-
When I try to resume a checkpoint from last training, it raises an error:
```
[rank12]: Traceback (most recent call last):
[rank12]: File "/cpfs/29f69eb5e2e60f26/user/GPT/pretrain/mm_intern/duyif…
-
### What new functionality do you need?
Using the standard `cp` can result in downstream programs using incomplete copies, particularly for large files. `prod_util` provides a utility, `cpfs`, that…
-
**Is your feature request related to a problem? Please describe.**
On a large deployment of CP4D/watsonx the underlying CPFS components need to be scaled up to avoid performance issues and instabilit…
-
When mv is used on WCOSS2 to move data from DATA to COM, a downstream job that depends on that data may on rare occasion check a file that has not fully moved to COM yet, which can trigger a process i…
-
012.849.332-10
036.068.160-30
-
Podemos criar vários cadastros no sistema com o mesmo CPF, realizando o teste eu consegui criar um atendido e cadastrar um parente dele com o mesmo CPF. Imagino que no resto do sistema aconteça o mesm…
-
## Description
Provide a clear and concise description of the requested feature/capability.
From NCO SPA: Improve the HAFS scripts to use cpreq and cpfs production utilities at the next upgrade.
…
-
Traceback (most recent call last):
File "/cpfs/user/mingcanma/workspace/code/openseg/APE-main/demo/demo_lazy.py", line 135, in
demo = VisualizationDemo(cfg, args=args)
File "/cpfs/user/min…
-
- Contexto:
No inciso II, escolhendo a categoria CPF sem Espaço Físico, adicionar CPFs com formatos e valores errados.
- Erro:
Os CPFs são salvos, mesmo com formatação errada. Ex: 999999999999999…
-
Hi YiLin,
When I tried to run fine-tuning code on VQA(4 gpus):
```
python run.py with data_root=${data_dir} num_gpus=4 num_nodes=1 task_finetune_vqa_square_randaug_base_image384_ufo \
exp_name…