-
Allow the user to share elements like track, album, playlist and others. A comprehensive list of things to be shared needs to be set up first.
- [x] generate right share link (https://github.com/bc…
-
**Describe the bug**
**ttl.tensor** **bert** operations break on wormhole mashines.
1. bert_large_fused_qkv_matmul
2. bert_large_pre_softmax_bmm
3. bert_large_post_softmax_bmm
4. bert_large…
-
Hi Paul,
Currently brms requires that all families have a mu parameter, which is then interpreted as the parameter predicted by the formula with the response variable. For custom families, this the…
-
**Describe the bug**
When using a `bmmformula` that includes grouped random effects, such as the formula we use in the second example of our Tutorial manuscript:
```
LS_formula
-
### 🐛 Describe the bug
TestMathBitsXPU , totally 200 cases got RuntimeError: Double and complex datatype matmul is not supported in oneDNN
ONEDNN_VERBOSE=2 PYTORCH_ENABLE_XPU_FALLBACK=1 PYTORCH_TE…
-
### This issue is to have a centralized place to list and track work on adding support to new ops for the MPS backend.
[**PyTorch MPS Ops Project**](https://github.com/users/kulinseth/projects/1/vi…
-
When loading just Broker_MicroMenu, or BMM with a display like ChocolateBar, plus BugGrabber and BugSack, the below lua errors appear on load. The root cause appears to be that line 14 references Ace…
-
This command often fails on A100 and H100:
```
$ bin/nvfuser_bench --benchmark_filter='NvFuserScheduler_Matmul_Manual/nvfuser_splitk_NT/M:1024/N:2048/K:4096/warps:8/stages:3/splitk_factor:5/smem_ep…
-
Reporting potential quirk Mechfile errors (discovered as I parsed through the mech to sort my own mech list out).
### Environment
These will be from MML 0.49.16 (when I made my personal fork); you…
-
### 🐛 Describe the bug
#RuntimeError: Double and complex datatype matmul is not supported in oneDNN
"test_fn_fwgrad_bwgrad___rmatmul___xpu_complex128",
"test_fn_fwgrad_bwgrad___rmatmul___x…