-
when load the optimizer.pt display the key is different
KeyError: 'base_model.model.model.layers.0.self_attn.q_proj.lora_A.default.weight'
the items in optimizer.pt state is 0~255.
-
### Describe the bug?
It seems the Asset Optimization Wizard only affects objects who have their materials stored in the asset folders. Objects whos materials are stored on the object itself are un…
-
- [ ] Emit all files only at the end. Evaluate performance impact
- [ ] Avoid emitting files which are never actually used (previous point)
- [ ] Apply optimizations during image conversion in sharp…
-
BPO | [40151](https://bugs.python.org/issue40151)
--- | :---
Nosy | @pfmoore, @tjguk, @zware, @zooba, @ariccio
PRs | python/cpython#19298
*Note: these values reflect the state of the issue at the tim…
-
| | |
| --- | --- |
| Bugzilla Link | [24536](https://llvm.org/bz24536) |
| Version | trunk |
| OS | FreeBSD |
| Attachments | [Pulse Table .cc](https://user-images.githubusercontent.com/609449…
-
With all PRs merged, tree-sitter will generate a very large parser.c file (about 83M). And it takes about 1 min to load the parser by tree-sitter python binding.
Unfortunately I don't have an idea…
pplam updated
2 years ago
-
配置:
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 torchrun --rdzv_backend=c10d --rdzv_endpoint=localhost:29402 --nnodes=1 --nproc_per_node=8 train.py
config.tp_size = 1
config.dp_size = 1 # 8 无所谓
config.pp…
-
```
Current implementation of u-s has several limitation on providing several
features to control how JIT should works.
1. A more flexible hotness threshold value. The JIT compiler will be
executed …
-
I need to run the Triton Server using an ONNX model that generates a TensorRT engine on-the-fly. I'm aware that I could use the trtexec utility to generate the TensorRT engine, but I have multiple ty…
-
As part of [adding wide gamut color support to Flutter](https://github.com/flutter/flutter/issues/127855) I'm looking to migrate [Color](https://api.flutter.dev/flutter/dart-ui/Color-class.html) to ho…