unsubscribe_predict/data_analysize.py

import pandas as pd
from common.log_utils import logFactory
from common.database_utils import database_util
from common import constant
import pickle
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

click_client = database_util.get_client()
logger = logFactory("data analysis").log

if __name__:
    # train_pos_data = pd.read_pickle('./data/pkl/train_pos_all.pkl')
    # train_neg_data = pd.read_pickle('./data/pkl/train_neg_all.pkl')
    # train_neg_data = pd.concat(
    #     [train_neg_data, train_neg_data, train_neg_data, train_neg_data, train_neg_data, train_neg_data, train_neg_data,
    #      train_neg_data, train_neg_data, train_neg_data], axis=0)
    #
    # train_neg_data['mark'] = 0
    # train_pos_data['mark'] = 1
    #
    # all_data = pd.concat([train_pos_data, train_neg_data], axis=0)
    all_data = pd.read_pickle('./data/pkl/analysize_all.pkl')

    # first_EVENT_FLUX_V
    # second_EVENT_FLUX_V
    # temp = pd.melt(all_data[['first_EVENT_FLUX_V', 'second_EVENT_FLUX_V', 'mark']], id_vars='mark', var_name="Features",
    #                value_name="Values")
    # sns.violinplot(
    #     x="Features",
    #     y="Values",
    #     hue="mark",
    #     data=temp,
    #     split=True,
    #     palette='muted'
    # )
    # plt.show()
    #
    # # first_EVENT_CONSUM_V
    # # second_EVENT_CONSUM_V
    # # all_data = all_data[(all_data.EVENT_FLUX_V < 100000) & (all_data.EVENT_FLUX_V > 0)]
    # temp = all_data[(all_data.first_EVENT_CONSUM_V < 300) & (all_data.second_EVENT_CONSUM_V < 300)]
    # temp = pd.melt(temp[['first_EVENT_CONSUM_V', 'second_EVENT_CONSUM_V', 'mark']], id_vars='mark',
    #                var_name="Features",
    #                value_name="Values")
    # sns.violinplot(
    #     x="Features",
    #     y="Values",
    #     hue="mark",
    #     data=temp,
    #     split=True,
    #     palette='muted'
    # )
    # plt.show()

    # temp = all_data[(all_data.first_EVENT_VIDEO_FLUX_V < 100) & (all_data.second_EVENT_VIDEO_FLUX_V < 100)]
    # temp = pd.melt(temp[['first_EVENT_VIDEO_FLUX_V', 'second_EVENT_VIDEO_FLUX_V', 'mark']], id_vars='mark',
    #                var_name="Features",
    #                value_name="Values")
    # sns.violinplot(
    #     x="Features",
    #     y="Values",
    #     hue="mark",
    #     data=temp,
    #     split=True,
    #     palette='muted'
    # )
    # plt.show()

    # temp = pd.melt(all_data[['first_MAvg_TOTAL_FLUX_1_3_zs', 'second_MAvg_TOTAL_FLUX_1_3_zs',
    #                      'first_MPer1_TOTAL_FLUX_zs', 'second_MPer1_TOTAL_FLUX_zs',
    #                      'mark']], id_vars='mark',
    #                var_name="Features",
    #                value_name="Values")
    # sns.violinplot(
    #     x="Features",
    #     y="Values",
    #     hue="mark",
    #     data=temp,
    #     split=True,
    #     palette='muted'
    # )
    # sns.set(rc={'figure.figsize': (30, 30)})
    # plt.show()
    #
    # temp = pd.melt(all_data[[
    #                          'first_MAvg_TOTAL_VIDEO_FLUX_1_3_zs', 'second_MAvg_TOTAL_VIDEO_FLUX_1_3_zs',
    #                          'first_MPer1_TOTAL_VIDEO_FLUX_zs', 'second_MPer1_TOTAL_VIDEO_FLUX_zs',
    #                          'mark']], id_vars='mark',
    #                var_name="Features",
    #                value_name="Values")
    # sns.violinplot(
    #     x="Features",
    #     y="Values",
    #     hue="mark",
    #     data=temp,
    #     split=True,
    #     palette='muted'
    # )
    # sns.set(rc={'figure.figsize': (30, 30)})
    # plt.show()
    #
    # temp = pd.melt(all_data[[
    #                          'first_MAvg_Flow_kuaishou_1_3_zs', 'second_MAvg_Flow_kuaishou_1_3_zs',
    #                          'first_MPer1_Flow_kuaishou_zs', 'second_MPer1_Flow_kuaishou_zs',
    #                          'first_Div_kuaishou_vFlux_1_3', 'second_Div_kuaishou_vFlux_1_3',
    #                          'mark']], id_vars='mark',
    #                var_name="Features",
    #                value_name="Values")
    # sns.violinplot(
    #     x="Features",
    #     y="Values",
    #     hue="mark",
    #     data=temp,
    #     split=True,
    #     palette='muted'
    # )
    # sns.set(rc={'figure.figsize': (30, 30)})
    # plt.show()

#, 'diff_EVENT_CONSUM_V',
                             # 'diff_EVENT_VIDEO_FLUX_V', 'diff_kuaishou_use',
    temp = all_data[(all_data.diff_EVENT_FLUX_V < 50000) & (all_data.diff_EVENT_FLUX_V < 50000)]
    temp = pd.melt(all_data[[
                             'diff_EVENT_FLUX_V',
                             'mark']], id_vars='mark',
                   var_name="Features",
                   value_name="Values")
    sns.violinplot(
        x="Features",
        y="Values",
        hue="mark",
        data=temp,
        split=True,
        palette='muted'
    )
    plt.show()

    temp = all_data[(all_data.diff_EVENT_VIDEO_FLUX_V < 50) & (all_data.diff_EVENT_VIDEO_FLUX_V < 50)]
    temp = pd.melt(all_data[[
                             'diff_EVENT_VIDEO_FLUX_V',
                             'mark']], id_vars='mark',
                   var_name="Features",
                   value_name="Values")
    sns.violinplot(
        x="Features",
        y="Values",
        hue="mark",
        data=temp,
        split=True,
        palette='muted'
    )
    plt.show()

    temp = pd.melt(all_data[[
        'diff_kuaishou_use',
        'mark']], id_vars='mark',
                   var_name="Features",
                   value_name="Values")
    sns.violinplot(
        x="Features",
        y="Values",
        hue="mark",
        data=temp,
        split=True,
        palette='muted'
    )
    plt.show()

    pass