-
Reenable transpose propagation to clean up lots of transpose/copy/broadcast dispatches.
-
In order to most accurately carry out clustering and reproduce the position of the hit a NN model should be made. This should then be used to compare the resolutions of the simulated and generated sam…
-
torch version: 2.5.0.dev20240616+cu121
python version: python 3.8
I run the llama example with torchrun --nproc-per-node 2 pippy_llama.py. It got an Error
```
Loading checkpoint shards: 100%|███…
-
### 🚀 The feature, motivation and pitch
When creating a heterogeneous model by using `torch_geometric.nn.to_hetero_with_bases()`, using `torch.nn.parameter.Parameter` directly as a layer is unsupport…
-
According to the OpenLCA 2 manual, I should be able to right click on a process in the model graph of a product system and select "Search providers for" if I want to build provider links manually.
…
-
### Description
In TG there is a subsystem responsible for generation of domain metadata.
It was created specifically for EQL.
EQL3 metadata has the following structure:
* Unordered set of ent…
-
When initially getting the pipeline working, I struggled to resolve an issue that would crash Olinda with a Graph Execution Error where the tuner class trains the final model after identifying the bes…
-
When calling `numpyro.render_model(model, filename=my_path)` I was expecting the generated file to be saved in `my_path`. But it does not.
Reading the source code, it seems that it deliberately us…
-
I did just catch this bug: It comes from the `make_compute_graph` which causes a self loop
```python
from pymc.model_graph import ModelGraph
coords = {
"obs": range(5),
}
with pm.Model(…
-
Can I understand that the current dataset "cora" in the project only contains one graph, and how can I modify the code to utilize data from multiple graphs for training and generate overall graph embe…