Add training of first order motion #256
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
qingqing01
reviewed
Apr 14, 2021
| @@ -0,0 +1,250 @@ | |||
| import logging | |||
ppgan/datasets/firstorder_dataset.py
Outdated
| import tqdm | ||
| from imageio import imread, mimread, imwrite | ||
| from paddle.io import Dataset | ||
| from sklearn.model_selection import train_test_split |
There was a problem hiding this comment.
如果是用户训练自己的数据，需要对数据进行随机分train和test吧
ppgan/datasets/firstorder_dataset.py
Outdated
| from paddle.io import Dataset | ||
| from sklearn.model_selection import train_test_split | ||
| from skimage.transform import resize | ||
| from skimage import io, img_as_float32 |
There was a problem hiding this comment.
skimage是否必须？ 如果是，需要 https://github.com/PaddlePaddle/PaddleGAN/blob/develop/requirements.txt 增加依赖
|
|
||
| @DATASETS.register() | ||
| class FirstOrderDataset(Dataset): | ||
| def __init__(self, **cfg): |
There was a problem hiding this comment.
add comments, cfg都包含那些字段？觉得还是铺开好一些，否则不知道要输入那些参数
There was a problem hiding this comment.
输入参数不是一般不要超过6个么，这里有9个，是不是铺开不太好？
ppgan/datasets/firstorder_dataset.py
Outdated
| out['name'] = video_name | ||
| return out | ||
|
|
||
| def getSample(self, idx): |
There was a problem hiding this comment.
getSample -> get_sample
|
|
||
| class DownBlock2d(nn.Layer): | ||
| """ | ||
| Simple block for processing video (encoder). |
There was a problem hiding this comment.
add comments for input arguments
| return out | ||
|
|
||
|
|
||
| class Discriminator(nn.Layer): |
| @@ -0,0 +1,282 @@ | |||
| import paddle | |||
ppgan/models/firstorder_model.py
Outdated
| import paddle.nn as nn | ||
| import numpy as np | ||
| from skimage.draw import circle | ||
| import matplotlib.pyplot as plt |
There was a problem hiding this comment.
matplotlib还是lazy import吧，否则ppgan整个得依赖matplotlib
| return loss_values, generated | ||
|
|
||
|
|
||
| class Vgg19(nn.Layer): |
There was a problem hiding this comment.
can this reuse criterion/perceptual_loss.py?
LielinJiang
reviewed
Apr 14, 2021
| kernel_size=kernel_size) | ||
| if sn: | ||
| self.conv = spectral_norm(self.conv) | ||
| #if sn: |
LielinJiang
reviewed
Apr 14, 2021
ppgan/modules/first_order.py
Outdated
| @@ -285,6 +319,19 @@ def forward(self, input): | |||
|
|
|||
| out = F.pad(input, [self.ka, self.kb, self.ka, self.kb]) | |||
| out = F.conv2d(out, weight=self.weight, groups=self.groups) | |||
| out = F.interpolate(out, scale_factor=[self.scale, self.scale]) | |||
| out.stop_gradient = False | |||
| # The high version of pytorch has a bug that affects the convergence of this model | |||
LielinJiang
approved these changes
Apr 19, 2021
… rid of sklearn, skimage
qingqing01
approved these changes
Apr 21, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix #253