def main():
app = QApplication([])
window = Mainapp()
window.show()
app.exec_()
if __name__ == '__main__':
main()

复制代码

@hrp

rsj0315 · 发表于 2021-1-21 15:55:20

打包命令是 pyinstaller -w -D xxx.py
形成的文件夹下多个文件的形式。
问题是：
首次双击exe打开时，启动时间比较久。我打开了任务管理器后。
给其他人电脑（没有python）启动时，时间快10s了，令人发指。。

rsj0315 · 发表于 2021-1-21 15:58:57

各位帮忙看看哪款还有优化空间。
或者怎么写才更python

_2_ · 发表于 2021-1-21 21:27:49

rsj0315 发表于 2021-1-21 15:58
各位帮忙看看哪款还有优化空间。
或者怎么写才更python

你的代码好长啊，不大想看了 = =
可以找找论坛里别的打包 exe 的方法，以前似乎发过一个方案但我忘记了

hrp · 发表于 2021-1-21 21:54:29

本帖最后由 hrp 于 2021-1-21 23:33 编辑

rsj0315 发表于 2021-1-21 15:58
各位帮忙看看哪款还有优化空间。
或者怎么写才更python

目前粗略测试发现matplotlib和seaborn和pandas这三个包导入比较慢，加起来占2~3秒时间，matplotlib和seaborn占大头。
其中seaborn是对matplotlib的封装，如果可能的话把它们其中之一干掉，全部用另一个来实现功能，可以节省不少启动时间。
还有就是python环境安装太多的包，打包的时候会带入其他不需要的库，所以尝试用Anaconda3创建一个干净的虚拟python环境，只安装需要的包再打包，可能启动耗时会有好转。

技术水平有限，实在不好意思

yywwf2003 · 发表于 2021-1-21 23:59:58

这么厉害

rsj0315 · 发表于 2021-1-22 08:25:20

hrp 发表于 2021-1-21 21:54
目前粗略测试发现matplotlib和seaborn和pandas这三个包导入比较慢，加起来占2~3秒时间，matplotlib和se ...

收到感谢，画图的两个包去掉一个的话，估计也是seaborn，但是全matplotlib也画风也忒单调了。
我准备尝试安装个新的环境。打包下试试。

有一点我要说明下，我每次打包都是把这两个py文件和ui拿到桌面，放到一个文件夹里，然后开始打包的。
我要是装一个新的虚拟环境应该如何操作呢？还是拖到桌面？

hrp · 发表于 2021-1-22 09:15:02

rsj0315 发表于 2021-1-22 08:25
收到感谢，画图的两个包去掉一个的话，估计也是seaborn，但是全matplotlib也画风也忒单调了。
我准备尝 ...

文件放在哪里跟用什么py环境没有关系的。
如果你会用anaconda3的话打开anaconda3的命令行窗口，输入

conda create -n 环境名字比如py38 python=3.8

复制代码

创建一个虚拟环境，激活它，然后安装pyinstaller和运行你程序所需的第三方包，然后用pyinstaller打包，具体用法版主有发帖

weiter · 发表于 2021-1-25 21:15:18

建议把install改成from……install……？
不知道只导入需要的功能会不会好一些，毕竟导这么多包不卡是肯定不可能的
但是我的想法在您这个的规模下有点高工作量……

Ⅲ只小猪 · 发表于 2021-1-26 09:08:35

rsj0315 发表于 2021-1-21 15:58
各位帮忙看看哪款还有优化空间。
或者怎么写才更python

可以参考一下这个方法：
https://fishc.com.cn/thread-90726-1-1.html

19924269098 · 发表于 2021-1-26 09:24:41

厉害

yywwf2003 · 发表于 2021-1-26 12:33:05

hrp 发表于 2021-1-21 21:54
目前粗略测试发现matplotlib和seaborn和pandas这三个包导入比较慢，加起来占2~3秒时间，matplotlib和se ...

大佬厉害

账号		自动登录	找回密码
密码			立即注册

[技术交流] 打包exe后，启动速度慢的问题

马上注册，结交更多好友，享用更多功能^_^

评分