ORTStableDiffusionPipeline Support + Unexpected input data type. Actual: (tensor(float)) , expected: (tensor(int64))

[m4a1carbin4]

Ok... so I just want to use my stable diffusion model combine with my own lora + LCM_LoRa

I Use Optimum to convert model [SD1.5 base model + my lora + LCM_LoRa] to onnx

But I think this repo doesn't have ORTStableDiffusionPipeline Support.

SO I test it with LatentConsistencyModelPipeline, and StableDiffusionPipeline + LCMScheduler (convert DiffusionPipeline.ts)

But I always got same error.

 ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ 0% | ETA: 0s | 0/31/home/waganawa/Documents/Code/node/diffusers.js/node_modules/onnxruntime-node/dist/backend.js:45
                    resolve(__classPrivateFieldGet(this, _OnnxruntimeSessionHandler_inferenceSession, "f").run(feeds, fetches, options));
                                                                                                           ^

Error: Unexpected input data type. Actual: (tensor(float)) , expected: (tensor(int64))
    at /home/waganawa/Documents/Code/node/diffusers.js/node_modules/onnxruntime-node/dist/backend.js:45:108
    at process.processTicksAndRejections (node:internal/process/task_queues:77:11)

well I don't Know what is causing this error.... Someone help please?

my onnx repo : https://huggingface.co/WGNW/chamcham_v1_checkpoint_onnx

[m4a1carbin4]

I found what is problem from this comment

// for some reason v1.4 takes int64 as timestep input. ideally we should get input dtype from the model // but currently onnxruntime-node does not give out types, only input names

[kungfooman]

Did you make it work @m4a1carbin4?

I found an ONNX model and it causes the same error. I didn't dig too deep into the issue yet, but I assume we have to wait for an update or prototype patch onnxruntime in order to receive the dtype and act accordingly?

[m4a1carbin4]

In My case it's just misstake of mine

link see this link will help you. it just a matter of settings.