google

/

ddpm-cifar10-32

like 72

Follow

Google 31.1k

https://github.com/mseitzer/pytorch-fid.
To generate the images I used code that similar to the provided example.

\n","updatedAt":"2022-09-21T13:07:53.319Z","author":{"_id":"62fbbc5c21c444a56f7a564b","avatarUrl":"/avatars/bc5afad3cad6af8b58840b7aebb1acf5.svg","fullname":"Ori Matityahu","name":"ori-m-2022a","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"editors":["ori-m-2022a"],"editorAvatarUrls":["/avatars/bc5afad3cad6af8b58840b7aebb1acf5.svg"],"reactions":[],"isReport":false}},{"id":"63adbfc18b50b36a189bd305","author":{"_id":"6144d541fa30a4ba3ef7ef15","avatarUrl":"/avatars/eeb849eb29e249fd1b40e5c10d9bf9e5.svg","fullname":"Alex Havrilla","name":"Dahoas","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":73,"isOwner":false,"isOrgMember":false},"createdAt":"2022-12-29T16:26:41.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Any update on this? I also observe the same FID.","html":"

Any update on this? I also observe the same FID.

\n","updatedAt":"2022-12-29T16:26:41.472Z","author":{"_id":"6144d541fa30a4ba3ef7ef15","avatarUrl":"/avatars/eeb849eb29e249fd1b40e5c10d9bf9e5.svg","fullname":"Alex Havrilla","name":"Dahoas","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":73}},"numEdits":0,"editors":["Dahoas"],"editorAvatarUrls":["/avatars/eeb849eb29e249fd1b40e5c10d9bf9e5.svg"],"reactions":[],"isReport":false}},{"id":"63bd791408409827334d07d7","author":{"_id":"6144d541fa30a4ba3ef7ef15","avatarUrl":"/avatars/eeb849eb29e249fd1b40e5c10d9bf9e5.svg","fullname":"Alex Havrilla","name":"Dahoas","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":73,"isOwner":false,"isOrgMember":false},"createdAt":"2023-01-10T14:41:24.000Z","type":"comment","data":{"edited":true,"hidden":false,"latest":{"raw":"@patrickvonplaten @anton-l can you perhaps comment?","html":"

\n\n@patrickvonplaten\n\t \n\n@anton-l\n\t can you perhaps comment?

\n","updatedAt":"2023-01-10T14:44:38.544Z","author":{"_id":"6144d541fa30a4ba3ef7ef15","avatarUrl":"/avatars/eeb849eb29e249fd1b40e5c10d9bf9e5.svg","fullname":"Alex Havrilla","name":"Dahoas","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":73}},"numEdits":1,"editors":["Dahoas"],"editorAvatarUrls":["/avatars/eeb849eb29e249fd1b40e5c10d9bf9e5.svg"],"reactions":[],"isReport":false}},{"id":"63ec5d64308bd1f7fcbff129","author":{"_id":"63ec5cde59a46fd5f4b739ca","avatarUrl":"/avatars/efb1ce91a4ae5e42da4951e2aac8d91b.svg","fullname":"Jaeyun_Lee","name":"TurtleNeck","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-02-15T04:19:48.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Any update for this? I trained improved-ddpm code from https://github.com/openai/improved-diffusion and check FID score which is much worse than reported in the paper... :(","html":"

Any update for this? I trained improved-ddpm code from https://github.com/openai/improved-diffusion and check FID score which is much worse than reported in the paper... :(

\n","updatedAt":"2023-02-15T04:19:48.911Z","author":{"_id":"63ec5cde59a46fd5f4b739ca","avatarUrl":"/avatars/efb1ce91a4ae5e42da4951e2aac8d91b.svg","fullname":"Jaeyun_Lee","name":"TurtleNeck","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"editors":["TurtleNeck"],"editorAvatarUrls":["/avatars/efb1ce91a4ae5e42da4951e2aac8d91b.svg"],"reactions":[],"isReport":false}},{"id":"63ff06a75caf40b7417fd318","author":{"_id":"5dfcb1aada6d0311fd3d5448","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg","fullname":"Patrick von Platen","name":"patrickvonplaten","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":725,"isOwner":false,"isOrgMember":false},"createdAt":"2023-03-01T08:02:47.000Z","type":"comment","data":{"edited":true,"hidden":false,"latest":{"raw":"Hey @TurtleNeck,\n\nDiffusers DDPM is an exact clone of: https://github.com/pesser/pytorch_diffusion\n\nwhich is the PyTorch implementation of orig TF Version","html":"

