Zhe Zhang

In https://github.com/microsoft/garnet/issues/152 We found that when entering only the SCAN command we would not receive the return value ### **Now I fixed the bug.** The successful screenshot is below: ...

Since I found that **sismember** command was still missing, I added it

fix issue: https://github.com/microsoft/garnet/issues/190

fix: https://github.com/microsoft/garnet/issues/277

In Redis, when LEFT or RIGHT is missed, it will return syntax error.  However, We didn't handle this situation 

图片存储路径是否能从image换到其他

2

您好，有一个问题想请教一下： 之前的模板图片都统一放到了image目录下，但是我的博客（或者全部）的图片想放到_post/static目录下，请问有啥设置办法吗？现在的问题是github仓库里面看md文件是没问题的，但是渲染出来的html页面是有问题的 

[Bug]: Incorrect Device Handling in DeviceMemoryProfiler Causes Zero Memory Usage Log

1

### Your current environment The output of `python collect_env.py` ```text INFO 02-10 20:39:49 __init__.py:190] Automatically detected platform cuda. Collecting environment information... PyTorch version: 2.5.1+cu124 Is debug build: False CUDA used...

RuntimeError: batch size must be positive in flash_attn_varlen_forward when use_remove_padding=True

9

**Environment**: • Framework: `verl` • Components: `PPO Trainer`, `FSDP Worker`, `Flash Attention` **Problem Description**: When running GRPO training with `use_remove_padding=True`, the code fails with: ``` RuntimeError: batch size must be...

``` export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 # Assume 8 GPUs export HYDRA_FULL_ERROR=1 # show more error message export NCCL_DEBUG=INFO # show more error message fairseq-hydra-train -m --config-dir examples/t5/config/pl \ --config-name base1m_rpe_mask25_10_2e-4_8gpus_bs8192 \ task.data=/path/to/data/githubcorpus/data-bin...