Open furkancoskun opened 1 year ago
Hi furkancoskun, Thanks for reporting the issue. We'll try to reproduce and look into it. just to confirm, the issue shows up in the latest 2.10 neuron-sdk?
Yes, the issue shows up in 2.10
Hi @furkancoskun , We have reproduced the issue and are currently looking at fixing this in a future release. However, you can continue to use batch=1 in the meantime.
If you are looking to use higher batch sizes to improve performance, note that our batch=1 configuration is expected to be performant. We will continue to improve batch=1 performance as well as support multiple batches in future releases.
I have changed the batch sizes of the trace tensor inputs in hf_pretrained_sd2_512_inference.ipynb notebook. Although text encoder, unet and vae_post_quant_conv were compiled, vae wasn't compiled.
batch=2
I get error message:
I have used inf2.8xlarge instance and set 100GB swap space. Any ideas on this batch input compilation problem?