CLIP-ONNX

CLIP-ONNX copied to clipboard

Replace the operator of "torch.einsum"

2

q, k, v = (torch.einsum("tbh, oh -> tbo", x, self.attn.in_proj_weight) + self.attn.in_proj_bias).contiguous().chunk( 3, dim=-1) @Lednik7 Thanks for your great work on Clip-ONNX. for the pytorch operator of "torch.einsum" , if...

zhangnju

Textual .forward() override bug.

3

Hi! First of all, thanks a lot for a great example project! However, I'm pretty sure I have found a bug. onnx_model.encode_text() produces results that are not consistent with regular...

Great-Frosty