-
Per in-person discussion, we should move cell graph creation from the `cg-gnn` pip package into SPT.
This was motivated by two things:
* Clarifying the separation of concerns between SPT and `cg-g…
-
I'm not sure this is the correct place for this, and I'm even less confident that it's my place to say any of it, but having spent a good deal of time playing with masked tensors I figured I'd share m…
jackd updated
6 months ago
-
### 🐛 Describe the bug
Using MPS for BERT inference appears to produce about a 2x slowdown compared to the CPU. Here is code to reproduce the issue:
```python
# MPS Version
from transformers i…
-
微博内容精选
-
Hi, great works! I have questions about datasets mentioned in your paper and models which you use to compare with PDFormer.
I noticed the max number of nodes of datasets is 1024(T-Drive), which is…
-
We currently construct cell graphs using DGL, but it's used relatively rarely compared to something like saving a feature and adjacency matrix in pytorch or numpy, even in the GNN space. It's simple e…
-
![image](https://github.com/rhythmcao/astormer/assets/80022154/2db658b4-523e-404e-bc54-7e17b4bbf31d)
Hello~ rhythm~ I want to ask why word level schema linking is used instead of using an subword l…
-
This issue is meant to summarize the current status and likely future direction of the NumPy array protocols, and their relevance to the array API standard.
**What are these array protocols?**
I…
-
-
I created a conda environment, activated it and installed pip in it; I then followed steps 1 and 2 in https://github.com/deepsphere/deepsphere-cosmo-tf2, but it returns
"
note: This error origi…