Help!! I have a problem! As I introduced on the title, I have a problem with map memory and openCL.
I am Korean, and I am sorry for my bad English wording.
My Computer's GPU has 2GB, and the RAM has 8GB and CPU has intel-i3 3.7GHz, and 64bit Windows.
I used the original Trainer and my batch size is 48.
And I also have question. I want to change video of Alpha with the face of Beta. Then how to set the Alpha, Beta with input A, B?
Following is the code that the program showed to me.
Loading...
Setting Faceswap backend to AMD
04/21/2020 21:22:42 INFO Log level set to: INFO
04/21/2020 21:22:43 INFO Setting up for PlaidML
04/21/2020 21:22:43 INFO Setting GPU to largest available supported device. If you want to override this selection, run plaidml-setup
from the command line.
04/21/2020 21:22:43 INFO Using GPU: ['opencl_nvidia_geforce_gtx_750_ti.0', 'opencl_nvidia_geforce_gtx_750_ti.0']
04/21/2020 21:22:43 INFO Successfully set up for PlaidML
Using plaidml.keras.backend backend.
04/21/2020 21:22:46 INFO Model A Directory: C:\Users\work\faceswap\faces\dahyeon
04/21/2020 21:22:46 INFO Model B Directory: C:\Users\work\faceswap\faces\yeong
04/21/2020 21:22:46 INFO Training data directory: C:\Users\work\faceswap\yeongmodel
04/21/2020 21:22:46 INFO ===================================================
04/21/2020 21:22:46 INFO Starting
04/21/2020 21:22:46 INFO Press 'Stop' to save and quit
04/21/2020 21:22:46 INFO ===================================================
04/21/2020 21:22:47 INFO Loading data, this may take a while...
04/21/2020 21:22:47 INFO Loading Model from Original plugin...
04/21/2020 21:22:47 INFO No existing state file found. Generating.
04/21/2020 21:22:47 INFO Opening device "opencl_nvidia_geforce_gtx_750_ti.0"
04/21/2020 21:22:48 INFO Creating new 'original' model in folder: 'C:\Users\work\faceswap\yeongmodel'
04/21/2020 21:22:48 INFO Loading Trainer from Original plugin...
04/21/2020 21:22:48 INFO Enabled TensorBoard Logging
Unable to map memory: CL_MEM_OBJECT_ALLOCATION_FAILURE
04/21/2020 21:22:50 ERROR Unable to map memory: CL_MEM_OBJECT_ALLOCATION_FAILURE
04/21/2020 21:22:50 CRITICAL Error caught! Exiting...
04/21/2020 21:22:50 ERROR Caught exception in thread: 'training_0'
04/21/2020 21:22:54 ERROR Got Exception on main handler:
Traceback (most recent call last):
File "C:\Users\work\faceswap\lib\cli.py", line 128, in execute_script
process.process()
File "C:\Users\work\faceswap\scripts\train.py", line 161, in process
self.end_thread(thread, err)
File "C:\Users\work\faceswap\scripts\train.py", line 201, in end_thread
thread.join()
File "C:\Users\work\faceswap\lib\multithreading.py", line 121, in join
raise thread.err[1].with_traceback(thread.err[2])
File "C:\Users\work\faceswap\lib\multithreading.py", line 37, in run
self.target(*self.args, **self.kwargs)
File "C:\Users\work\faceswap\scripts\train.py", line 226, in _training
raise err
File "C:\Users\work\faceswap\scripts\train.py", line 216, in training
self.run_training_cycle(model, trainer)
File "C:\Users\work\faceswap\scripts\train.py", line 305, in run_training_cycle
trainer.train_one_step(viewer, timelapse)
File "C:\Users\work\faceswap\plugins\train\trainer\_base.py", line 316, in train_one_step
raise err
File "C:\Users\work\faceswap\plugins\train\trainer\_base.py", line 283, in train_one_step
loss[side] = batcher.train_one_batch()
File "C:\Users\work\faceswap\plugins\train\trainer\_base.py", line 424, in train_one_batch
loss = self.model.predictors[self.side].train_on_batch(model_inputs, model_targets)
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\keras\engine\training.py", line 1216, in train_on_batch
self.make_train_function()
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\keras\engine\training.py", line 509, in make_train_function
loss=self.total_loss)
File "C:\Users\work\faceswap\lib\model\optimizers.py", line 51, in get_updates
ms, vs, vhats = self.update_1(params)
File "C:\Users\work\faceswap\lib\model\optimizers.py", line 78, in update_1
ms = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
File "C:\Users\work\faceswap\lib\model\optimizers.py", line 78, in <listcomp>
ms = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\plaidml\keras\backend.py", line 1743, in zeros
return constant(0.0, shape=shape, dtype=dtype, name=prepend_name_scope(name, 'zeros'))
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\plaidml\keras\backend.py", line 482, in constant
return variable(np_value, dtype=dtype, name=prepend_name_scope(name, 'constant'))
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\plaidml\keras\backend.py", line 1735, in variable
with tensor.mmap_discard(ctx) as view:
File "C:\Users\work\anaconda3\envs\faceswap\lib\contextlib.py", line 112, in enter
return next(self.gen)
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\plaidml\__init.py", line 1252, in mmap_discard
mapping = _lib().plaidml_map_buffer_discard(ctx, self.buffer)
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\plaidml\__init.py", line 777, in _check_err
self.raise_last_status()
File "C:\Users\work\anaconda3\envs\faceswap\lib\site-packages\plaidml\library.py", line 131, in raise_last_status
raise self.last_status()
plaidml.exceptions.Unknown: Unable to map memory: CL_MEM_OBJECT_ALLOCATION_FAILURE
04/21/2020 21:22:54 CRITICAL An unexpected crash has occurred. Crash report written to 'C:\Users\work\faceswap\crash_report.2020.04.21.212254189152.log'. You MUST provide this file if seeking assistance. Please verify you are running the latest version of faceswap before reporting
Process exited.