search

vitoplantamura / OnnxStream

Lightweight inference library for ONNX files, written in C++. It can run Stable Diffusion XL 1.0 on a RPI Zero 2 (or in 298MB of RAM) but also Mistral 7B on desktops and servers. ARM, x86, WASM, RISC-V supported. Accelerated by XNNPACK.
https://yolo.vitoplantamura.com/
Other
1.86k stars 84 forks source link

=== ERROR === Model::get_tensor_data: input tensor not found: input_5F_ids #23

Open JonRiv opened 1 year ago

JonRiv commented 1 year ago

I'm getting tensor errors after the build is completed. Not sure I do things right, as I had to create a symbolic link text_encoder_fp32 -> stable-diffusion-xl-base-1.0-onnxstream/sdxl_text_encoder_1_fp32 to do so.

./sd
----------------[start]------------------
positive_prompt: a photo of an astronaut riding a horse on mars
negative_prompt: ugly, blurry
output_png_path: ./result.png
steps: 10
----------------[prompt]------------------
=== ERROR === Model::get_tensor_data: input tensor not found: input_5F_ids
vitoplantamura commented 1 year ago

hi,

Please specify the "--xl" parameter for Stable Diffusion XL. You can use the "--models-path" parameter to specify the directory that contains the files downloaded from Hugging Face. Use the "--rpi" parameter if you are using a Raspberry Pi.

Vito