import numpy as np
import pandas as pd
import seaborn as sns
import statsmodels.formula.api as sm
import statsmodels as stamo
import patsy
from statsmodels.regression.linear_model import OLS
from statsmodels.sandbox.regression.predstd import wls_prediction_std
from collections import defaultdict
import matplotlib.pyplot as plt
from sklearn.feature_selection import VarianceThreshold
from sklearn.preprocessing import PolynomialFeatures
from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import chi2
from sklearn import linear_model
from sklearn.linear_model import Ridge
from sklearn.metrics import mean_squared_error, r2_score
from ggplot import *
import regressions

myds = pd.read_csv('all_vars_with_social.csv')

data_frame = pd.DataFrame({
    "spotify_passive_streams": myds['SPOTIFY_STREAMS_PASSIVE_DISC_TOTAL'],
    "spotify_playlist_streams": myds['SPOTIFY_PLAYLISTS_STREAMS_TOTAL'],
    "spotify_active_streams": myds['SPOTIFY_STREAMS_ACTIVE_DISC_TOTAL'],
    "spotify_collection_streams": myds['SPOTIFY_STREAMS_COLLECTION_TOTAL'],
    "apple_passive_streams": myds['APPLE_STREAMS_PASSIVE_DISC_TOTAL'],
    "apple_active_streams": myds['APPLE_STREAMS_ACTIVE_DISC_TOTAL'],
    "apple_collection_streams": myds['APPLE_STREAMS_COLLECTION_TOTAL'],
    "apple_playlist_streams": myds['APPLE_PLAYLISTS_STREAMS_TOTAL'],
    "amazon_total_streams": myds['AMAZON_OVERALL_STREAMS_TOTAL'],
    "pandora_total_streams": myds['PANDORA_OVERALL_STREAMS_TOTAL'],
    "youtube_audio_streams": myds['YOUTUBE_AUDIO_OVERALL_STREAMS_TOTAL'],
    "shazam_total_streams": myds['SHAZAM_OVERALL_STREAMS_TOTAL'],
    "alexa_total_streams": myds['ALEXA_OVERALL_STREAMS_TOTAL'],
    "itunes_total_downloads": myds['ITUNES_TOTAL_DOWNLOADS'],
    "amazon_total_downloads": myds['AMAZON_TOTAL_DOWNLOADS'],
    "google_total_downloads": myds['GOOGLE_PLAY_DOWNLOADS_TOTAL'],
    "total_gross_revenue_last_year": myds['TOTAL_GROSS_REVENUE_USD_LAST_YEAR'],
    "apple_playlists_streams_last_year": myds['APPLE_PLAYLISTS_STREAMS_LAST_YEAR'],
    "spotify_playlists_streams_last_year": myds['SPOTIFY_PLAYLISTS_STREAMS_LAST_YEAR'],
    "total_physical_rev": myds['PHYSICAL_GROSS_REVENUE'],
    "alexa_overall_streams_last_year": myds['ALEXA_OVERALL_STREAMS_LAST_YEAR'],
    "wiki_pageviews_ly": myds['WIKI_PAGEVIEWS_LY'],
    "songkick_followers_ly": myds['SONGKICK_FOLLOWERS_LY'],
    "facebook_likes_total": myds['FB_LIKES_LTD'],
    "instagram_followers_total": myds['INSTA_FOLLOWERS_LTD'],
    'y': myds['TOTAL_GROSS_REVENUE'],
    # "release_genre": myds["RELEASE_GENRE"],
    # "spotify_total_streams": myds["SPOTIFY_OVERALL_STREAMS_TOTAL"],
    # "apple_total_streams": myds["APPLE_OVERALL_STREAMS_TOTAL"],
})

data_frame.fillna(0, inplace=True)
data_frame = data_frame.replace(0, 1)
genres = pd.get_dummies(myds["RELEASE_GENRE"])
rock = genres['Rock']
country = pd.get_dummies(myds["LABEL_COUNTRY"])
usa = country["USA"]
data_frame = np.log(data_frame)
# data_frame['y'] = myds['TOTAL_GROSS_REVENUE']
data_frame = data_frame.replace([np.inf, -np.inf, np.nan], 0)
data_frame = pd.concat([data_frame, rock, usa], axis=1)
#
# fml = "y ~ np.log(total_physical_rev) + np.log(spotify_playlist_streams) + np.log(spotify_active_streams) + np.log(spotify_passive_streams) + np.log(spotify_collection_streams) + np.log(" \
#       "apple_active_streams) + np.log(apple_passive_streams) + np.log(apple_collection_streams) + np.log(pandora_total_streams) + np.log(shazam_total_streams)"



# Ols Regulization
# model = sm.ols(fml, data=data_frame)
# unregularized_regression = model.fit()
# result = sm.ols(fml, data=data_frame).fit_regularized(method='elastic_net', L1_wt=0, alpha=5.0, start_params=unregularized_regression.params, profile_scale=False, refit=False)
# print(result.params)
# final = stamo.regression.linear_model.OLSResults(model,
#                                               result.params,
#                                               model.normalized_cov_params)
# print(final.summary())


alpha_ridge = [1e-15, 1e-10, 1e-8, 1e-4, 1e-3, 1e-2, (1e-2 * 2), (1e-2 * 4), (1e-2 * 8), 1, 5, 10, 15, 20]
#
# predictors = ["spotify_passive_streams", "spotify_playlist_streams", "spotify_active_streams", "spotify_collection_streams",
# "apple_passive_streams", "apple_active_streams", "apple_collection_streams", "apple_playlist_streams",
# "amazon_total_streams", "pandora_total_streams", "youtube_audio_streams", "shazam_total_streams",
# "alexa_total_streams", "itunes_total_downloads", "amazon_total_downloads", "google_total_downloads",
# "total_gross_revenue_last_year", "apple_playlists_streams_last_year", "spotify_playlists_streams_last_year",
# "alexa_overall_streams_last_year", "Metal", "Blues", "Classical", "Country", "Electronic", "Hip-hop/Rap",
# "New Age", "Pop", "Punk", "R&B", "Rock", "wiki_pageviews_ly", "songkick_followers_ly", "facebook_likes_total", "instagram_followers_total"]
##

predictors = ["spotify_passive_streams", "spotify_playlist_streams", "spotify_active_streams", "spotify_collection_streams",
"apple_passive_streams", "apple_active_streams", "apple_collection_streams", "apple_playlist_streams",
"amazon_total_streams", "pandora_total_streams", "youtube_audio_streams", "shazam_total_streams",
"alexa_total_streams", "itunes_total_downloads", "amazon_total_downloads", "google_total_downloads",
"total_gross_revenue_last_year", "apple_playlists_streams_last_year", "spotify_playlists_streams_last_year",
"alexa_overall_streams_last_year", "Rock", "USA"]


for i in range(len(alpha_ridge)):
    print(regressions.ridge_regression(data_frame, predictors, alpha_ridge[i]))