Training the XSeg model for Deepfacelabs fails due to memory error - training-data

Question

I'm new to deepfakes and I'm trying to do the 5XSeg) train.bat and everytime it finishes the filtering I get the following error. I use wf, and tried batch sizes from 1-8, always the same result. I have a Ryzen 5 3600, a 3080 Ti and 16 GB of RAM.
Using 26519 xseg labeled samples.
Traceback (most recent call last):
File "multiprocessing\queues.py", line 234, in _feed
File "multiprocessing\reduction.py", line 51, in dumps
MemoryError
Error:
Traceback (most recent call last):
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 1375, in _do_call
return fn(*args)
Traceback (most recent call last):
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 1360, in _run_fn
target_list, run_metadata)
File "multiprocessing\queues.py", line 234, in _feed
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 1453, in _call_tf_sessionrun
run_metadata)
File "multiprocessing\reduction.py", line 51, in dumps
tensorflow.python.framework.errors_impl.InternalError: 2 root error(s) found.
(0) Internal: Attempting to perform BLAS operation using StreamExecutor without BLAS support
[[{{node MatMul}}]]
[[concat_6/concat/_3]]
(1) Internal: Attempting to perform BLAS operation using StreamExecutor without BLAS support
[[{{node MatMul}}]]
0 successful operations.
0 derived errors ignored.
MemoryError
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\ModelBase.py", line 263, in update_sample_for_preview
self.get_history_previews()
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\ModelBase.py", line 383, in get_history_previews
return self.onGetPreview (self.sample_for_preview, for_history=True)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\Model_XSeg\Model.py", line 209, in onGetPreview
I, M, IM, = [ np.clip( nn.to_data_format(x,"NHWC", self.model_data_format), 0.0, 1.0) for x in ([image_np,mask_np] + self.view (image_np) ) ]
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\Model_XSeg\Model.py", line 141, in view
return nn.tf_sess.run ( [pred], feed_dict={self.model.input_t :input_np})
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 968, in run
run_metadata_ptr)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 1191, in _run
feed_dict_tensor, options, run_metadata)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 1369, in _do_run
run_metadata)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\client\session.py", line 1394, in _do_call
raise type(e)(node_def, op, message) # pylint: disable=no-value-for-parameter
tensorflow.python.framework.errors_impl.InternalError: 2 root error(s) found.
(0) Internal: Attempting to perform BLAS operation using StreamExecutor without BLAS support
[[node MatMul (defined at E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\layers\Dense.py:66) ]]
[[concat_6/concat/_3]]
(1) Internal: Attempting to perform BLAS operation using StreamExecutor without BLAS support
[[node MatMul (defined at E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\layers\Dense.py:66) ]]
0 successful operations.
0 derived errors ignored.
Errors may have originated from an input operation.
Input Source operations connected to node MatMul:
XSeg/dense1/weight/read (defined at E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\layers\Dense.py:47)
Reshape_60 (defined at E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\ops\__init__.py:182)
Input Source operations connected to node MatMul:
XSeg/dense1/weight/read (defined at E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\layers\Dense.py:47)
Reshape_60 (defined at E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\ops\__init__.py:182)
Original stack trace for 'MatMul':
File "threading.py", line 884, in _bootstrap
File "threading.py", line 916, in _bootstrap_inner
File "threading.py", line 864, in run
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\mainscripts\Trainer.py", line 58, in trainerThread
debug=debug)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\Model_XSeg\Model.py", line 17, in __init__
super().__init__(*args, force_model_class_name='XSeg', **kwargs)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\ModelBase.py", line 193, in __init__
self.on_initialize()
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\Model_XSeg\Model.py", line 103, in on_initialize
gpu_pred_logits_t, gpu_pred_t = self.model.flow(gpu_input_t, pretrain=self.pretrain)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\facelib\XSegNet.py", line 85, in flow
return self.model(x, pretrain=pretrain)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\models\ModelBase.py", line 117, in __call__
return self.forward(*args, **kwargs)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\models\XSeg.py", line 124, in forward
x = self.dense1(x)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\layers\LayerBase.py", line 14, in __call__
return self.forward(*args, **kwargs)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\leras\layers\Dense.py", line 66, in forward
x = tf.matmul(x, weight)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\util\dispatch.py", line 206, in wrapper
return target(*args, **kwargs)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\ops\math_ops.py", line 3655, in matmul
a, b, transpose_a=transpose_a, transpose_b=transpose_b, name=name)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\ops\gen_math_ops.py", line 5713, in mat_mul
name=name)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 750, in _apply_op_helper
attrs=attr_protos, op_def=op_def)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\framework\ops.py", line 3569, in _create_op_internal
op_def=op_def)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\python-3.6.8\lib\site-packages\tensorflow\python\framework\ops.py", line 2045, in __init__
self._traceback = tf_stack.extract_stack_for_node(self._c_op)
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\mainscripts\Trainer.py", line 58, in trainerThread
debug=debug)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\Model_XSeg\Model.py", line 17, in __init__
super().__init__(*args, force_model_class_name='XSeg', **kwargs)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\ModelBase.py", line 216, in __init__
self.update_sample_for_preview(choose_preview_history=self.choose_preview_history)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\ModelBase.py", line 265, in update_sample_for_preview
self.sample_for_preview = self.generate_next_samples()
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\models\ModelBase.py", line 461, in generate_next_samples
sample.append ( generator.generate_next() )
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\samplelib\SampleGeneratorBase.py", line 21, in generate_next
self.last_generation = next(self)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\samplelib\SampleGeneratorFace.py", line 112, in __next__
return next(generator)
File "E:\DeepFaceLab_NVIDIA_RTX3000_series\_internal\DeepFaceLab\core\joblib\SubprocessGenerator.py", line 73, in __next__
gen_data = self.cs_queue.get()
File "multiprocessing\queues.py", line 94, in get
File "multiprocessing\connection.py", line 216, in recv_bytes
File "multiprocessing\connection.py", line 318, in _recv_bytes
File "multiprocessing\connection.py", line 344, in _get_more_data
MemoryError
Reducing the batch size didn't help as well as increasing the page file. I tried to Google it but I couldn't find a solution.