Wow, thank you so much Dat, I really appreciate your help!
Rasmus Møhring Madsen
1

So I just converted the images to xml and also did the training. No problem on my side… Here is the data that I generated and this is the command that I used for training:

python object_detection/train.py \
— logtostderr \
— pipeline_config_path=object_detection/samples/configs/ssd_mobilenet_v1_pets.config \
— train_dir=object_detection/train_test/
One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.