sunbo11112

> 嗨，这[行](https://github.com/TMElyralab/MuseTalk/blob/main/train.py#L562)代码将以格式保存检查点。如果设置为，代码将自动找到最新的检查点并从那里恢复训练。`.bin``resume_from_checkpoint``True` save_model_epoch_interval=1 Training for one epoch, the code will save the model in bin format, right @zzzweakman

Can we use mediapipe instead of FaceAlignment, or do you suggest any faster models for modification

image_pred.shape: torch.Size([32, 3, 256, 256]) concat.shape: torch.Size([8, 3, 1280, 256]) Image saved successfully: ./exp_out/stage1//test/samples/sample_10_cuda_SyncNetScore_1.jpg Steps: 0%| | 14/250000 [02:39

Unable to load model training from training node, prompting inconsistent weights