Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / FoundationVision/VAR issues and pull requests
#89 - Training resource for multi-scale VAE
Issue -
State: open - Opened by FRESHMAN2016 about 1 month ago
- 3 comments
#88 - [New!] Try ImageFolder🚀 tokenizer for next-scale AR with shorter tokens length🔥, faster speed 🔥 and better FID🔥
Issue -
State: open - Opened by lxa9867 about 1 month ago
- 2 comments
#87 - Rejection sampling
Issue -
State: open - Opened by wangf3014 about 2 months ago
#86 - Conflicting Instructions for computing FID
Issue -
State: open - Opened by Kumbong about 2 months ago
#85 - FID of VQVAE
Issue -
State: open - Opened by RohollahHS 2 months ago
- 2 comments
#84 - generate images with arbitrary resolutions
Issue -
State: open - Opened by Leiii-Cao 2 months ago
#83 - Cannot reproduce results in Table 1, especially the IS score
Issue -
State: open - Opened by adreamwu 3 months ago
- 1 comment
#82 - How can I train VAR with my own data? Does the num_class parameter in VAR have any effect?
Issue -
State: open - Opened by zhangqingwu 3 months ago
- 1 comment
#81 - class token replaces first token
Issue -
State: closed - Opened by cyw-3d 3 months ago
#80 - Question about the training dataset - tokenzier
Issue -
State: open - Opened by Lucky-Light-Sun 3 months ago
#79 - How can i train this model by custom dataset
Issue -
State: open - Opened by morestart 4 months ago
- 1 comment
#78 - Update VAR code for continuous optimisation
Pull Request -
State: closed - Opened by BartekPog 4 months ago
#77 - Weird/Inconsistency evaluation IS score.
Issue -
State: open - Opened by ChenDRAG 4 months ago
- 2 comments
#76 - Usage of classifier-free guidance
Issue -
State: closed - Opened by ChenDRAG 4 months ago
#75 - we use VAR-CLIP train a TextToImage model on ImageNet,The result seems pretty good.
Issue -
State: open - Opened by daixiangzi 4 months ago
- 3 comments
#74 - About in-painting tasks
Issue -
State: open - Opened by zhiyuanyou 4 months ago
- 1 comment
#73 - Why multi-scale features partially shared a convolution network via PhiPartiallyShared
Issue -
State: open - Opened by sunset-clouds 5 months ago
- 1 comment
#72 - 换个角度,ms codebbok是不是也可以等价于另一种latent diffusion?
Issue -
State: open - Opened by YilanWang 5 months ago
- 1 comment
#71 - Computation and Memory Consumption When Training Models
Issue -
State: open - Opened by walking-shadow 5 months ago
#70 - for in/out painting
Issue -
State: open - Opened by Youngwoo-git 5 months ago
#69 - Can not align FID with provided checkpoint
Issue -
State: open - Opened by LiCHH 5 months ago
- 6 comments
#68 - https://var.vision/demo is borked
Issue -
State: closed - Opened by yosun 5 months ago
- 1 comment
#67 - question about ema_vocab_hit_SV
Issue -
State: open - Opened by shliu0 6 months ago
- 1 comment
#66 - Finetuning on own data
Issue -
State: open - Opened by LetsGoFir 6 months ago
- 1 comment
#65 - the size of latent space
Issue -
State: open - Opened by xinding64 6 months ago
- 1 comment
#64 - T2I Generation
Issue -
State: open - Opened by ucasyjz 6 months ago
- 3 comments
#63 - VQVAE training code
Issue -
State: open - Opened by Junda24 6 months ago
#62 - the patch_nums of 256*256 image
Issue -
State: closed - Opened by xinding64 6 months ago
#61 - Inference code keeps generating the same image
Issue -
State: closed - Opened by moeinheidari7829 6 months ago
- 2 comments
#60 - Training code for VQVAE
Issue -
State: open - Opened by zhangjingze21 6 months ago
#59 - There was no increase in speed after installing flash-attn and xformer.
Issue -
State: open - Opened by kongwanbianjinyu 6 months ago
#58 - Inference after training on own dataset
Issue -
State: closed - Opened by moeinheidari7829 6 months ago
- 2 comments
#57 - The 512 ckp?
Issue -
State: open - Opened by FanqingM 6 months ago
- 2 comments
#56 - Question on autoregressive_infer_cfg
Issue -
State: closed - Opened by sparse-mvs-2 6 months ago
- 2 comments
#55 - Image reconstruction via Transformer.
Issue -
State: open - Opened by minimini-1 6 months ago
- 1 comment
#54 - 请问下训练512x512分辨率的图像也使用16x16的codebook size吗
Issue -
State: closed - Opened by YilanWang 6 months ago
- 2 comments
#53 - Question about the cross-antropy loss average?
Issue -
State: closed - Opened by Yheechou 6 months ago
- 4 comments
#52 - 大佬 关于demo生成的图像的问题
Issue -
State: closed - Opened by BigConsin 6 months ago
- 1 comment
#51 - Scalability to multimodal large language models?
Issue -
State: open - Opened by DEBIHOOD 6 months ago
#51 - Scalability to multimodal large language models?
Issue -
State: open - Opened by DEBIHOOD 6 months ago
#50 - 请问下两个阶段ablation的细节
Issue -
State: closed - Opened by YilanWang 6 months ago
- 3 comments
#49 - training code for VQVAE
Issue -
State: open - Opened by Junda24 6 months ago
- 2 comments
#48 - Can the training log be released?