Hey \n\n@TurtleNeck\n\t,

\n

Diffusers DDPM is an exact clone of: https://github.com/pesser/pytorch_diffusion

\n

which is the PyTorch implementation of orig TF Version

\n","updatedAt":"2023-03-01T08:03:00.268Z","author":{"_id":"5dfcb1aada6d0311fd3d5448","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg","fullname":"Patrick von Platen","name":"patrickvonplaten","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":725}},"numEdits":1,"editors":["patrickvonplaten"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg"],"reactions":[],"isReport":false}},{"id":"63ff06d46bb959b94e6970a7","author":{"_id":"5dfcb1aada6d0311fd3d5448","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg","fullname":"Patrick von Platen","name":"patrickvonplaten","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":725,"isOwner":false,"isOrgMember":false},"createdAt":"2023-03-01T08:03:32.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Can you check whether you get better results with presser/pytorch_diffusion?","html":"

Can you check whether you get better results with presser/pytorch_diffusion?

\n","updatedAt":"2023-03-01T08:03:32.869Z","author":{"_id":"5dfcb1aada6d0311fd3d5448","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg","fullname":"Patrick von Platen","name":"patrickvonplaten","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":725}},"numEdits":0,"editors":["patrickvonplaten"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg"],"reactions":[],"isReport":false}},{"id":"63ff075625b278f4117d8ab6","author":{"_id":"5dfcb1aada6d0311fd3d5448","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg","fullname":"Patrick von Platen","name":"patrickvonplaten","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":725,"isOwner":false,"isOrgMember":false},"createdAt":"2023-03-01T08:05:42.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Also note that this is the non ema checkpoint, maybe they use the ema checkpoint in their paper for FID evaluation.","html":"

Also note that this is the non ema checkpoint, maybe they use the ema checkpoint in their paper for FID evaluation.

\n","updatedAt":"2023-03-01T08:05:42.794Z","author":{"_id":"5dfcb1aada6d0311fd3d5448","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg","fullname":"Patrick von Platen","name":"patrickvonplaten","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":725}},"numEdits":0,"editors":["patrickvonplaten"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1584435275418-5dfcb1aada6d0311fd3d5448.jpeg"],"reactions":[],"isReport":false}},{"id":"6423104bae7fe472976f2899","author":{"_id":"62fbbc5c21c444a56f7a564b","avatarUrl":"/avatars/bc5afad3cad6af8b58840b7aebb1acf5.svg","fullname":"Ori Matityahu","name":"ori-m-2022a","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-03-28T16:05:31.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Is there an ema checkpoint for this model?","html":"

Is there an ema checkpoint for this model?

\n","updatedAt":"2023-03-28T16:05:31.277Z","author":{"_id":"62fbbc5c21c444a56f7a564b","avatarUrl":"/avatars/bc5afad3cad6af8b58840b7aebb1acf5.svg","fullname":"Ori Matityahu","name":"ori-m-2022a","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"editors":["ori-m-2022a"],"editorAvatarUrls":["/avatars/bc5afad3cad6af8b58840b7aebb1acf5.svg"],"reactions":[],"isReport":false}},{"id":"6464947b1054f08ddba099f1","author":{"_id":"63ce831481065f5e3ffff2e3","avatarUrl":"/avatars/4d0b9d0d58ab3cb7097fe714a5377a7d.svg","fullname":"Peter Lorenz","name":"j35t3r","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-05-17T08:46:51.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"here is one ema checkpoint: https://github.com/pesser/pytorch_diffusion","html":"

