【角度加强工具】此方法实现了训练时侧脸和上下脸都优化.

1427523462

有人用过试过效果没

ajno22

没有看懂楼主的方法，所以就全部替换了字符，启动训练程序后出来了一大段的字码，最后到了平时的训练那里，然后就没有任何反应了。应该是替换字符替换的不对，烦请楼主再发个讲解，说一下替换字符，是替换哪一部分。

Headmaster1615

ajno22 发表于 2024-4-20 00:28
没有看懂楼主的方法，所以就全部替换了字符，启动训练程序后出来了一大段的字码，最后到了平时的训练那里， ...

你做的没错，就是替换全部字符。把你说的“一大段的字码”复制过来看看，应该不是我代码的问题。

ajno22

Headmaster1615 发表于 2024-4-21 00:58
你做的没错，就是替换全部字符。把你说的“一大段的字码”复制过来看看，应该不是我代码的问题。 ...

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "multiprocessing\process.py", line 258, in _bootstrap
  File "multiprocessing\process.py", line 93, in run
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_src\aligned\03026_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

Process Process-23:
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "multiprocessing\process.py", line 258, in _bootstrap
  File "multiprocessing\process.py", line 93, in run
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_src\aligned\04519_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

Process Process-24:
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "multiprocessing\process.py", line 258, in _bootstrap
  File "multiprocessing\process.py", line 93, in run
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_src\aligned\01298_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

Process Process-21:
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "multiprocessing\process.py", line 258, in _bootstrap
  File "multiprocessing\process.py", line 93, in run
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_src\aligned\00574_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

侧脸和上下都优化: 100%|######################| 128/128 [01:47<00:00,  1.19it/s]
Process Process-28:
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)
Process Process-26:

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "multiprocessing\process.py", line 258, in _bootstrap
Process Process-29:
  File "multiprocessing\process.py", line 93, in run
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
ValueError: too many values to unpack (expected 1)
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_dst\aligned\17137_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)


During handling of the above exception, another exception occurred:

Traceback (most recent call last):
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
  File "multiprocessing\process.py", line 258, in _bootstrap
ValueError: too many values to unpack (expected 1)
  File "multiprocessing\process.py", line 93, in run

During handling of the above exception, another exception occurred:

  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
  File "multiprocessing\process.py", line 258, in _bootstrap
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_dst\aligned\16131_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

  File "multiprocessing\process.py", line 93, in run
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_dst\aligned\07453_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

Process Process-27:
Traceback (most recent call last):
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "multiprocessing\process.py", line 258, in _bootstrap
  File "multiprocessing\process.py", line 93, in run
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\core\jobl
ib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 142, in batch_func
    raise Exception ("Exception occured in sample %s. Error: %s" % (sample.filen
ame, traceback.format_exc() ) )
Exception: Exception occured in sample G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\
workspace\data_dst\aligned\10276_0.jpg. Error: Traceback (most recent call last)
:
  File "G:\DeepFaceLab_NVIDIA_RTX2080Ti_MVE_AIBL\_internal\DeepFaceLab\samplelib
\SampleGeneratorFace.py", line 140, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.ou
tput_sample_types, self.debug, ct_sample=ct_sample)
ValueError: too many values to unpack (expected 1)

ajno22

Headmaster1615 发表于 2024-4-21 00:58
你做的没错，就是替换全部字符。把你说的“一大段的字码”复制过来看看，应该不是我代码的问题。 ...

特别长，上面顶到没影了，可能复制的不够全。麻烦大佬看一下，是什么原因。

qq313586234

ajno22 发表于 2024-4-21 15:49
特别长，上面顶到没影了，可能复制的不够全。麻烦大佬看一下，是什么原因。 ...

怎样？解决了吗？

pasanonic

vancheap 发表于 2024-3-25 11:49
关键部分移到MVE版本，好像没有报错，不知道有没有效果

哪几段是关键代码  我今天开始转移到ME了

vancheap

pasanonic 发表于 2024-5-5 23:48
哪几段是关键代码  我今天开始转移到ME了

其实就一段

        if uniform_yaw_distribution:  # Now it means uniform distribution for both pitch and yaw 都优化
            samples_pyr = [(idx, sample.get_pitch_yaw_roll()) for idx, sample in enumerate(samples)]
            
            grads = 128
            pitch_space = np.linspace(-1.2, 1.2, grads)  # Assuming the same range for pitch
            yaw_space = np.linspace(-1.2, 1.2, grads)  # Yaw range

            grid_samples_list = [[None]*grads for _ in range(grads)]
            for g_pitch in io.progress_bar_generator(range(grads), "侧脸和上下脸排序中"):
                pitch = pitch_space[g_pitch]
                next_pitch = pitch_space[g_pitch+1] if g_pitch < grads-1 else pitch + (pitch_space[1] - pitch_space[0])
                for g_yaw in range(grads):
                    yaw = yaw_space[g_yaw]
                    next_yaw = yaw_space[g_yaw+1] if g_yaw < grads-1 else yaw + (yaw_space[1] - yaw_space[0])
                    
                    grid_samples = []
                    for idx, pyr in samples_pyr:
                        s_pitch, s_yaw, _ = pyr
                        s_pitch, s_yaw = -s_pitch, -s_yaw  # Adjusting the signs if necessary
                        
                        if (pitch <= s_pitch < next_pitch) and (yaw <= s_yaw < next_yaw):
                            grid_samples.append(idx)
                    if len(grid_samples) > 0:
                        if grid_samples_list[g_pitch][g_yaw] is None:
                            grid_samples_list[g_pitch][g_yaw] = []
                        grid_samples_list[g_pitch][g_yaw].extend(grid_samples)
            
            # Flatten the list and filter out None values
            yaws_sample_list = [item for sublist in grid_samples_list for item in sublist if item is not None]

            index_host = mplib.Index2DHost(yaws_sample_list)
        else:
            index_host = mplib.IndexHost(self.samples_len)

pasanonic

vancheap 发表于 2024-5-6 19:33
其实就一段

感谢感谢

728187887

pasanonic 发表于 2024-5-6 20:23
感谢感谢

有效吗，我用了半个月没有分辨出区别