Commit History
-
Upload PixArtSigmaPipeline
Verified signaturef45be97
-
Trained for 5 epochs and 6500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix1613317 -
Trained for 5 epochs and 6000 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fixacb522c -
Trained for 5 epochs and 5500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fixd29036f -
Trained for 4 epochs and 5000 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix9cf439f -
Trained for 4 epochs and 4500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix2266c91 -
Trained for 3 epochs and 4000 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix12fe4cb
-
Trained for 3 epochs and 3500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fixb50e31d -
Trained for 2 epochs and 3000 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix3e6f26b -
Trained for 2 epochs and 2500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix8e45edf -
Trained for 1 epochs and 2000 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix01b2859 -
Trained for 1 epochs and 1500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 4 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fixa17b715
-
Trained for 0 epochs and 1000 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 3 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fix559cc5a -
Trained for 0 epochs and 500 steps.
Verified signatureTrained with datasets ['text-embeds', 'mj-v6'] Learning rate 8e-06, batch size 32, and 3 gradient accumulation steps. Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False Using 'trailing' timestep spacing. Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS VAE: madebyollin/sdxl-vae-fp16-fixb659c17