Testing not working

[lawrenceztang]

When the network is trained the error goes down, but during testing it says the average recall and precision is 0. Is there a reason for this? Perhaps I broke something messing with the code?

Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
2019-02-27 16:38:50,123 maskrcnn_benchmark.inference INFO: OrderedDict([('bbox', OrderedDict([('AP', 0.0), ('AP50', 0.0), ('AP75', 0.0), ('APs', 0.0), ('APm', 0.0), ('APl', 0.0)])), ('segm', OrderedDict([('AP', 0.0), ('AP50', 0.0), ('AP75', 0.0), ('APs', 0.0), ('APm', 0.0), ('APl', 0.0)]))])

Also, the function do_train in maskrcnn_benchmark/engine/trainer.py debugging in PyCharm breaks for me, I'm not sure why this is.

[fmassa]

@proptrot13 it might indicate that your model didn't learn anything maybe?
Or the codebase was broken with the changes, but hard to say only with the information you provided

[lawrenceztang]

I'm using this config: configs/e2e_mask_rcnn_R_50_FPN_1x.yaml. You're right, I just figured out that this config doesn't load trained weights. One other thing: how would I train mask-rcnn on a dataset that has masks labeled not using polygons, but by pixel? On the main page it says: "You can also add extra fields to the boxlist, such as segmentation masks (using structures.segmentation_mask.SegmentationMask), or even your own instance type." but I don't understand how to make the network train on this.

[fmassa]

@proptrot13 this is supported in #473 have a look there.

Closing this issue as the original question has already been addressed.