Emma Thompson

I want to install autodiff by copying, so I create a new folder named autodiff_example, then under this root directory, I create a subfolder named 3rdparty, I copy autodiff into...

Where is the code related to “variable resolutions and aspect ratios” ?

2

I'm very interested in this feature, could you please tell me where the relevant code is ?

What is the algorithm of accelation ?

1

I see the acceleration seems being achieved by `origin_by_velocity_and_sample` and `psuedo_velocity_wrt_noisy_and_timestep` funtions, could you please tell us the algorithm behind it ?

[Performance] How to reduce gpu memory consumption ？

1

### Describe the issue I have a onnx model whose size is only 204.57MB，but when I create the session, gpu memory consumpation comes 1.16GB, when inferencing, the gpu memory consumpation...

How to strictly limiting the maximum GPU memory usage and clear GPU memory cache?

3

My use case scenario is deploying model inference services in the cloud, utilizing GPU virtualization technology to split one GPU into multiple instances. Each instance runs a model, and since...

some questions about trtexec's batch size and workspace

4

I have a onnx model which has two input and one output with dynamic shape:  I convert it to trt model with trtexec with command:`trtexec --onnx=weights/wav2lip/wav2lip.onnx --saveEngine=weights/wav2lip/wav2lip_2048_minoptmax.trt --workspace=2048 --minShapes="...

How to improve the accuracy of FP16 model ?

2

I used Polygraphy to compare the accuracy of ONNX FP32 and TensorRT FP16 with following command : ``` polygraphy run weights/model.onnx \ --onnxrt --trt \ --workspace=4096M \ --atol 1e-3 --rtol...

Thanks for opensourceing the evaluation code ! I run the command`python scripts/synthesize_hierarchy.py eval TED-Gesture-output/train_hierarchy/ted_gesture_hierarchy_checkpoint_best.bin 10`, then this is its output: ``` loading checkpoint TED-Gesture-output/train_hierarchy/ted_gesture_hierarchy_checkpoint_best.bin epoch 98 Embedding size is 32,...

I notice there is a function `def check_torch_gpu()` says it may not be able to train FLUX model properly with torch version ≤ 2.3.0, what exactly is the problem ?

Definition of 21 implicit 3D keypoint

3

I want to know each of 21 implicit 3D keypoint control where ? Is there a 2d projection landmark visualization image like this ?