-
Hi RIM dev team,
Code fails when device = 'cuda'
It can be easily solved adding an extra parameter "device" to all "Group" classes.
Thank you for the great RIM implementation,
Ildefons
-
Good understanding of deep learning architectures like Multi-Layer Perceptron, Recurrent Neural Networks (RNNs), Long Short Term Memory models (LSTMs), Gated Recurrent Units (GRUs), and Convolutional …
-
This was recently brought to my attention. I am glad that you are able to get better performance than standard fsspec.
First a couple of notes
- fsspec provides multiple possible (memory) caching…
-
# Qwen1.5-MoE Support
With the increasing attention on mixture-of-experts (MoE) models, especially following the advancements heralded by Mixtral, I propose considering the integration of the Qwen1.5…
-
When we are developing some library, we need to include logisim-evolution with gradle.
If you want to publish this to maven repositry, `maven-publish` plugin should added to `build.gradle` and set …
-
Hello,
May I know if the future layer prediction (FLP) exclusively forecasts the future control points on a layer-wise basis, without considering interactions or dependencies with other layers.
-
### Model description
This model is is a Self-supervised Vision Transformer that uses patch reconstruction as the spectrogram task. It extends MAE (which is already on HuggingFace) for audio. This mo…
-
Currently, we collect as many errors as possible via the CompoundError mechanism. However, these "errors" are not typical Rust errors, but rather diagnostic messages intended for user reporting via th…
-
### What happened?
Every time I access my linux dashboard, it's telling me to migrate from Angular, which I do. Over and over and over and over again
### What did you expect to happen?
I expe…
-
## 🐛 Bug
The following code snippet from multihead attention module is using tensor.equal method to compare query, key and value to determine if the attention module is being used as self-attention…