Closed kayleeliyx closed 5 years ago
Hi, you need to place the training checkpoint in the ckpt folder. you can use the ones from your training or the one I provided (the link is in the readme).
For training, you can just let it run all the way as in my train.sh (which runs for 2+70 epochs). That should be sufficient for the model to converge.
When I run
inference_example.sh
, I faced the following error. I have create a new folder called ckpt.The training step works well. By the way, what loss is recommanded to stop training? Thank you so much for your help.