confused with coco map meaning and can't get good result

Prerequisites

Please answer the following question for yourself before submitting an issue.

• I checked to make sure that this issue has not been filed already.

1. The entire URL of the documentation with the issue

https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md#user-content-fn-1-389938ae008e4b49d16b1c9c0c403066

2. Describe the issue

I'm confused with the coco map metrics for "ssd_mobilenet_v1_coco" since coco has 12 evaluate metrics now, does it mean (AP) @IoU=0.50:0.95 or (AP) @IoU=0.50? Also, I have converted the "ssd_mobilenet_v1_coco" into tflite file without postprocess node, and then I decoded the two raw outputs by myself(all postprocess arguments were correctly taken from config file and the process was similar to "detection_postprocess.cc", but was written in python by myself). But the coco metrics I got seems not good. I'm not sure if it's my program bug, or using the default argument is not correct. How to reproduce the 21% result? Also, I have tried to generate a tflite file with the postprocess node, and I tried several inputs and found that the results of the 4 outputs are similar to what I have got by my python script.
This is my result below.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.184
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.307
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.192
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.017
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.157
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.372
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.180
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.242
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.242
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.026
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.206
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.491
In addition, I found that this issue is similar to https://github.com/tensorflow/models/issues/6636, and changing "nms_score_threshold" is not working(in fact, I have already noticed this while I'm reading sample config file and remarks of model zoo page, it helps increasing map actually, but 18.4% is the limitation, impossible to get 21% for me).