here is one ema checkpoint: https://github.com/pesser/pytorch_diffusion

\n","updatedAt":"2023-05-17T08:46:51.182Z","author":{"_id":"63ce831481065f5e3ffff2e3","avatarUrl":"/avatars/4d0b9d0d58ab3cb7097fe714a5377a7d.svg","fullname":"Peter Lorenz","name":"j35t3r","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"editors":["j35t3r"],"editorAvatarUrls":["/avatars/4d0b9d0d58ab3cb7097fe714a5377a7d.svg"],"reactions":[],"isReport":false}},{"id":"646a1aec96cfe72aef787baf","author":{"_id":"646a1939c37ca1e12308fe81","avatarUrl":"/avatars/752e9d86018e7d33ad8bcd741203fd86.svg","fullname":"Gongfan Fang","name":"Vinnnf","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isOwner":false,"isOrgMember":false},"createdAt":"2023-05-21T13:21:48.000Z","type":"comment","data":{"edited":true,"hidden":false,"latest":{"raw":"Hey guys. We converted the [official EMA checkpoint]( https://github.com/pesser/pytorch_diffusion) to Diffusers. Hope this will be helpful: https://github.com/VainF/Diff-Pruning/releases/download/v0.0.1/ddpm_ema_cifar10.zip\n\nA bash script for converting is available at https://github.com/VainF/Diff-Pruning/blob/main/tools/convert_cifar10_ddpm_ema.sh \n\nThis checkpoint achieves an FID score of 4.5 with 100 DDIM steps.","html":"

Hey guys. We converted the official EMA checkpoint to Diffusers. Hope this will be helpful: https://github.com/VainF/Diff-Pruning/releases/download/v0.0.1/ddpm_ema_cifar10.zip

\n

A bash script for converting is available at https://github.com/VainF/Diff-Pruning/blob/main/tools/convert_cifar10_ddpm_ema.sh

\n

This checkpoint achieves an FID score of 4.5 with 100 DDIM steps.

\n","updatedAt":"2023-05-21T15:45:43.911Z","author":{"_id":"646a1939c37ca1e12308fe81","avatarUrl":"/avatars/752e9d86018e7d33ad8bcd741203fd86.svg","fullname":"Gongfan Fang","name":"Vinnnf","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6}},"numEdits":3,"editors":["Vinnnf"],"editorAvatarUrls":["/avatars/752e9d86018e7d33ad8bcd741203fd86.svg"],"reactions":[{"reaction":"👍","users":["Mqleet"],"count":1}],"isReport":false}},{"id":"64f99c34a5067f6b6552f064","author":{"_id":"64b94bf05067873176e4d49b","avatarUrl":"/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg","fullname":"peiqin sun","name":"speiqin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-09-07T09:47:32.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"I use the ema checkpoint, but I get an FID score of 5.23 in 1000 DDPM steps. can you give me a link to you FID scripts? ","html":"

I use the ema checkpoint, but I get an FID score of 5.23 in 1000 DDPM steps. can you give me a link to you FID scripts?

\n","updatedAt":"2023-09-07T09:47:32.962Z","author":{"_id":"64b94bf05067873176e4d49b","avatarUrl":"/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg","fullname":"peiqin sun","name":"speiqin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8925860524177551},"editors":["speiqin"],"editorAvatarUrls":["/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg"],"reactions":[],"isReport":false}},{"id":"64f9d253615b4ffedf390467","author":{"_id":"646a1939c37ca1e12308fe81","avatarUrl":"/avatars/752e9d86018e7d33ad8bcd741203fd86.svg","fullname":"Gongfan Fang","name":"Vinnnf","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isOwner":false,"isOrgMember":false},"createdAt":"2023-09-07T13:38:27.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"> I use the ema checkpoint, but I get an FID score of 5.23 in 1000 DDPM steps. can you give me a link to you FID scripts?\n\nHi @speiqin, I modified the original diffusers codebase to fix [this issue](https://github.com/huggingface/diffusers/issues/2585). You can try [this script](https://github.com/VainF/Diff-Pruning/blob/main/scripts/sample_ddpm_cifar10_pretrained.sh) to reproduce my results (50k samples, FID=4.5, 100-step DDIM). But I think DDPM sampling is not affected by this issue. How many samples did you use for FID estimation?","html":"

\n

I use the ema checkpoint, but I get an FID score of 5.23 in 1000 DDPM steps. can you give me a link to you FID scripts?

\n

\n

Hi \n\n@speiqin\n\t, I modified the original diffusers codebase to fix this issue. You can try this script to reproduce my results (50k samples, FID=4.5, 100-step DDIM). But I think DDPM sampling is not affected by this issue. How many samples did you use for FID estimation?

\n","updatedAt":"2023-09-07T13:38:27.361Z","author":{"_id":"646a1939c37ca1e12308fe81","avatarUrl":"/avatars/752e9d86018e7d33ad8bcd741203fd86.svg","fullname":"Gongfan Fang","name":"Vinnnf","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8265745043754578},"editors":["Vinnnf"],"editorAvatarUrls":["/avatars/752e9d86018e7d33ad8bcd741203fd86.svg"],"reactions":[],"isReport":false}},{"id":"64faf75ada4429f25ddce8e2","author":{"_id":"64b94bf05067873176e4d49b","avatarUrl":"/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg","fullname":"peiqin sun","name":"speiqin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-09-08T10:28:42.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Hi, @Vinnnf . I get the FID result which close to ddpm paper. the most importance error in my experiment( FID only 5.23) is that I only use 1w images to calculate the FID score. Now, I use 5w is get the right result. Thanks for your checkpoint. ","html":"

Hi, \n\n@Vinnnf\n\t . I get the FID result which close to ddpm paper. the most importance error in my experiment( FID only 5.23) is that I only use 1w images to calculate the FID score. Now, I use 5w is get the right result. Thanks for your checkpoint.

\n","updatedAt":"2023-09-08T10:28:42.242Z","author":{"_id":"64b94bf05067873176e4d49b","avatarUrl":"/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg","fullname":"peiqin sun","name":"speiqin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8345819115638733},"editors":["speiqin"],"editorAvatarUrls":["/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg"],"reactions":[{"reaction":"👍","users":["j35t3r"],"count":1}],"isReport":false}},{"id":"653fe07a01b0d514c8978893","author":{"_id":"6537815a62dd8126cf97d4af","avatarUrl":"/avatars/3f20b124ac6f62343d2085c5e43765fa.svg","fullname":"Tang Fuzhi","name":"the-weakest","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-10-30T16:57:30.000Z","type":"comment","data":{"edited":true,"hidden":false,"latest":{"raw":"@ori-m-2022a Hello. Could you tell me the finally FID score of 50k samples of the checkpoints \"google/ddpm-cifar10-32\" when T = 1000? I got 12.9786970050562 I don't know whether it 's correct. Thank you very much!","html":"

\n\n@ori-m-2022a\n\t Hello. Could you tell me the finally FID score of 50k samples of the checkpoints \"google/ddpm-cifar10-32\" when T = 1000? I got 12.9786970050562 I don't know whether it 's correct. Thank you very much!

\n","updatedAt":"2023-10-30T17:02:30.044Z","author":{"_id":"6537815a62dd8126cf97d4af","avatarUrl":"/avatars/3f20b124ac6f62343d2085c5e43765fa.svg","fullname":"Tang Fuzhi","name":"the-weakest","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.7941372990608215},"editors":["the-weakest"],"editorAvatarUrls":["/avatars/3f20b124ac6f62343d2085c5e43765fa.svg"],"reactions":[],"isReport":false}},{"id":"65505c384c6f4e3300234ce3","author":{"_id":"64b94bf05067873176e4d49b","avatarUrl":"/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg","fullname":"peiqin sun","name":"speiqin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2023-11-12T05:01:44.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"hi, @Vinnnf. I found the dropout rate in converted checkpoint is 0.0, but the offical model is setting 0.1 in the training. This is a wrong?","html":"

hi, \n\n@Vinnnf\n\t. I found the dropout rate in converted checkpoint is 0.0, but the offical model is setting 0.1 in the training. This is a wrong?

\n","updatedAt":"2023-11-12T05:01:44.098Z","author":{"_id":"64b94bf05067873176e4d49b","avatarUrl":"/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg","fullname":"peiqin sun","name":"speiqin","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.964205801486969},"editors":["speiqin"],"editorAvatarUrls":["/avatars/ad322d0bcf5a6ac3bcca228b1b10f4c0.svg"],"reactions":[],"isReport":false}},{"id":"65cc1fb39a4ee02bbbbaab58","author":{"_id":"6382bdcb2a85b6714b84a619","avatarUrl":"/avatars/b36fc22c2290934820c2c0faebb55087.svg","fullname":"isaacNingLee","name":"ilee0022","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2024-02-14T02:04:35.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Hi all, I am sampling 10k examples with DDPM(T=1000) and 50k DDIM(T=100) and 10k PNDM(T=100). The best FID score I can get is just around 36. I am computing the sampled image in respect to the images from torchvision cifar10 dataset saved in a directory. The script that I use to compute FID is https://github.com/mseitzer/pytorch-fid . I am very new to FID and I am not too sure which part did I made some mistakes. I really appreciate any help and advices.\n\nCheers","html":"

Hi all, I am sampling 10k examples with DDPM(T=1000) and 50k DDIM(T=100) and 10k PNDM(T=100). The best FID score I can get is just around 36. I am computing the sampled image in respect to the images from torchvision cifar10 dataset saved in a directory. The script that I use to compute FID is https://github.com/mseitzer/pytorch-fid . I am very new to FID and I am not too sure which part did I made some mistakes. I really appreciate any help and advices.

