xuweidongkobe

ImportError: No module named 'imgaug.augmentables'; 'imgaug' is not a package

8

I am run the demo with boundingbox, but get the error, How can I solve the problem?

how to change code to get final_result

2

First, thank you for your work! The question is I tried to change write.py to get final_result in demo_inference.py, but it does not work. Can you give me some advice...

when I use mms , the url is like http:ip:port/my_name/predict. The question is how to change the "predict" to my own words. if I want url is http:ip:port/my_name/predictFile, how can...

Service will get stuck when predict large size images

1

My MMS version is 1.0.4. When I predict a large size image(about >6M), the service will get stuck. I check the doc, it limits the size of the input image....

How to handle invalid input!

1

Given a MMS custom service that expects JSON as input, If the input is invalid JSON such as { "image": "I missed a quote} MMS will return { "code": 500,...

readAddress(..) failed: Connection reset by peer

1

 My config file is shown in the above picture, the log file is shown in the following figure.  the error is : [ERROR] epollEventLoopGroup-3-8 com.amazonaws.ml.mms.http.HttpRequestHandler - io.netty.channel.unix.Errors$NativeIoException: readAddress(..)...

Using MaskRcnn InternImage for Object Detection without Segmentation

1

hi，I am using mask_rcnn_internimage_t_fpn_3x_coco.pth model for detection.The following config is for detection and segmentation How can I change the config file for detection only without segmentation, thank you !