Issue -
State: open - Opened by daixiangzi 6 months ago
#47 - training with multi-gpu but stuck
Issue -
State: closed - Opened by Erisura 6 months ago
- 2 comments
#46 - class VQVAE forward function error
Issue -
State: closed - Opened by woldier 6 months ago
- 1 comment
#45 - FID misalignment
Issue -
State: closed - Opened by ckczzj 6 months ago
- 4 comments
#44 - How many GPU did you use for training?
Issue -
State: closed - Opened by daixiangzi 7 months ago
- 3 comments
#43 - AdaLNSelfAttn.forward 请问下这句目的是什么
Issue -
State: closed - Opened by yanghu819 7 months ago
- 2 comments
#42 - 请问VQVAE(stage1)阶段是怎样使用多级VectorQuantizer的?
Issue -
State: closed - Opened by YilanWang 7 months ago
- 3 comments
#41 - Abnormal sample results with `demo_sample.ipynb`
Issue -
State: closed - Opened by karrykkk 7 months ago
- 2 comments
#40 - move compute condition out of loop in inference
Pull Request -
State: closed - Opened by FEIFEIEIAr 7 months ago
#38 - FID on Class-Conditioned Evaluation & Normalized Attn in Depth 16/30
Issue -
State: closed - Opened by lxa9867 7 months ago
- 2 comments
#37 - About progressive training
Issue -
State: closed - Opened by ParanoidHW 7 months ago
- 3 comments
#36 - what is your position embeding ? is 2d RoPE good choice ?
Issue -
State: closed - Opened by renjingneng 7 months ago
- 1 comment
#35 - AR Time Complexity?
Issue -
State: closed - Opened by isaacrob 7 months ago
- 5 comments
#34 - Question about the shared codebook
Issue -
State: closed - Opened by jiho314 7 months ago
- 2 comments
#33 - Error reported during training phase
Issue -
State: closed - Opened by zshuo7615 7 months ago
- 1 comment
#32 - The 512*512 model checkpiont?
Issue -
State: closed - Opened by FanqingM 7 months ago
- 2 comments
#31 - Question about the unconditional generation
Issue -
State: closed - Opened by minimini-1 7 months ago
- 2 comments
#30 - Dtype error with flash-attention
Issue -
State: closed - Opened by ThisisBillhe 7 months ago
- 1 comment
#29 - How to set progressive training in traing the VAR?
Issue -
State: closed - Opened by HaozheZhao 7 months ago
- 6 comments
#28 - Training code for VAE
Issue -
State: open - Opened by kl2004 7 months ago
- 22 comments
#27 - flash-atten相关问题
Issue -
State: closed - Opened by Silentssss 7 months ago
- 4 comments
#26 - Training resources
Issue -
State: closed - Opened by songtianhui 7 months ago
- 1 comment
#25 - RuntimeError: Error(s) in loading state_dict for VAR in demo_sample.ipynb
Issue -
State: closed - Opened by sunqiang85 7 months ago
- 3 comments
#24 - 关于并行解码的疑问(Questions About Parallel Decoding)
Issue -
State: closed - Opened by helloheshee 7 months ago
- 3 comments
#23 - Actual Training Loss Curve
Issue -
State: closed - Opened by chenllliang 7 months ago
- 3 comments
#22 - training code
Issue -
State: open - Opened by Jiawei-Yang 7 months ago
- 3 comments
#21 - is it possible to get embedding of last layer of network and add softmax to get a Classifier
Issue -
State: closed - Opened by roodkcab 7 months ago
- 2 comments
#19 - There are errors in the model folders .py files
Issue -
State: closed - Opened by Aibecool174 7 months ago
#18 - Keep getting this error in demo_sample.ipynb
Issue -
State: closed - Opened by Aibecool174 7 months ago
- 2 comments
#17 - Keep getting this error in demo_sample.ipynb
Issue -
State: closed - Opened by Aibecool174 7 months ago
#16 - The performance of VAR Tokenizer
Issue -
State: closed - Opened by youngsheen 7 months ago
- 5 comments
#15 - question about 2 stage training
Issue -
State: closed - Opened by plutoyuxie 7 months ago
- 1 comment
#14 - Implications of Classifier-Free Guidance in Auto-regressive Models
Issue -
State: open - Opened by fkcptlst 7 months ago
- 1 comment
#13 - how to use this for inpaint?
Issue -
State: open - Opened by WenmuZhou 7 months ago
- 5 comments
#12 - module 'fused_dense_lib' has no attribute 'linear_act_forward'
Issue -
State: closed - Opened by tanshuai0219 7 months ago
- 3 comments
#11 - Refactoring
Pull Request -
State: closed - Opened by elysia090 7 months ago
- 1 comment
#10 - Resource consumption
Issue -
State: closed - Opened by xiaosu-zhu 7 months ago
- 2 comments
#9 - Make sure HF download metrics work, better discoverability on Hugging Face
Pull Request -
State: closed - Opened by NielsRogge 7 months ago
- 5 comments
#8 - About the input resolution scale
Issue -
State: closed - Opened by luohao123 7 months ago
- 6 comments
#7 - About testing Zero-shot generalizability
Issue -
State: closed - Opened by Youngwoo-git 7 months ago
- 2 comments
#6 - Training and Inference Scripts
Issue -
State: closed - Opened by achen46 7 months ago
- 5 comments
#5 - about the training code
Issue -
State: closed - Opened by WeianMao 7 months ago
- 3 comments
#4 - fix date error in README
Pull Request -
State: closed - Opened by zhtjtcz 7 months ago
#3 - can it do super resolution?
Issue -
State: open - Opened by francqz31 7 months ago
- 3 comments
#2 - 9 am UTC already past, code and models?
Issue -
State: closed - Opened by yqy2001 7 months ago
- 6 comments
#1 - How do you make Transformer generate tokens in parallel?
Issue -
State: open - Opened by jzhang38 7 months ago
- 36 comments