-
Notifications
You must be signed in to change notification settings - Fork 125
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
training error information #5
Comments
Is it necessary to run install step 2 and 3? I don't install them. |
Validation is not computed for epoch 0, as the network has not completed an epoch, so it is set to infinity until an epoch has been completed, see the initialised param 'val_error' on line 153: epoch_par = {'epoch_size': len(args.train_s_list), 'epoch_comp': 0, As with your error, can you tell me what line of the script you are getting the error? |
Thank you! 2019-07-16 14:28:08.124081: W tensorflow/core/framework/op_kernel.cc:1192] Invalid argument: TensorArray map/TensorArray_4_7603: Could not read from TensorArray index 0 because it has not yet been written to. During handling of the above exception, another exception occurred: Traceback (most recent call last): Caused by op 'map/while/TensorArrayReadV3_4', defined at: InvalidArgumentError (see above for traceback): TensorArray map/TensorArray_4_7603: Could not read from TensorArray index 0 because it has not yet been written to. |
can you please check your args.val_s, args.val_d, args.val_s_len, args.val_d_len, and args.val_snr numpy arrays? e.g. print(args.val_s.shape) to check if they are the right shape. |
@anicolson , Thank you for your help! I get it. Snr information don't be included in my val data's file name. |
Only for the validation set. It's just so the validation set is always the same. Please read the following and can you tell me if you understand it: For the validation set only: Identical filenames for the clean speech and noise must be placed in 'val_clean_speech' As an example: This must be done, as the clean speech and noise used to create the noisy speech Thank you |
like this:
E1: 98% (train err 160.54 ) E0 val err: inf, 3a, GPU:1.
why is E0 err inf?
And then some error information is dumped out:
2019-07-15 11:40:25.994360: W tensorflow/core/framework/op_kernel.cc:1318] OP_REQUIRES failed at tensor_array_ops.cc:497 : Invalid argument: TensorArray map/TensorArray_4_7605: Could not read from TensorArray index 0. Furthermore, the element shape is not fully defined: . It is possible you are working with a resizeable TensorArray and stop_gradients is not allowing the gradients to be written. If you set the full element_shape property on the forward TensorArray, the proper all-zeros tensor will be returned instead of incurring this error.
......
The text was updated successfully, but these errors were encountered: