There is no evaluation during training in the Google Colab Notebook for detectron2 as can be seen here: https://colab.research.google.com/drive/16jcaJoc6bCFAQ96jDe2HwtXj7BMD_-m5. It does not use the validation data while training a custom dataset. How should I add it?
Reference Github repo - https://github.com/facebookresearch/detectron2
Colab notebooks are usually slow and meant for showing the basic usage of a repo. The fact that it is not evaluating during training could be just as simple as they don't consider it necessary having that step in a simple notebook. The more complex examples with periodic evaluation during training are in their repo.
However if you still want to evaluate in the notebook, I see that they create a train and val split here:
for d in ["train", "val"]:
DatasetCatalog.register("balloon_" + d, lambda d=d: get_balloon_dicts("balloon/" + d))
MetadataCatalog.get("balloon_" + d).set(thing_classes=["balloon"])
But is is not evaluating under training because of this line of code
cfg.DATASETS.TEST = ()
Try
cfg.DATASETS.TEST = ("balloon_val",)
instead and then set the hook for the trainer such that it suits your evaluation needs
Results obtained by setting eval_period to 50 and the custom evaluation to COCOEvaluator on balloon_val:
[07/14 07:23:52 d2.engine.train_loop]: Starting training from iteration 0
[07/14 07:24:02 d2.utils.events]: eta: 0:02:14 iter: 19 total_loss: 2.246 loss_cls: 0.7813 loss_box_reg: 0.6616 loss_mask: 0.683 loss_rpn_cls: 0.03956 loss_rpn_loc: 0.008304 time: 0.4848 data_time: 0.0323 lr: 1.6068e-05 max_mem: 5425M
[07/14 07:24:12 d2.utils.events]: eta: 0:02:01 iter: 39 total_loss: 1.879 loss_cls: 0.6221 loss_box_reg: 0.5713 loss_mask: 0.615 loss_rpn_cls: 0.04036 loss_rpn_loc: 0.01448 time: 0.4721 data_time: 0.0108 lr: 3.2718e-05 max_mem: 5425M
[07/14 07:24:17 d2.evaluation.evaluator]: Start inference on 13 batches
[07/14 07:24:26 d2.evaluation.evaluator]: Inference done 11/13. Dataloading: 0.0013 s/iter. Inference: 0.1507 s/iter. Eval: 0.1892 s/iter. Total: 0.3412 s/iter. ETA=0:00:00
[07/14 07:24:27 d2.evaluation.evaluator]: Total inference time: 0:00:02.799655 (0.349957 s / iter per device, on 1 devices)
[07/14 07:24:27 d2.evaluation.evaluator]: Total inference pure compute time: 0:00:01 (0.149321 s / iter per device, on 1 devices)
[07/14 07:24:27 d2.evaluation.coco_evaluation]: Preparing results for COCO format ...
[07/14 07:24:27 d2.evaluation.coco_evaluation]: Saving results to ./output/coco_instances_results.json
[07/14 07:24:27 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API...
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:24:27 d2.evaluation.fast_eval_api]: Evaluate annotation type *bbox*
[07/14 07:24:27 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:24:27 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:24:27 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.029
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.063
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.021
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.002
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.044
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.035
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 ] = 0.176
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.444
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.100
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.388
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.510
[07/14 07:24:27 d2.evaluation.coco_evaluation]: Evaluation results for bbox:
| AP | AP50 | AP75 | APs | APm | APl |
|:-----:|:------:|:------:|:-----:|:-----:|:-----:|
| 2.906 | 6.326 | 2.098 | 0.193 | 4.398 | 3.484 |
Loading and preparing results...
DONE (t=0.02s)
creating index...
index created!
[07/14 07:24:27 d2.evaluation.fast_eval_api]: Evaluate annotation type *segm*
[07/14 07:24:27 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.02 seconds.
