a312863063 / seeprettyface-generator-wanghong Goto Github PK

W tensorflow/core/common_runtime/bfc_allocator.cc:211] Allocator (GPU_0_bfc) ran out of memory trying to allocate 2.14GiB. The caller indicates that this is not a failure, but may mean that there could be performance gains if more memory were available.

Hey man, really love your result and I was trying to run this repository but I kept getting this out of memory... any idea how I go about fixing this?

I build is:

• Windows 10
• GTX 970 and 1070 (separate PC respectively)
• Python 3.6.9
• Tensorflow-gpu 1.14
• VS2017&2019 Community
• CUDA 10.0 + cuDNN 7.6.5

For some reason, my PC with the 1070 Graphic Card always freezes instead of showing me the message above. At least my GTX 970 PC show me the message above.

请问该怎么与generator_wanghong_hd.pkl一起使用生成码 dlatent?

如果使用 generator_wanghong_hd.pkl，它会绘制错误的图像

# http://www.seeprettyface.com/generate_code/Fanbingbing.npy
dlatent= np.load('Fanbingbing.npy')
model_path = 'generator_wanghong_hd.pkl'

synthesis_kwargs = dict(output_transform=dict(func=tflib.convert_images_to_uint8, nchw_to_nhwc=True), minibatch_size=1)
with dnnlib.util.open_url(model_path) as f:
    _G, _D, Gs = pickle.load(f)

img = Gs.components.synthesis.run(dlatent.reshape((-1, 18, 512)), randomize_noise=False, **synthesis_kwargs)[0]
plt.imshow(img)
plt.axis('off')
plt.show()

只有使用 karras2019stylegan-ffhq-1024x1024.pkl 时才会绘制正确的图像

# http://www.seeprettyface.com/generate_code/Fanbingbing.npy
dlatent= np.load('Fanbingbing.npy')
model_path = 'karras2019stylegan-ffhq-1024x1024.pkl'

synthesis_kwargs = dict(output_transform=dict(func=tflib.convert_images_to_uint8, nchw_to_nhwc=True), minibatch_size=1)
with dnnlib.util.open_url(model_path) as f:
    _G, _D, Gs = pickle.load(f)

img = Gs.components.synthesis.run(dlatent.reshape((-1, 18, 512)), randomize_noise=False, **synthesis_kwargs)[0]
plt.imshow(img)
plt.axis('off')
plt.show()

Traceback (most recent call last):
File "generate_wanghong.py", line 10, in
import PIL.Image
ModuleNotFoundError: No module named 'PIL'

How can I Deal with it？

我的qq 994562751

作者你好，我在做视频的人脸性别转换相关的工作，我没有用stylegan，而是用cyclegan，基本就是把视频中的每一帧人脸检测出来，然后通过cyclegan生成异性的图片，但是我在替换的时候遇到了下面两个问题

1. 将生成人脸替换到原视频后，有很明显的边框
2. 替换后的人脸，经常会闪烁
   但我看作者你生成的视频很自然，所以想了解下你有没有遇到我说的这两个问题，或者是怎么解决的

你好~ 运行genrate_wanghong.py的时候报了这个错：
AttributeError: module 'tensorflow' has no attribute 'Dimension'
我使用的是tensorflow-gpu 2.0.0 ,不知道应该怎么改，大神可以教教我吗~百度了半天也查不到，谢谢了

CPU上可否生成？

https://www.youtube.com/watch?v=Y2r91i0NXDI

anaconda搭建python3.6，TensorFlow==1.10.0，numpy==1.14.3但是运行后报错
InvalidArgumentError (see above for traceback): Cannot assign a device for operation 'Gs/_Run/Gs/latents_in': Operation was explicitly assigned to /device:GPU:0 but available devices are [ /job:localhost/replica:0/task:0/device:CPU:0 ]. Make sure the device specification refers to a valid device.
[[Node: Gs/_Run/Gs/latents_in = IdentityT=DT_FLOAT, _device="/device:GPU:0"]]
解决下代码问题...

感谢作者~能不能将http://www.seeprettyface.com网站中的数据集打个包再放网盘，现在数量太大没法一次性转存或下载，十分感谢