YYinWild

> the generated onnx input width which is image lenght is fixed? Then how to handle different lenght charactors input prediction when using onnx or tensorrt inference? Thanks for your...

> The model is imported into onnx, and converted into an engine file to infer on tensorrt7.0, which can be successfully inferred with trtexec, but the result of inference with...

Same problem, did you solve this problem?

> Thank you for your comments. I implemented this a while ago and modified the backbone based on pix2pixHD, so there might be some discrepancy compared to the original paper....

Hi, Thanks for your interest in our project. I will take a look into your proposed situation combined with quantization and pruning. If it is a bug, I will try...

Hello Thomas, Thanks for your further information. I think that combining quantization and pruning in one training process seems to make sense. I will investigate this situation and find out...

Hi! Sorry for the late response. I think the method you proposed seems not supported in our current API. However, I believe your idea is interesting, and I will look...

I found out why prompt_embeds's dim zero is always 2 times of id_embeds. This is because the [num_tokens](https://github.com/TencentARC/PhotoMaker/blob/main/photomaker/pipeline.py#L226) = 2. Could anyone give a hint of this parameter's correspondence in...