import pandas as pd
import numpy as np
import statsmodels.api as sm
from biglearning.module2.common.data import Outputs
from zipline.finance.commission import PerOrder
import os
import copy


from bigdatasource.api import DataSource
from biglearning.api import M
from biglearning.api import tools as T
from biglearning.module2.common.data import Outputs
from joblib import Parallel, delayed

import warnings
warnings.filterwarnings('ignore')

#import fai
# 连接和初始化fai
#fai.init(cluster="fai-anthonywan-emzbzcpl.fai-anthonywan",token="exEOSXJhSFJGAJYYINsu36YDFwKxGhKW")

sd = '2022-01-04'
ed = '2022-12-31'

def calc_data(instrument,sd,ed):
    import warnings
    warnings.filterwarnings('ignore')


    try:
        df = DataSource('bar1m_CN_STOCK_A').read(start_date=sd,end_date=ed,instruments = instrument)

        def data_preprocess(df):
            df['分钟收益率'] = (df['close']/df['close'].shift())-1
            df['增量成交量'] = df['volume'] - df['volume'].shift()
            df['增量成交量t-1'] = df['增量成交量'].shift()
            df['增量成交量t-2'] = df['增量成交量'].shift(2)
            df['增量成交量t-3'] = df['增量成交量'].shift(3)
            df['增量成交量t-4'] = df['增量成交量'].shift(4)
            df['增量成交量t-5'] = df['增量成交量'].shift(5)

            return df

        df['year'] = df.date.dt.year
        df['month'] = df.date.dt.month
        df['day'] = df.date.dt.day
        df = df.groupby(['year','month','day']).apply(data_preprocess)

        df.dropna(inplace=True)

        def calc_factor(df):
            y = np.array(df['分钟收益率'])

            x = df[['增量成交量','增量成交量t-1','增量成交量t-2','增量成交量t-3','增量成交量t-4','增量成交量t-5']]
            X = sm.add_constant(x)
            result = sm.OLS(y,X).fit()

            inter_ = result.params.const
            thea0 = result.params[1]
            thea1 = result.params[2]
            thea2 = result.params[3]
            thea3 = result.params[4]
            thea4 = result.params[5]
            thea5 = result.params[6]
            f_value = result.fvalue

            #回归系数 F-Value intercept 赋值
            df['inter_'] = inter_
            df['thea0'] = thea0
            df['thea1'] = thea1
            df['thea2'] = thea2
            df['thea3'] = thea3
            df['thea4'] = thea4
            df['thea5'] = thea5
            df['f_value'] = f_value
            std_ = np.std([thea1,thea2,thea3,thea4,thea5])
            
            #计算因子和截距的绝对值
            df['factor1'] = std_
            df['abs_inter'] = abs(df['inter_'])

            return df
        
        df = df.groupby(['year','month','day']).apply(calc_factor)
        df.drop_duplicates(subset=['year','month','day'],inplace=True)

        df['date'] = df['date'].dt.to_period('D')
        df = df[['instrument','date','inter_','thea0','thea1','thea2','thea3','thea4','thea5','f_value','factor1']]

        def f_mean(df):
            df['f_mean'] = df['f_value'].mean()
            return df

        df = df.groupby('date').apply(f_mean)

        df['factor2'] = np.where(df['f_value']<df['f_mean'], abs(df['inter_']) *-1,df['inter_']  )

        return df
        
    except:
        return
    


    

df = calc_data('300170.SZA',sd,ed)
df

df

tmp = DataSource('bar1d_CN_STOCK_A').read(start_date='2022-01-01',end_date='2022-12-31')
instrument_list = tmp.instrument.unique()

results = Parallel(n_jobs=32)(delayed(calc_data)(ins,sd,ed) for ins in instrument_list)

df_all = pd.concat(results)
df_all.reset_index(inplace=True,drop=True)
df_all['date'] = df_all['date'].dt.to_timestamp()

m1 = M.instruments.v2(
    start_date=sd,
    end_date=ed,
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m2 = M.input_features.v1(
    features="""
收盘=close_0

市值=market_cap_0

行业=industry_sw_level1_0

"""
)


m3 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m2.data,
    start_date='',
    end_date='',
    before_start_days=500
)

m4 = M.derived_feature_extractor.v3(
    input_data=m3.data,
    features=m2.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m5 = M.chinaa_stock_filter.v1(
    input_data=m4.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

df = m5.data.read()

df = pd.merge(df_all,df[['instrument','date','收盘','市值','行业']],left_on=['instrument','date'],right_on=['instrument','date'],how='left')

df.sort_values(by='date',inplace=True)

df_2 = df.copy()

df.dropna(inplace=True)

df.describe()

import plotly.graph_objs as go
import numpy as np


# 使用 Histogram 类创建概率分布图
fig = go.Figure(data=[go.Histogram(x=df.factor1, histnorm='probability')])

# 设置图形的标题和轴标签
fig.update_layout(title='Probability Distribution', xaxis_title='Value', yaxis_title='Probability')

# 显示图形
fig.show()