def generate_result_file():
dsets = []
for lang in ['pt', 'es']:
X = util.get_X_test(data_type='keras_tokenized_tri', lang=lang, file_type="dump")
model, epoch = load_lastest(lang=lang)
y_pred = util.one_hot_decode(model.predict(process_x(X)))
index = np.load('./data/test_index_'+lang+'.npy')
df = pd.DataFrame({'id': index, 'category': y_pred})
df.index = df['id']
dsets.append(df)
print('y_pred '+lang+' unique: ', len(np.unique(y_pred)))
df = pd.concat(dsets)
df = df.sort_index()
df[['id', 'category']].to_csv('./data/results-'+NAME+'.csv', index=False)
#weights for each epoch according with the number of epochs trained
weigths_epoch = {
1: [1],
2: [0.35, 0.65],
3: [0.15, 0.35, 0.5],
4: [0.1, 0.2, 0.3, 0.4],
5: [0.1, 0.15, 0.2, 0.25, 0.3]
}
num_classes = len(util.get_categories())
#Load test data for each language
data = {}
for lang in ['es', 'pt']:
X_test = util.get_X_test(data_type='keras_tokenized_tri',
lang=lang,
file_type="dump")
index = np.load(DATA_PATH + 'test_index_' + lang + '.npy')
data[lang] = {'index': index, 'X_test': process_x(X_test)}
del X_test, index
gc.collect()
paths = {}
for model_name in model_list:
PATH = DATA_PATH + 'models/' + model_name + '/'
files = {'pt': {}, 'es': {}}
if (len(os.listdir(PATH)) > 0):
for file in os.listdir(PATH):
if (file.startswith('weights')):
epoch = int(file.split('-')[1])
lang = file.split('-')[-1].split('.')[0]