For calibration, it’s not advised to use many images. From past experiece 64 images is a good size of dataset for calibration. The reason is, that during the calibration profcess the qunatizer collects max/min/zero-point data of each layer. Using large calibration set pushes the max and the min farther away, thus makeing more outlier nodes. This in turn force a large dynamic range that increases the quantization noise.
but i geting waring if i use the below 1024 images not in the accepted size @Nadav
1 Like
[warning] Reducing optimization level to 1 (the accuracy won’t be optimized and compression won’t be used) because there’s less data than the recommended amount (1024) i geting this waring
As said, 1024 is recommended, it will enable more optimization options.
I don’t think that is true. I get the following warnings:
[warning] Reducing optimization level to 0 (the accuracy won’t be optimized and compression won’t be used) because there’s less data than the recommended amount (1024), and there’s no available GPU
[warning] Running model optimization with zero level of optimization is not recommended for production use and might lead to suboptimal accuracy results
Why would the optimizer indicate this if less images are enough?
Is there a way to set the optimization level? The documentation at
does not mention an optimization level.
It’s a requirement, but not the sole one. You also need to have access to a supported GPU.
Yes, you you were just looking at the wrong place in the doc. this is the right link 