Fix LetterResize bug with imrescale api

[Nioolek]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

·mmcv.imrescale· is used for keeping ratio resize. It may result in unequal input and output shape.
I changed mmcv.imrescale to mmcv.imresize. And optimize the code in transform.py.

Modification

Fix:
transform.py

dataset: COCO

Now, the LetterResize is used in test. So, we need to verify the test mAP results before and after the modification.
Use pretrained model yolov5s to test:
mAP results in README.md:

After:

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.377
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.571
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.410
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.217
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.425
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.488
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.310
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.516
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.572
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.384
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.629
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.720

[PeterH0323]

Hi @Nioolek

I saw that you are using train to eval it . But when we using coco dataset to train, the LetterResize is not been use, we need to use pre-trained model to run test.py to compare mAP both before and after refactor

[Nioolek]

Hi @Nioolek

I saw that you are using train to eval it . But when we using coco dataset to train, the LetterResize is not been use, we need to use pre-trained model to run test.py to compare mAP both before and after refactor

I used the method you mentioned, and updated the doc.