shutil.copy(image_filename_source, image_filename_dest)
print(image_filename_dest)
print('read csv file, copy subclass10 to dir OK!')
#region preprocess 512
# dir_original = '/tmp2/SubClass10/original/'
dir_original = dir_dest
dir_preprocess512 = '/tmp2/SubClass10/preprocess512/'
if DO_PREPROCESS:
from LIBS.ImgPreprocess.my_preprocess_dir import do_process_dir
do_process_dir(dir_original, dir_preprocess512, image_size=512)
print('preprocess OK!')
#endregion
#region crop optic disc 112
dir_source = dir_preprocess512
dir_dest = '/tmp2/SubClass10_new/Crop_optic_disc_112/'
crop_optic_disc_dir(dir_source=dir_source, dir_dest=dir_dest, server_port=21000, mask=True)
print('crop optic disc 112 OK!')
#endregion
import os, sys, csv
import pandas as pd
from sklearn.utils import shuffle
from LIBS.DataPreprocess import my_data
DO_PREPROCESS = True
GENERATE_CSV = True
from LIBS.ImgPreprocess.my_preprocess_dir import do_process_dir
dir_original = '/media/ubuntu/data2/其它数据集/筛查集//original'
dir_preprocess = '/media/ubuntu/data2/其它数据集/筛查集/preprocess384'
if DO_PREPROCESS:
do_process_dir(dir_original,
dir_preprocess,
image_size=384,
add_black_pixel_ratio=0.02)
filename_csv = 'screening.csv'
current_dir = os.path.abspath(os.path.dirname(__file__))
filename_csv = os.path.join(current_dir, filename_csv)
from LIBS.DataPreprocess.my_data import write_csv_dir_nolabel
if GENERATE_CSV:
write_csv_dir_nolabel(filename_csv, dir_preprocess)
print('OK')
filename_csv = os.path.abspath(os.path.join(sys.path[0], "..",
'datafiles', predict_type_name + '.csv'))
filename_csv = os.path.abspath(os.path.join(sys.path[0], "..",
'datafiles', 'Subclass_0.3_a.csv'))
dir_dest_confusion = os.path.join(DIR_DEST_BASE, predict_type_name, 'confusion_matrix', 'files')
dir_dest_predict_dir = os.path.join(DIR_DEST_BASE, predict_type_name, 'dir')
pkl_prob = os.path.join(DIR_DEST_BASE, predict_type_name + '_prob.pkl')
pkl_confusion_matrix = os.path.join(DIR_DEST_BASE, predict_type_name + '_cf.pkl')
if DO_PREPROCESS:
from LIBS.ImgPreprocess import my_preprocess_dir
image_size = 512
my_preprocess_dir.do_process_dir(dir_original, dir_preprocess, image_size=image_size)
if GEN_CSV:
if not os.path.exists(os.path.dirname(filename_csv)):
os.makedirs(os.path.dirname(filename_csv))
if GET_LABELS_FROM_DIR:
dict_mapping = {}
for i in range(30):
dict_mapping[str(i)] = str(i)
my_data.write_csv_based_on_dir(filename_csv, dir_preprocess, dict_mapping)
else:
my_data.write_csv_dir_nolabel(filename_csv, dir_preprocess)
import sys, os
import pandas as pd
import csv
from LIBS.DataPreprocess.my_data import get_big_classes
DO_PREPROCESS = False
dir_original = '/media/ubuntu/data1/multi_labels_2919_1_15/'
dir_preprocess = '/home/ubuntu/multi_labels_2919_1_15/preprocess384/'
if DO_PREPROCESS:
from LIBS.ImgPreprocess.my_preprocess_dir import do_process_dir
do_process_dir(dir_original, dir_preprocess, image_size=299)
def gen_subclass_csv(filename_csv_all,
filename_csv_subclass,
subclass_no,
one_one_class=True):
str_subclass_no = str(subclass_no)
df = pd.read_csv(filename_csv_all)
if os.path.exists(filename_csv_subclass):
os.remove(filename_csv_subclass)
with open(filename_csv_subclass, 'w', newline='') as csvfile:
csv_writer = csv.writer(csvfile, delimiter=',')
csv_writer.writerow(['images', 'labels'])
for i, row in df.iterrows():
labels = row["labels"]
DO_PREPROCESS = True
GEN_CSV = True
COMPUTE_DIR_FILES = True
DIR_DEST_BASE = '/tmp5/测试集分子类/results'
dir_original = '/tmp5/测试集分子类/original'
dir_preprocess = '/tmp5/测试集分子类/preprocess384'
if DO_PREPROCESS:
from LIBS.ImgPreprocess import my_preprocess_dir
image_size = 384
my_preprocess_dir.do_process_dir(dir_original,
dir_preprocess,
image_size=image_size,
add_black_pixel_ratio=0.02)
print('Preprocess OK')
for subclass_type in ['0.1', '0.2', '1', '2', '29']:
# for subclass_type in ['0.1', '0.2', '1', '2', '5', '15', '29']:
dir_original_subclass = os.path.join(dir_original, subclass_type)
dir_preprocess_subclass = os.path.join(dir_preprocess, subclass_type)
predict_type_name = 'Subclass' + subclass_type
filename_csv = os.path.join(DIR_DEST_BASE, predict_type_name + '.csv')
if GEN_CSV:
if not os.path.exists(os.path.dirname(filename_csv)):