calmiLovesAI

Bottom-up测试过程中的图像resize过程究竟是怎样的？我发现它并不会将输入图像resize到我在config文件data_cfg中设置的那样

5

这是我在配置文件中设置的： ``` data_cfg = dict( image_size=256, base_size=128, base_sigma=2, heatmap_size=[64, 128], num_joints=channel_cfg['dataset_joints'], dataset_channel=channel_cfg['dataset_channel'], inference_channel=channel_cfg['inference_channel'], num_scales=2, scale_aware_sigma=False, ) ``` 我知道这里的base_size是用作multi scale test的，但是我在test_cfg里的设置是`scale_factor=[1]`，也就是说只有一个缩放比例，就是它本身，但是我在测试过程中打印了data_loader，修改代码如下： ``` def single_gpu_test(model, data_loader): """Test model with a single gpu....

为什么所有Bottom_up模型的config文件中的输入图片大小都是宽高相同的？

6

例如在`res50_crowdpose_512x512.py`中： ``` data_cfg = dict( image_size=512, base_size=256, base_sigma=2, heatmap_size=[128], num_joints=channel_cfg['dataset_joints'], dataset_channel=channel_cfg['dataset_channel'], inference_channel=channel_cfg['inference_channel'], num_scales=1, scale_aware_sigma=False, ) ``` 如果我想让输入图片size变为[512, 384]，应该怎么设置参数？

Cannot evaluate results on CrowdPose datatset

1

hello, Thanks for your work. I've trained HigherHRNet on CrowdPose dataset, but when I evaluate my model on the test set by using the command `python tools/valid.py --cfg xxx.xml TEST.MODEL_FILE...

This is how I understand the installation steps in README: 1. Install the `xformers` . (I got pytorch, cuda, cudnn installed on my pc already.) 2. Run `git clone https://github.com/Stability-AI/stablediffusion.git`...

[Bug]: ModuleNotFoundError: No module named 'scripts.hypertile_xyz'

5

### Checklist - [ ] The issue exists after disabling all extensions - [X] The issue exists on a clean installation of webui - [ ] The issue is caused...

RuntimeError: The size of tensor a (0) must match the size of tensor b (2) at non-singleton dimension 0

4

``` TRAINING FROM SCRATCH NHEADS 4 NO WEIGHT SHARING!!! /environment/miniconda3/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/ddp.py:21: LightningDeprecationWarning: The `pl.plugins.training_type.ddp.DDPPlugin` is deprecated in v1.6 and will be removed in v1.8. Use `pl.strategies.ddp.DDPStrategy` instead. "The `pl.plugins.training_type.ddp.DDPPlugin` is deprecated...