Samjith Raj CP
Samjith Raj CP
Have you tried with lesser batch size?
@mingxingtan @fsx950223 I have the same doubt object size 8x8 Image resolution : 3840x2160
@gtnbssn Did you trained on a custom dataset?
Getting the same error, they are using some old version of tensorflow , so you have to downgrade the version . (They didn't specify the tf version ) `tensorflow 19.10...
Sorry . I meant the Nvidia tf docker container version You can try tf 1.9 . It's working for me
> Moin, > > please take a lock at: > https://github.com/ChristianMarzahl/ObjectDetection/blob/master/examples/CocoTiny_Retina_Net.ipynb > > That model is trained with the COCO format. I got following error, ``` from BoundingBox import BoundingBox...
> Please make sure that you added the lib folder to your system path "****/Object-Detection-Metrics/lib" My custom dataset is in kitti format. I have a python script which read those...
> > > Not sure why this is happening. I don't temember encountering this. https://github.com/fizyr/keras-retinanet/issues/941 https://github.com/fizyr/keras-retinanet/issues/957 I got the same issue wile detection. Is it related to anchor params? you...
I have solved this issue. We need to add '--config config.ini' into the command while converting the trained model . Refer the 53 line (parser.add_argument('--config', help='Path to a configuration parameters...
Installed by using 'pip3 install telethon' , but still getting the error