Pradeep

cc @GeorgiosSmyrnis

@GeorgiosSmyrnis was it intentional that you had to call blurring between these two transformations? https://github.com/rom1504/img2dataset/blob/fc3fb2eeb3b4b0c352f4dff622440b5c13100312/img2dataset/resizer.py#L182-L186

Thanks @GeorgiosSmyrnis, We can keep the blurring flow as is. w.r.t CROP, @rom1504 is of the opinion that it should happen before any resize happens and the bounding box dimensions...

Since the published metrics are mean execution times for an address, information of number of cores is probably irrelevant? However the actual type of CPU device will matter I guess?

I am not sure where the problem is, but the way this library is [looking to find opencv](https://github.com/peterbraden/node-opencv/blob/f0a03a4be7064ca4d75ff2f16f8cf206fe982456/utils/find-opencv.js#L23) is not working. opencv4 is what is present. looking at this script,...

I am also looking for information on any health endpoint that can be configured in the load balancer for health monitoring of the container/pod.

It is working in chrome after setting the Content Disposition **_response header_**.

@jparismorgan My images are stored in S3, so I just set the metadata so that content disposition response header is sent. I have set it to `Content-disposition: attachment` like suggested...

I don't have much technical insight to add, but I have seen such samples (everything black) being generated occassionally while doing transfer learning on Colab using below implementation. https://github.com/afiaka87/glide-finetune