mirza

Results 5 comments of mirza

Significant Precision Loss After Converting to FP16 in TensorRT Inference

@ganessh22 Hello, Have you succeeded in running inference with the TensorRT model at the proposed speeds of 2-13ms (depending on the model)? I have tried the D-FINE-S (Objects365+COCO) with the...

Warnings When Converting PTH Model to ONNX and slow TensorRT inference

@KozlovKY Thank you for your response. I'm still encountering issues achieving the proposed 2-3ms latency for the small model using TensorRT. My TensorRT inference time remains around 80ms. I've found...

Loss staying high while trying to tune network

Hi there, what is the size of your dataset? How many epochs did you train for? Have you experimented with the hyperparameters? What is the goal of your dataset—what are...

Loss staying high while trying to tune network

If you had a perfect architecture for your problem you'd only need a few datapoints, I would add more images to the dataset: https://en.m.wikipedia.org/wiki/Neural_scaling_law I haven't tried these models on...

Encountering "Expected to have finished reduction in the prior iteration before starting a new one" Error During Training

@LemonWei111 Have you solved the issue? I have the same problem now.