from feature_extract.features import(

calc_chroma_features

)

from utils.read import(

csv_2_dict_va,

read_csv_song_features,

read_eric_va,

read_fake_chroma

)

from utils.calc_utils import(

find_a_v_mens_va,

regression,

average_distance_va,

nearest_dist_average_va,

no_stdev_average_va,

find_in_dict,

valence_distance_va,

arousal_distance_va

)

from utils.plot import(

plot_all_va

)

import numpy as np

import random

import sys

# y, sr = load_files('audio/101.mp3')

# mfcc_v = mfcc(y, sr)

# get exsisting valence and arousal data

valence, arousal = csv_2_dict_va('csv/survery2dataMin1.csv')

# calculate fetures for song in train set

ids, feat = read_fake_chroma('features/fakechroma')

best_avg = sys.maxint

best_near = sys.maxint

best_std = sys.maxint

best_val = sys.maxint

best_aro = sys.maxint

for i in range(50):

train_ids = ids

random.shuffle(train_ids)

all_ids = train_ids[141:]

train_ids = train_ids[0:140]

# calcultae valence and arousal find_a_v_mens

val_mean, aro_mean = find_a_v_mens_va(train_ids, valence, arousal)

train_feat = find_in_dict(feat, train_ids)

test_feat = find_in_dict(feat, all_ids)

# use regression

X_v, X_a = regression(train_feat, val_mean, aro_mean)

# calculating features for whole dataset

#print all_feat.shape

# use regresion function to calculate v and a

all_val = np.sum(np.array(test_feat) * X_v, axis=1)

all_aro = np.sum(test_feat * X_a, axis=1)

#print all_val.shape

#print all_aro.shape

print "ATTEMPT" + str(i)

avg = average_distance_va(all_val, all_aro, valence, arousal, all_ids)

nearest = nearest_dist_average_va(all_val, all_aro, valence, arousal, all_ids)

standdev = no_stdev_average_va(all_val, all_aro, val_mean, aro_mean, valence, arousal, all_ids)

valence_dist = valence_distance_va(all_val, all_aro, valence, arousal, all_ids)

arousal_dist = arousal_distance_va(all_val, all_aro, valence, arousal, all_ids)

print 'Average distance: ' + str(avg)

print 'Nearest distance: ' + str(nearest)

print 'Nearest distance: ' + str(standdev)

if avg < best_avg:

best_avg = avg

best_near = nearest

best_std = standdev

if best_val > valence_dist:

best_val = valence_dist

if best_aro > arousal_dist:

best_aro = arousal_dist

plot_all_va(all_val, all_aro, all_ids, valence, arousal)

print "BEST"

print 'Average distance: ' + str(best_avg)

print 'Nearest distance: ' + str(best_near)

print 'Std distance: ' + str(best_std)

print best_val

print best_aro