[Example] Add RegAE example #660
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
HydrogenSulfate
requested changes
Nov 29, 2023
docs/zh/examples/RegAE.md
Outdated
There was a problem hiding this comment.
markdown文档用 vscode markdownlint 插件格式化一下
examples/RegAE/RegAE.py
Outdated
Comment on lines
27
to
33
| criterion = nn.MSELoss() | ||
| kl_loss = KLLoss() | ||
|
|
||
|
|
||
| def loss_expr(output_dict, label_dict, weight_dict=None): | ||
|
|
||
| return kl_loss(output_dict) + criterion(output_dict["p_hat"], label_dict["p_hat"]) |
There was a problem hiding this comment.
- 这里用F.mse_loss和KLLoss里的函数吧，简化下代码
loss_expr函数定义放到sup_constraint = ...的上方，不用作为全局函数- import不要直接导入某函数或者类，通过导入上一级的模块再访问
examples/RegAE/RegAE.py
Outdated
| # set random seed for reproducibility | ||
| ppsci.utils.misc.set_random_seed(cfg.seed) | ||
| # initialize logger | ||
| logger.init_logger("ppsci", osp.join(cfg.output_dir, "train.log"), "info") |
examples/RegAE/dataloader.py
Outdated
| @@ -0,0 +1,142 @@ | |||
| """ | |||
| 输入数据类型 10^5 * 100 * 100 | |||
There was a problem hiding this comment.
数据类型为什么是一个乘法数值？不应该是float或者double吗？还是说10^5表示数值范围，后面的100*100表示样本数？感觉表述可以更改得准确一点
examples/RegAE/dataloader.py
Outdated
|
|
||
| def transform(self, data): | ||
| mean = ( | ||
| paddle.to_tensor(self.mean).type_as(data).to(data.device) |
There was a problem hiding this comment.
==> to_tensor(self.mean, dtype=data.dtype)
examples/RegAE/dataloader.py
Outdated
| self.train_data = data[: self.train_len] | ||
| self.test_data = data[self.train_len :] | ||
|
|
||
| self.scaler = ScalerStd() |
There was a problem hiding this comment.
scaler跟AMP的scaler存在歧义，改为 normalizer
ppsci/arch/__init__.py
Outdated
There was a problem hiding this comment.
API文档里加一下AutoEncoder
ppsci/arch/vae.py
Outdated
| from ppsci.arch import base | ||
|
|
||
|
|
||
| # copy from AISTUDIO |
ppsci/arch/vae.py
Outdated
Comment on lines
26
to
34
| class AutoEncoder(base.Arch): | ||
| def __init__( | ||
| self, | ||
| input_keys: Tuple[str, ...], | ||
| output_keys: Tuple[str, ...], | ||
| input_dim, | ||
| latent_dim, | ||
| hidden_dim, | ||
| ): |
There was a problem hiding this comment.
完善docstring、type hint
Comment on lines
+40
to
+45
| mu, log_sigma = output_dict["mu"], output_dict["log_sigma"] | ||
|
|
||
| base = paddle.exp(2.0 * log_sigma) + paddle.pow(mu, 2) - 1.0 - 2.0 * log_sigma | ||
| loss = 0.5 * paddle.sum(base) / mu.shape[0] | ||
|
|
||
| return loss |
There was a problem hiding this comment.
这是KL散度损失吗？但是看起来跟nn.KLDivLoss不是很像，如果不是很通用的Loss的话还是放在案例里通过FunctionalLoss使用吧，如果是比较通用的，就需要写成KLDiv的形式，即
huohuohuohuohuo123
pushed a commit
to huohuohuohuohuo123/PaddleScience
that referenced
this pull request
Aug 12, 2024
* add RegAE example * add RegAE --------- Co-authored-by: HydrogenSulfate <490868991@qq.com>
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.
PR types
New features
PR changes
Others
Describe