from data import data, Dataset

import evaluate

from sklearn.metrics import accuracy_score,average_precision_score,cohen_kappa_score,hamming_loss,roc_auc_score,recall_score,confusion_matrix,precision_recall_curve,auc

import scipy.misc as misc

import efficientnet.keras as efn

import os

os.environ["CUDA_VISIBLE_DEVICES"]="1"

import numpy as np

from glob import glob

import pandas as pd

import sklearn.metrics as metrics

import joblib

f = open('./train_set.txt', 'r')

fv = open('./validation_set.txt', 'r')

ft = open('./test_set.txt', 'r')

ft_test = ft.readlines()

fv_set = fv.readlines()

f_set = f.readlines()

test_set = []

validation_set = []

train_set = []

for line in ft_test:

line = line.replace('\n', '')

test_set.append(line)

test_set.sort()

for line in fv_set:

line = line.replace('\n', '')

validation_set.append(line)

validation_set.sort()

for line in f_set:

line = line.replace('\n', '')

train_set.append(line)

print(train_set)

print(test_set)

print(validation_set)

seed = config.SEED

def predict(df):

return features,truth_label

from keras.models import load_model

model = load_model('/media/user1/model.h5')

test_features,test_label = predict(test_set)

np.save('./test_each.npy',test_features)

np.save('./test_label.npy',test_label)

validation_features,validation_label = predict(validation_set)

np.save('./validation_each.npy',validation_features)

np.save('./validation_label.npy',validation_label)

train_features, train_label = predict(train_set)

np.save('./train_each.npy',train_features)

np.save('./train_label.npy',train_label)