Open chillaxkrish opened 3 years ago
dreamyuchen
commented
3 years ago Could please tell me how to solve this runtime error ?
Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
chillaxkrish
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!!
But I got another issue
File "/usr/local/lib/python3.7/dist-packages/torch/tensor.py", line 621, in array return self.numpy() TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
Can you please help me in this ?
dreamyuchen
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!!
But I got another issue
File "/usr/local/lib/python3.7/dist-packages/torch/tensor.py", line 621, in array return self.numpy() TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
Can you please help me in this ?
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
chillaxkrish
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!! But I got another issue
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
Thanks you so much !!!
Is their any way to reduce the dataset for the train.py ? . Its very slow ..
and then where we can do the test image validation in this project ? can you help me in this please !
dreamyuchen
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!! But I got another issue
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
Thanks you so much !!!
Is their any way to reduce the dataset for the train.py ? . Its very slow ..
and then where we can do the test image validation in this project ? can you help me in this please !
If you want to reduce the number of dataset for train you have to change the settings in KarpathSplit.py. If you want to do test, you should write a test function ,you can learn how to write test functions from other image caption codes which have test function.
chillaxkrish
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!! But I got another issue
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
Thanks you so much !!! Is their any way to reduce the dataset for the train.py ? . Its very slow .. and then where we can do the test image validation in this project ? can you help me in this please !
If you want to reduce the number of dataset for train you have to change the settings in KarpathSplit.py. If you want to do test, you should write a test function ,you can learn how to write test functions from other image caption codes which have test function.
Thank you so much for helping me in this project !!!
Can you share code for the test function ? or any related test function for this project image caption. I have searched before but I can't link with in this project... please
chillaxkrish
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!! But I got another issue
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
Thanks you so much !!! Is their any way to reduce the dataset for the train.py ? . Its very slow .. and then where we can do the test image validation in this project ? can you help me in this please !
If you want to reduce the number of dataset for train you have to change the settings in KarpathSplit.py. If you want to do test, you should write a test function ,you can learn how to write test functions from other image caption codes which have test function.
Thank you so much for helping me in this project !!!
Can you share code for the test function ? or any related test function for this project image caption. I have searched before but I can't link with in this project... please
Can you please tell me what input have to give in this below image . After computing METEOR score... its keep on loading .. Will you please
dreamyuchen
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!! But I got another issue
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
Thanks you so much !!! Is their any way to reduce the dataset for the train.py ? . Its very slow .. and then where we can do the test image validation in this project ? can you help me in this please !
If you want to reduce the number of dataset for train you have to change the settings in KarpathSplit.py. If you want to do test, you should write a test function ,you can learn how to write test functions from other image caption codes which have test function.
Thank you so much for helping me in this project !!! Can you share code for the test function ? or any related test function for this project image caption. I have searched before but I can't link with in this project... please
Can you please tell me what input have to give in this below image . After computing METEOR score... its keep on loading .. Will you please
sorry,I don't have the same problems running the code as you do.I guess there is some problem with your pycocoevalcap...Perhaps you can download the latest PyCocoTools and the latest PyCocoevalcap to replace the referenced parts of the code
chillaxkrish
commented
3 years ago Could please tell me how to solve this runtime error ? Tensor flow dimension
I solved this problem, you shuld add h_t = h_t.squeeze(1) before hiddens[:, time_step, :] = h_t
yeah its working thank you !!!! But I got another issue
You should change the code loss.data to loss.item() and you'll be able to run the code smoothly
Thanks you so much !!! Is their any way to reduce the dataset for the train.py ? . Its very slow .. and then where we can do the test image validation in this project ? can you help me in this please !
If you want to reduce the number of dataset for train you have to change the settings in KarpathSplit.py. If you want to do test, you should write a test function ,you can learn how to write test functions from other image caption codes which have test function.
Thank you so much for helping me in this project !!! Can you share code for the test function ? or any related test function for this project image caption. I have searched before but I can't link with in this project... please
Can you please tell me what input have to give in this below image . After computing METEOR score... its keep on loading .. Will you please
sorry,I don't have the same problems running the code as you do.I guess there is some problem with your pycocoevalcap...Perhaps you can download the latest PyCocoTools and the latest PyCocoevalcap to replace the referenced parts of the code
Yeah ok ..
Now am facing new error . Can you please help me ?
Could please tell me how to solve this runtime error ?
Tensor flow dimension