def gen_deeper_pipeline_predictor_test(tongue_image_arrays,
tongue_yaofangs,
tongue_image_shape,
nb_yao,
trained_gen_model,
use_tfidf_tensor=False):
'''
@param use_tfidf_tensor: flag of use tfidf tensor or not with different tensorization function
'''
''' load test_x & test_y '''
if use_tfidf_tensor == True:
total_x, total_y = tongue2text_gen.data_tensorization_tfidf(
tongue_image_arrays, tongue_yaofangs, tongue_image_shape, nb_yao)
else:
total_x, total_y = tongue2text_gen.data_tensorization(
tongue_image_arrays, tongue_yaofangs, tongue_image_shape, nb_yao)
# test_x = total_x[: 500]
# test_x = total_x[2000 : 2500]
# test_x = total_x[4000 : 4500]
# test_x = total_x[6000 : 6500]
test_x = total_x[len(total_x) - 500:]
gen_output = tongue2text_gen.predictor(trained_gen_model, test_x)
return gen_output
def gen_biggertimes_cnn_withlda_predictor_test(tongue_image_arrays, tongue_yaofangs, tongue_image_shape, nb_yao, trained_gen_model,
lda_model_path, use_tfidf_tensor=False):
'''
@param use_tfidf_tensor: flag of use tfidf tensor or not with different tensorization function
@return: gen_output_list: [gen_output, aux_output]
'''
# in test process, lda_model and dictionary can be only load from disk(can
# not be replaced)
lda_model, dictionary = lda.loadModelfromFile(
lda_model_path, readOnly=True)
total_tongue_x, total_y, total_aux_y = tongue2text_gen.data_tensorization_lda(
tongue_image_arrays, tongue_yaofangs, tongue_image_shape, nb_yao, lda_model.num_topics,
lda_model, dictionary, use_tfidf_tensor=use_tfidf_tensor)
# test_tongue_x = total_tongue_x[: 500]
# test_tongue_x = total_tongue_x[2000 : 2500]
# test_tongue_x = total_tongue_x[4000 : 4500]
# test_tongue_x = total_tongue_x[6000 : 6500]
test_tongue_x = total_tongue_x[len(total_tongue_x) - 500:]
gen_output_list = tongue2text_gen.predictor(
trained_gen_model, test_tongue_x)
# gen_output in here is a tuple as (main_output, aux_output), get the
# first one
return gen_output_list