Scout
Scout
AFAIK, In Pytorch Lightning, there are 4 running modes including [train/val/test/predict] (or maybe 5? including tune). But now in lightning-hydra-template, for using these different running modes, different *entry files* [train.py/val.py/test.py(eval.py)/predict.py]...
https://github.com/menyifang/ADGAN/blob/4dd70649ad136829b92dd6a1a823af7594a0220f/models/model_adgen.py#L37-L40 I noticed that you add an `LinearBlock` after obtain the style code by encoding every part of an image and its segment, It seems that this part is not...
 
Currently, it seems that for input pytorch float tensor between -1 and 1, a min-max normalization is taken: https://github.com/xwying/torchshow/blob/3008c23ebac2811f07c314119b14ba1725b215f7/torchshow/visualization.py#L264-L287 However, usually people use `transforms.Normalize((0.5,), (0.5,))` to normalize image. So I...
When displaying a large number of images, it seems that the browser loads too much memory causing the browser to crash. Is it possible to add a dynamic loading feature?...
same reason as #1, thanks!
公司内网有自己的dns服务器，公用的DNSPod无法解析，我把primaryDNS设置为了内网的dns，还把内网域名写入了domain_primary_sample，都还是无法解析 详细配置可见https://github.com/shawn1m/overture/issues/292 ，怎样才能国内外分流的同时访问到内网dns呀？
公司内网有自己的dns服务器，用公用的DNSPod无法解析，我把primaryDNS设置为了内网的dns，还把内网域名写入了domain_primary_sample，都还是无法解析，config.yaml配置如下： ```yaml primaryDNS: - name: LAN address: 10.x.x.x:53 # - name: DNSPod # address: 119.29.29.29:53 protocol: udp socks5Address: timeout: 6 ednsClientSubnet: policy: disable externalIP: noCookie: true ... domainFile: primary: ./domain_primary_sample...
Is there any comparison with lmdb? And is it possible to generate ffcv dataset directly from raw data without going through torch Dataset?
Fix https://github.com/feifeibear/long-context-attention/issues/161 Refine AttnType.TORCH Recommend AttnType.TORCH_EFFICIENT when FA is unavailable