thtrieu / darkflow

Translate darknet to tensorflow. Load trained weights, retrain/fine-tune using tensorflow, export constant graph def to mobile devices
GNU General Public License v3.0
6.13k stars 2.09k forks source link

Training error #1186

Open youngDaLee opened 3 years ago

youngDaLee commented 3 years ago

Hi, I have a problem when trying to train on a custom dataset. How can I fix this problem? thanks for any help.

C:\darkflow>python flow --model ./cfg/my-tiny-yolo.cfg --labels ./labels.txt --trainer adam --dataset ../data/dataset/ --annotation ../data/annotations/ --train --summary ./logs --batch 5 --epoch 100 --save 50 --keep 5 --lr 1e-04

Enter training ...

./cfg/my-tiny-yolo.cfg parsing ../data/annotations/ Parsing for ['stairs'] [=> ]7% stairs_1.xmlTraceback (most recent call last): File "flow", line 6, in cliHandler(sys.argv) File "C:\darkflow\darkflow\cli.py", line 33, in cliHandler print('Enter training ...'); tfnet.train() File "C:\darkflow\darkflow\net\flow.py", line 39, in train for i, (x_batch, datum) in enumerate(batches): File "C:\darkflow\darkflow\net\yolo\data.py", line 97, in shuffle data = self.parse() File "C:\darkflow\darkflow\net\yolo\data.py", line 18, in parse dumps = pascal_voc_clean_xml(ann, meta['labels'], exclusive) File "C:\darkflow\darkflow\utils\pascal_voc_clean_xml.py", line 37, in pascal_voc_clean_xml tree=ET.parse(in_file) File "C:\Users\ldy09\anaconda3\envs\wiset\lib\xml\etree\ElementTree.py", line 1195, in parse tree.parse(source, parser) File "C:\Users\ldy09\anaconda3\envs\wiset\lib\xml\etree\ElementTree.py", line 596, in parse self._root = parser._parse_whole(source) UnicodeDecodeError: 'cp949' codec can't decode byte 0xeb in position 96: illegal multibyte sequence