\n

Cheers

\n","updatedAt":"2024-02-14T02:04:35.477Z","author":{"_id":"6382bdcb2a85b6714b84a619","avatarUrl":"/avatars/b36fc22c2290934820c2c0faebb55087.svg","fullname":"isaacNingLee","name":"ilee0022","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9119212031364441},"editors":["ilee0022"],"editorAvatarUrls":["/avatars/b36fc22c2290934820c2c0faebb55087.svg"],"reactions":[],"isReport":false}},{"id":"65df674b8f6ea1ca7079457b","author":{"_id":"65095f5af99720fea1297b9b","avatarUrl":"/avatars/d3c1afa1e004b8bff91b6ede48d65a48.svg","fullname":"Cassie Li","name":"Cassie-0121","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2024-02-28T17:03:07.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"> Hi all, I am sampling 10k examples with DDPM(T=1000) and 50k DDIM(T=100) and 10k PNDM(T=100). The best FID score I can get is just around 36. I am computing the sampled image in respect to the images from torchvision cifar10 dataset saved in a directory. The script that I use to compute FID is https://github.com/mseitzer/pytorch-fid . I am very new to FID and I am not too sure which part did I made some mistakes. I really appreciate any help and advices.\n> \n> Cheers\n\nsame process, how did you solve this? many thanks\n","html":"

\n

Hi all, I am sampling 10k examples with DDPM(T=1000) and 50k DDIM(T=100) and 10k PNDM(T=100). The best FID score I can get is just around 36. I am computing the sampled image in respect to the images from torchvision cifar10 dataset saved in a directory. The script that I use to compute FID is https://github.com/mseitzer/pytorch-fid . I am very new to FID and I am not too sure which part did I made some mistakes. I really appreciate any help and advices.

\n

Cheers

\n

\n

same process, how did you solve this? many thanks

\n","updatedAt":"2024-02-28T17:03:07.231Z","author":{"_id":"65095f5af99720fea1297b9b","avatarUrl":"/avatars/d3c1afa1e004b8bff91b6ede48d65a48.svg","fullname":"Cassie Li","name":"Cassie-0121","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9296927452087402},"editors":["Cassie-0121"],"editorAvatarUrls":["/avatars/d3c1afa1e004b8bff91b6ede48d65a48.svg"],"reactions":[],"isReport":false}},{"id":"667298be0560603efd36d4d5","author":{"_id":"6448b2f53e7b3c11be684348","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6448b2f53e7b3c11be684348/QvlUQG3pWf8ZyEVBV6F7w.jpeg","fullname":"Qianli Ma","name":"Mqleet","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isOwner":false,"isOrgMember":false},"createdAt":"2024-06-19T08:37:18.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Any one have solved this problem? I generated 50K images using 100 steps DDIMSampler, and I got a FID of 11.214451002914018 which is far away from 3.17....","html":"

Any one have solved this problem? I generated 50K images using 100 steps DDIMSampler, and I got a FID of 11.214451002914018 which is far away from 3.17....

\n","updatedAt":"2024-06-19T08:37:18.076Z","author":{"_id":"6448b2f53e7b3c11be684348","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6448b2f53e7b3c11be684348/QvlUQG3pWf8ZyEVBV6F7w.jpeg","fullname":"Qianli Ma","name":"Mqleet","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8826470971107483},"editors":["Mqleet"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6448b2f53e7b3c11be684348/QvlUQG3pWf8ZyEVBV6F7w.jpeg"],"reactions":[],"isReport":false}}],"pinned":false,"locked":false,"collection":"discussions","isPullRequest":false,"isReport":false},"primaryEmailConfirmed":false,"repo":{"name":"google/ddpm-cifar10-32","type":"model"},"discussionRole":0,"acceptLanguages":["*"],"hideComments":true,"repoDiscussionsLocked":false,"isDiscussionAuthor":false}">