However, when I try to run the code you provided, the actual model training doesn't seem to work correctly. For example, the performance of the model displayed in TensorBoard is not satisfactory, and the model output during training is not displayed correctly. Additionally, when I train the model, the output is represented as a black visual screen.
I only changed the data path in your code. Why is this happening?
Unlike the previous issue, the code seems to run fine, but the training doesn't appear to be working correctly.
What do you think?
Lastly, if the input is 192x640, is it correct for the output to be 96x320?
Dear,
The code provided is excellent.
However, when I try to run the code you provided, the actual model training doesn't seem to work correctly. For example, the performance of the model displayed in TensorBoard is not satisfactory, and the model output during training is not displayed correctly. Additionally, when I train the model, the output is represented as a black visual screen.
I only changed the data path in your code. Why is this happening?
Unlike the previous issue, the code seems to run fine, but the training doesn't appear to be working correctly.
What do you think?
Lastly, if the input is 192x640, is it correct for the output to be 96x320?
Sincerely,