simplew2011

### Describe the bug ``` xorbits.init(init_local=True, n_cpu=4) # do something code xorbits.shutdown() ``` ``` 2024-02-23 12:11:33,782 xorbits._mars.services.cluster.uploader 24675 ERROR Failed to upload node info Traceback (most recent call last): File...

BUG: FileNotFoundError: [Errno 2] No such file or directory

2

### Describe the bug A clear and concise description of what the bug is.  ### To Reproduce To help us to reproduce this bug, please provide information below: ```bash...

FEAT: how xorbits datastes export to json file

5

### Is your feature request related to a problem? Please describe - how xorbits datastes export to json file - in current, only support to arrow format file ### Describe...

使用场景

1

输入需要有HQ参考图像，还是不符合通用场景，大多数人脸都无法获取到参考HQ图像

在批量predict时显存上涨，最后OOM问题的解决思路

1

原因：keras.backend.get_value函数会添加新的节点到tf的图中；在每次循环图片预测时导致了tf的图越来越大，内存泄露。 参照： https://blog.csdn.net/mingshili/article/details/81941677 修改predict.py里的CTC解码部分，可解决。 每次重置session会打印tensorflow信息，比较难看，通过在predict.py开头添加下面关闭控制台输出的tensorflow信息： import os os.environ['TF_CPP_MIN_LOG_LEVEL']='2'

linux版本

1

是否有linux版本开发计划呢

Can you provide an example of large-scale text deduplication, such as the following example

1

- https://github.com/ChenghaoMou/text-dedup/blob/main/text_dedup/minhash_spark.py - https://github.com/phdinds-aim/alis/blob/68c7f56a08fa5cfe10638ea45292914620c9f5cf/notebooks/lsh-for-minhash/05_demo_minhash_lsh.ipynb - https://github.com/NVIDIA/NeMo-Curator/blob/main/nemo_curator/scripts/fuzzy_deduplication/README.md - https://xorbits.io/blogs/text-deduplicate

tarin error in ema

3

``` (python3.8) (base) wzp@vastai-NF5468M6:~/code/model_check/GAN/stylegan2-ada-lightning$ python trainer/train_stylegan.py wandb_main=True trainer/train_stylegan.py:226: UserWarning: The version_base parameter is not specified. Please specify a compatability version level, or None. Will assume defaults for version 1.1 @hydra.main(config_path='../config',...

Face Restoration

1

Can StyleGANEX be used for Face Restoration tasks without zooming in，like https://github.com/yangxy/GPEN