-
**Describe the bug**
When I was trying something like exponentially weighted moving average, I saw the gradients may be incorrect.
**To Reproduce**
```py
ti.init(arch=ti.cuda)
row_num, co…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Feature Description
A Multilayer Perceptron (MLP) is a class of feedforward artificial neural networks (ANN) t…
-
# Error Report
07/25 17:33:21 - mmengine - INF0 - Iter(test) [1550/4020] eta: 1:57:23 time: 3.3139 data_time: 2.4773 memory: 5238
[35/1829]
[E ProcessGroupNCCL. cpp:587] [Rank 1] Watchdog caught co…
-
I tried to train the swin large model on coco, with swin backbone pretrained initilaization. The initial loss starts around 2e4. The loss print looks like this:
```
iter: 19 total_loss: 2.163e+04 …
-
Hi,
Thank you very much for sharing the code. I noticed that you use the original cora dataset, rather than the processed one in GCN and GAT. I was also thinking of using the original one, but I f…
-
Version 3.0.1.dev5 seems to have broken one of my earlier LyCORIS configurations:
```json
{
"algo": "lokr",
"multiplier": 1.0,
"linear_dim": 10000,
"linear_alpha": 1,
"fac…
-
something to support OpenPilot support :)
"dump"
…
-
Hi, thank so much for your work.
I am testing the model on multiple configs. While using `step()` method to get both the output and the states, I observed that models with sLSTM layer does not have…
-
### ❓ Question
I want to modify the network structure for RecurrentPPO, but when I run the original network, I get the following error
error:
self.features_extractor = features_extractor_class(se…
-
Hello David.
I'm having doubts about the implementation of transfer functions in Neural2D, specifically - their derivatives. So far I've only analyzed and used the TanH and Logistic TFs, but let's …