def top_open_opportunities_callback(df):
edge1 = pd.read_csv('/home/ubuntu/bjur27r-gmail.com/model/upward.csv',
sep="\t",
encoding='utf8')
edge1 = pd.read_csv('/home/ubuntu/bjur27r-gmail.com/model/upward.csv',
sep="\t",
encoding='utf8')
df = edge1[['source', 'level', 'amount']].groupby(['source', 'level']).agg(
sum_trans_adrs=pd.NamedAgg(column='amount', aggfunc='sum'),
count_trans_adrs=pd.NamedAgg(column='amount', aggfunc='count'))
df = df.reset_index()
df = df[['level', 'count_trans_adrs',
'sum_trans_adrs']].groupby(['level']).agg(
count_trans_adrs_levl=pd.NamedAgg(column='count_trans_adrs',
aggfunc='count'),
mean_trans_adrs_levl=pd.NamedAgg(column='count_trans_adrs',
aggfunc='mean'),
mean_trans_val_levl=pd.NamedAgg(column='sum_trans_adrs',
aggfunc='mean'),
sum_trans_val_levl=pd.NamedAgg(column='sum_trans_adrs',
aggfunc='sum'))
# count_trans_adrs_levl: numero de address distintos por nivel.
# mean_trans_adrs_levl: transacciones medias de address.
# mean_trans_val_levl : transaccion media por nivel
# sum_trans_val_levl : suma todas las transacciones por nivel
df = df.reset_index()
df_2 = pd.DataFrame()
df_2['Level'] = df['level']
df_2['Distinct_Senders'] = round(df['count_trans_adrs_levl'], 1)
df_2['Average_Trans'] = round(df['mean_trans_adrs_levl'], 1)
df_2['Average_Trans_Value'] = round(df['mean_trans_val_levl'], 1)
df_2['Total_Trans_Value'] = round(df['sum_trans_val_levl'], 1)
return df_to_table(df_2)
def top_open_opportunities(df):
df = df.sort_values("Amount", ascending=True)
cols = ["CreatedDate", "Name", "Amount", "StageName"]
df = df[cols].iloc[:5]
# only display 21 characters
df["Name"] = df["Name"].apply(lambda x: x[:30])
return df_to_table(df)
def top_lost_opportunities(df):
df = df[df["StageName"] == "Closed Lost"]
cols = ["CreatedDate", "Name", "Amount", "StageName"]
df = df[cols].sort_values("Amount", ascending=False).iloc[:5]
# only display 21 characters
df["Name"] = df["Name"].apply(lambda x: x[:30])
return df_to_table(df)
def prod_mas_vendido(jsondf):
df = pd.read_json(jsondf, orient='split')
columns = [
'Etapa', 'Numero Unidad', 'Nombre', 'Estado', 'Programa',
'Nro Personas'
]
return df_to_table(df[columns].head())
def prod_menor_vendido(jsondf):
df = pd.read_json(jsondf, orient='split')
df = df.sort_values(by='Nro Personas', ascending=True)
columns = [
'Etapa', 'Numero Unidad', 'Nombre', 'Estado', 'Programa',
'Nro Personas'
]
return df_to_table(df[columns].head())
def tables(endDate, startDate, db, coll):
# print(type(startDate),startDate,endDate)
# if endDate==None:
# stDate = '2018-11-13'
# endDate = '2018-11-14'
print(startDate, endDate)
stDate = datetime.strptime(startDate, "%Y-%m-%d")
edDate = datetime.strptime(endDate, "%Y-%m-%d")
print("---------------", stDate, edDate)
return df_to_table(cpkinitalTable(stDate, edDate, db, coll))
def leads_table_callback(status, df):
df = pd.read_json(df, orient="split")
if status == "open":
df = df[(df["Status"] == "Open - Not Contacted")
| (df["Status"] == "Working - Contacted")]
elif status == "converted":
df = df[df["Status"] == "Closed - Converted"]
elif status == "lost":
df = df[df["Status"] == "Closed - Not Converted"]
df = df[["CreatedDate", "Status", "Company", "State", "LeadSource"]]
return df_to_table(df)
def nodes_des(df):
df_node_3 = df_nodes()
df_nodes_out = df_node_3[['level', 'label', 'cluster', 'prob_1']].groupby([
'level', 'label', 'cluster'
]).agg(LookAlike=pd.NamedAgg(column='prob_1', aggfunc='median'))
df_nodes_out = df_nodes_out.reset_index()
df_nodes_out['Level'] = df_nodes_out['level']
df_nodes_out['LookAlike'] = round(df_nodes_out['LookAlike'] * 100, 1)
df_nodes_out['Sannity Check'] = df_nodes_out['LookAlike'].apply(
lambda x: 'SUSPICIOUS' if x > 50 else "OK")
return df_to_table(df_nodes_out)
def top_lost_opportunities_callback(df):
df = pd.read_json(df, orient="split")
# df_clust= clusterice(df,10)
# df_clust.to_csv("c:\BC_DS\df_result_b.csv", sep = "\t", encoding='utf8')
df['Cluster'] = df['cluster']
df['Sample_Share(%)'] = round(df['cluster_Pob'] * 100, 0)
df['Lifetime(days)'] = round(df['lifetime_d'], 0)
df['Trans_In(x)'] = round(df['TX_IN'], 0)
df['Trans_Out(x)'] = round(df['TX_OUT'], 0)
df['Ratio_IN/OUT(x)'] = round(df['Trans_In(x)'] / df['Trans_Out(x)'], 2)
df['Average_Value_In(USD)'] = round(df['value_IN'] / df['Trans_In(x)'], 0)
df['Average_Value_Out(USD)'] = round(df['value_OUT'] / df['Trans_Out(x)'], 0)
df['Phising Lookalike(%)'] = round(df['prob_1'] * 100, 0)
df['balance_mean'] = round(df['TX_OUT'], 0)
df_Out = df[['Cluster', 'Sample_Share(%)', 'Lifetime(days)', 'Trans_In(x)', 'Trans_Out(x)',
'Ratio_IN/OUT(x)', 'Average_Value_In(USD)', 'Average_Value_Out(USD)', 'Phising Lookalike(%)']]
return df_to_table(df_Out)
def strategy_table_callback(df):
df = pd.read_json(df, orient="split")
return df_to_table(df)
margin=dict(
l=100, r=25, b=20, t=0, pad=4),
paper_bgcolor='white',
plot_bgcolor='white'))),
],
className="four columns chart_div"),
],
className="row",
style={"marginTop": "5"},
),
html.Div(
id="leads_table",
# className="row",
children=[
df_to_table(
cpkinitalTable(datetime(2018, 10, 12),
datetime(2018, 10, 13), '1521900003T0',
'DsQAM'))
],
style={
"maxHeight": "550px",
"overflowY": "scroll",
"padding": "10",
"marginTop": "5",
"backgroundColor": "white",
"border": "1px solid #C8D4E3",
"borderRadius": "3px"
},
),
modal(),
html.Div([
html.Div([
"width": '100%',
"Height": '40%'
}),
],
style={
"display": "inline-block",
"margin-left": 30,
"width": '45%'
})
],
),
#TM列表
html.Div(
id="TM_Table",
className="chart_div pretty_container",
children=df_to_table(),
),
#TM图表
html.Div(
id="TM_Chart",
className="chart_div pretty_container",
children=[
dcc.Graph(
id='lead_source',
figure={
'data': [
{
'x': [
"Topic 1", "Topic 2", "Topic 3",
"Topic 4", "Topic 5"
],
