Error in Train.py: I've been trying to figure it out for an hour now

[milkmilkMilktea]

So everything runs fine until the line:
batch_loss, batch_acc = model.train_on_batch([title_batch, past_batch], [pred_batch])

I figured it was a syntax error or something so I did some research. Although I've didn't see it used this way, there was nothing wrong with the syntax. So I updated all of my Libraries and tried it again. Still the same error. No matter what I did, I couldn't get it to work. If anyone cares, or is even out there (this is not a maintained repository so maybe I'm alone), your help would be immensely appreciated.

Here's the full error log. pls help me

Traceback (most recent call last):
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1322, in _do_call
return fn(*args)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1305, in _run_fn
self._extend_graph()
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1340, in _extend_graph
tf_session.ExtendSession(self._session)
tensorflow.python.framework.errors_impl.InvalidArgumentError: No OpKernel was registered to support Op 'CudnnRNN' with these attrs. Registered devices: [CPU], Registered kernels:

[[Node: cu_dnngru_1/CudnnRNN = CudnnRNN[T=DT_FLOAT, direction="unidirectional", dropout=0, input_mode="linear_input", is_training=true, rnn_mode="gru", seed=87654321, seed2=0](cu_dnngru_1/transpose, cu_dnngru_1/ExpandDims_1, cu_dnngru_1/Const_1, cu_dnngru_1/concat)]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "C:/Users/(Censored)/PycharmProjects/YouTubeCommenter/Train.py", line 170, in
batch_loss, batch_acc = model.train_on_batch([title_batch, past_batch], [pred_batch])
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 1449, in train_on_batch
outputs = self.train_function(ins)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\backend\tensorflow_backend.py", line 2947, in call
if hasattr(get_session(), '_make_callable_from_options'):
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\backend\tensorflow_backend.py", line 204, in get_session
[tf.is_variable_initialized(v) for v in candidate_vars])
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 900, in run
run_metadata_ptr)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1135, in _run
feed_dict_tensor, options, run_metadata)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1316, in _do_run
run_metadata)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1335, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: No OpKernel was registered to support Op 'CudnnRNN' with these attrs. Registered devices: [CPU], Registered kernels:

[[Node: cu_dnngru_1/CudnnRNN = CudnnRNN[T=DT_FLOAT, direction="unidirectional", dropout=0, input_mode="linear_input", is_training=true, rnn_mode="gru", seed=87654321, seed2=0](cu_dnngru_1/transpose, cu_dnngru_1/ExpandDims_1, cu_dnngru_1/Const_1, cu_dnngru_1/concat)]]

Caused by op 'cu_dnngru_1/CudnnRNN', defined at:
File "C:/Users/(Censored)/PycharmProjects/YouTubeCommenter/Train.py", line 94, in
x = CuDNNGRU(200, return_sequences=USE_OUT_SEQ)(x)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\layers\recurrent.py", line 533, in call
return super(RNN, self).call(inputs, **kwargs)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\base_layer.py", line 450, in call
output = self.call(inputs, **kwargs)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\layers\cudnn_recurrent.py", line 90, in call
output, states = self._process_batch(inputs, initial_state)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\layers\cudnn_recurrent.py", line 297, in _process_batch
is_training=True)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\contrib\cudnn_rnn\python\ops\cudnn_rnn_ops.py", line 1621, in call
seed=self._seed)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\contrib\cudnn_rnn\python\ops\cudnn_rnn_ops.py", line 1010, in _cudnn_rnn_no_input_c
direction, dropout, seed, name)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\contrib\cudnn_rnn\python\ops\cudnn_rnn_ops.py", line 922, in _cudnn_rnn
outputs, output_h, output_c, _ = gen_cudnn_rnn_ops.cudnn_rnn(**args)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\gen_cudnn_rnn_ops.py", line 143, in cudnn_rnn
is_training=is_training, name=name)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\ops.py", line 3414, in create_op
op_def=op_def)
File "C:\Users(Censored)\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\ops.py", line 1740, in init
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

InvalidArgumentError (see above for traceback): No OpKernel was registered to support Op 'CudnnRNN' with these attrs. Registered devices: [CPU], Registered kernels:

[[Node: cu_dnngru_1/CudnnRNN = CudnnRNN[T=DT_FLOAT, direction="unidirectional", dropout=0, input_mode="linear_input", is_training=true, rnn_mode="gru", seed=87654321, seed2=0](cu_dnngru_1/transpose, cu_dnngru_1/ExpandDims_1, cu_dnngru_1/Const_1, cu_dnngru_1/concat)]]

[HackerPoet]

This network can only be trained on a GPU because it uses CuDNNGRU, the error sounds like you're trying to run it on a CPU. You may be able to fix the problem by replacing that layer with a regular GRU.

[milkmilkMilktea]

Thank you so so much @HackerPoet how would I run it on my GPU? In the past I've just used Google Colab, but there must be another way?

[milkmilkMilktea]

Okay, I did a ton of research and found a tutorial. I'll make sure to close the issue if it works

[milkmilkMilktea]

woops I meant to link something, guess it didn't work
https://www.codingforentrepreneurs.com/blog/install-tensorflow-gpu-windows-cuda-cudnn/

[HackerPoet]

Installing CUDA and cuDNN is best because it will train significantly faster. But if you can't get it working, just replace the instance of 'CuDNNGRU' with 'GRU'.

[milkmilkMilktea]

After installing 5 different versions of CUDA and cuDNN, I just gave up and used GRU and it worked great! By chance could you tell me what versions you use for future reference?

[Ajeet-Yadav]

Make sure you have Nvidia graphics card as CUDA & cuDNN are Nvidia's library.