From 60c384bcd2b6f0cf9569fa8999ac8f7eff98b31a Mon Sep 17 00:00:00 2001 From: pink-red <94079189+pink-red@users.noreply.github.com> Date: Wed, 12 Oct 2022 23:43:37 +0400 Subject: [PATCH] Fix fine-tuning compatibility with deepspeed (#816) --- examples/text_to_image/train_text_to_image.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/text_to_image/train_text_to_image.py b/examples/text_to_image/train_text_to_image.py index 2a29b7fa2d..c3c3c75438 100644 --- a/examples/text_to_image/train_text_to_image.py +++ b/examples/text_to_image/train_text_to_image.py @@ -568,7 +568,7 @@ def main(): # Predict the noise residual and compute loss noise_pred = unet(noisy_latents, timesteps, encoder_hidden_states).sample - loss = F.mse_loss(noise_pred, noise, reduction="mean") + loss = F.mse_loss(noise_pred.float(), noise.float(), reduction="mean") # Gather the losses across all processes for logging (if we use distributed training). avg_loss = accelerator.gather(loss.repeat(args.train_batch_size)).mean()