-
1. after running `pip install -e .[tests]` in a new conda environment, `pytest tests` yields:`
```
ERROR: usage: pytest [options] [file_or_dir] [file_or_dir] [...]
pytest: error: unrecognized argum…
-
## description
- datasets: 1.1 GB Chinese corpus, about 2 million lines
- devices
- CPU=8 GPU=1 Memory=320G Node=1 Type=A100-SXM-80GB
## question
The training process was always killed …
-
where can i find the checkpoint/Train_10_images_t1_attention_transformer_COCO_10layers_NoClass/latest_checkpoint.pth
-
hello, flash-attention is really a great work!
However, I am trying to reproduce the TFLOPS data from the flash2 paper on an A100 40GB machine(The computing power of the A100 40GB should be the sam…
-
I'm trying to set the zoom level when there's only 1 marker. i'm not sure if it's set somewhere in the code or if it's default behaviour; but when there's only 1 marker, it zooms in to the max zoom le…
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
PyTorch version: 2.4.0+cu121
Is debug build: False
CUDA used to build PyTor…
-
### System Info
* CPU Architecture - x86_64
* CPU/Host memory size - 330gb from `/proc/meminfo`
* GPU name - NVIDIA H100 80GB HBM3
* GPU memory size - 81559MiB
* TensorRT-LLM branch - v0.7.1
*…
-
https://af.wikipedia.org/wiki/ISO_3166-1
-
下面是我在A100-sxm-80G上的测试结果：
vllm
`python -m vllm.entrypoints.api_server --model /code/llama-65b-hf --swap-space 16 --disable-log-requests --tensor-parallel-size 8`
`python benchmarks/benchmark_serving…
-
Just wanted to drop a note to say THANK YOU for making this available. It's just what I was looking for and reading through the history is rather sad to hear. It deserved so much more success. Seei…