Hongyang-Du commited on
Commit
cbf344e
·
verified ·
1 Parent(s): 86b6354

Add the FID reference statistics

Browse files
README.md CHANGED
@@ -32,6 +32,7 @@ Code, configs and instructions: [github.com/Hongyang-Du/FuseLift](https://github
32
  | `k23/h0_stats_raev2k23.pt` | h0 statistics for the K=23 decoder, same format |
33
  | `k23/decoder_raev2k23.pt` | official RAEv2 DINOv3-L K=23 decoder (layers 1, 2, ..., 23), unchanged |
34
  | `samples/ditxl_h0_raev2k7_ep080_ig178_50k.npz` | the 50,000 guided K=7 samples behind the 0.98 gFID: `arr_0`, uint8 `[50000, 256, 256, 3]`, 50 per class |
 
35
 
36
  The two decoders are byte-identical copies of `stage1/imagenet/dinov3l-k7/decoder.pt` and
37
  `stage1/imagenet/dinov3l-k23/decoder.pt` from
 
32
  | `k23/h0_stats_raev2k23.pt` | h0 statistics for the K=23 decoder, same format |
33
  | `k23/decoder_raev2k23.pt` | official RAEv2 DINOv3-L K=23 decoder (layers 1, 2, ..., 23), unchanged |
34
  | `samples/ditxl_h0_raev2k7_ep080_ig178_50k.npz` | the 50,000 guided K=7 samples behind the 0.98 gFID: `arr_0`, uint8 `[50000, 256, 256, 3]`, 50 per class |
35
+ | `reference/imagenet_256_fid_stats.npz` | the FID reference behind every gFID here: Inception `mu` (`[2048]`) and `sigma` (`[2048, 2048]`) of ImageNet-256 train images. Unchanged copy of the default fd_evaluator reference from [nanovisionx/nanogen-evals-stats](https://huggingface.co/datasets/nanovisionx/nanogen-evals-stats) (SHA-256 `925e8b5b…0775f1ac`) |
36
 
37
  The two decoders are byte-identical copies of `stage1/imagenet/dinov3l-k7/decoder.pt` and
38
  `stage1/imagenet/dinov3l-k23/decoder.pt` from
reference/imagenet_256_fid_stats.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:925e8b5b4ced42137f9847f97a63250a2bd59b70f33f3f356e03453d0775f1ac
3
+ size 33563124