FoundationVision / LlamaGen

Autoregressive Model Beats Diffusion: 🦙 Llama for Scalable Image Generation

https://arxiv.org/abs/2406.06525

FoundationVision/LlamaGen Issues

Training Results
Updated a month ago2
tokenizer of 4 dim
Updated a month ago
Questions about the discriminator
Closed a month ago1
Questions about the results of your experiment.
Updated a month ago
T2I Data
Updated a month ago
RuntimeError: shape '[1, 512, 1, 32, 2]' is invalid for input of size 16448
Updated a month ago5
你好，vq_ds16_c2i_training.pt 在私有的通用图片上finetune效果变差了，就试了在imgenet数据集接着finetune，效果也变差了，想问一下是哪里出问题了？
Updated a month ago1
Discriminator is not training properly?
Closed a month ago4
Inquiry about the OpenImages dataset
Closed a month ago2
FID Evaluation not matching paper results for VQ-16 checkpoint
Updated a month ago2
Text embedding inject
Closed a month ago2
Question about why not try using image tokenizer and a ready made llama3 etc LLM model with lora?
Updated a month ago9
你好，我使用您的代码训练了我自己的数据集，但是图片变得特别模糊
Updated a month ago9
Question about text-conditional generation.
Closed a month ago2
Which parameters are trainable? Are the encoder and decoder in VQGAN fixed? Is the llama fixed?
Updated a month ago1
Train script
Updated a month ago1
[Feature] Inpainting script
Updated a month ago
[Feature] ControlNet support via process similar to PixArt's ControlNet-Transformer
Updated a month ago
why using constant lr?
Closed a month ago1
Mismatched model weights document
Closed a month ago2
Issues about the 3B model
Updated 2 months ago1
Error in FID evaluation
Updated 2 months ago2
VQ-VAE ckpt optimizer states?
Updated 2 months ago3
Unexpected error when replicating experiments
Closed 2 months ago3
Will this line of code blur the image during evaluation? tokenizer/tokenizer_image/reconstruction_vq_ddp.py: transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5], inplace=True)
Updated 2 months ago1
Lacks Detailed Documentation
Updated 2 months ago
FSDP training gets stuck after saving the weights
Closed 2 months ago1
Training Details?
Closed 2 months ago4
Could you please tell me which trick has the greatest impact on the overall effect?
Closed 2 months ago5
can i train in colab free?
Updated 2 months ago1
How to use with diffusers?
Updated 2 months ago1
HF model
Updated 2 months ago2
Training cost
Updated 2 months ago2
ModuleNotFoundError: No module named 'vllm.engine.ray_utils'
Updated 2 months ago1
About t2i weights
Closed 2 months ago1