[07/14 07:24:27 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:24:27 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.040
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.081
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.039
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.002
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.049
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.062
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.004
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.214
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.532
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.100
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.465
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.613
[07/14 07:24:27 d2.evaluation.coco_evaluation]: Evaluation results for segm:
| AP | AP50 | AP75 | APs | APm | APl |
|:-----:|:------:|:------:|:-----:|:-----:|:-----:|
| 4.027 | 8.132 | 3.905 | 0.166 | 4.904 | 6.221 |
[07/14 07:24:32 d2.utils.events]: eta: 0:01:56 iter: 59 total_loss: 1.621 loss_cls: 0.4834 loss_box_reg: 0.6684 loss_mask: 0.4703 loss_rpn_cls: 0.03119 loss_rpn_loc: 0.006103 time: 0.4799 data_time: 0.0117 lr: 4.9367e-05 max_mem: 5425M
[07/14 07:24:42 d2.utils.events]: eta: 0:01:47 iter: 79 total_loss: 1.401 loss_cls: 0.3847 loss_box_reg: 0.6159 loss_mask: 0.3641 loss_rpn_cls: 0.03303 loss_rpn_loc: 0.00822 time: 0.4797 data_time: 0.0130 lr: 6.6017e-05 max_mem: 5425M
[07/14 07:24:51 d2.utils.events]: eta: 0:01:36 iter: 99 total_loss: 1.268 loss_cls: 0.3295 loss_box_reg: 0.6366 loss_mask: 0.2884 loss_rpn_cls: 0.01753 loss_rpn_loc: 0.00765 time: 0.4775 data_time: 0.0096 lr: 8.2668e-05 max_mem: 5425M
[07/14 07:24:51 d2.evaluation.evaluator]: Start inference on 13 batches
[07/14 07:25:01 d2.evaluation.evaluator]: Inference done 11/13. Dataloading: 0.0014 s/iter. Inference: 0.1493 s/iter. Eval: 0.1851 s/iter. Total: 0.3358 s/iter. ETA=0:00:00
[07/14 07:25:01 d2.evaluation.evaluator]: Total inference time: 0:00:02.778349 (0.347294 s / iter per device, on 1 devices)
[07/14 07:25:01 d2.evaluation.evaluator]: Total inference pure compute time: 0:00:01 (0.148906 s / iter per device, on 1 devices)
[07/14 07:25:02 d2.evaluation.coco_evaluation]: Preparing results for COCO format ...
[07/14 07:25:02 d2.evaluation.coco_evaluation]: Saving results to ./output/coco_instances_results.json
[07/14 07:25:02 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API...
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:25:02 d2.evaluation.fast_eval_api]: Evaluate annotation type *bbox*
[07/14 07:25:02 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:25:02 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:25:02 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.01 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.543
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.751
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.626
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.092
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.472
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.636
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.196
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.620
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.714
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.533
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.588
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.803
[07/14 07:25:02 d2.evaluation.coco_evaluation]: Evaluation results for bbox:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:-----:|:------:|:------:|
| 54.340 | 75.066 | 62.622 | 9.181 | 47.208 | 63.594 |
Loading and preparing results...
DONE (t=0.02s)
creating index...
index created!
[07/14 07:25:02 d2.evaluation.fast_eval_api]: Evaluate annotation type *segm*
[07/14 07:25:02 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.02 seconds.
[07/14 07:25:02 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:25:02 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.01 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.630
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.754
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.741
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.060
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.519
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.750
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.214
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.692
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.786
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.533
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.641
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.893
[07/14 07:25:02 d2.evaluation.coco_evaluation]: Evaluation results for segm:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:-----:|:------:|:------:|
| 62.959 | 75.390 | 74.088 | 5.987 | 51.899 | 74.988 |
[07/14 07:25:11 d2.utils.events]: eta: 0:01:26 iter: 119 total_loss: 1.158 loss_cls: 0.2745 loss_box_reg: 0.6951 loss_mask: 0.2165 loss_rpn_cls: 0.02461 loss_rpn_loc: 0.00421 time: 0.4773 data_time: 0.0101 lr: 9.9318e-05 max_mem: 5425M
[07/14 07:25:21 d2.utils.events]: eta: 0:01:16 iter: 139 total_loss: 1.015 loss_cls: 0.1891 loss_box_reg: 0.6029 loss_mask: 0.1745 loss_rpn_cls: 0.02219 loss_rpn_loc: 0.005621 time: 0.4766 data_time: 0.0111 lr: 0.00011597 max_mem: 5425M
[07/14 07:25:26 d2.evaluation.evaluator]: Start inference on 13 batches
[07/14 07:25:34 d2.evaluation.evaluator]: Inference done 11/13. Dataloading: 0.0013 s/iter. Inference: 0.1459 s/iter. Eval: 0.1786 s/iter. Total: 0.3258 s/iter. ETA=0:00:00
[07/14 07:25:35 d2.evaluation.evaluator]: Total inference time: 0:00:02.608437 (0.326055 s / iter per device, on 1 devices)
[07/14 07:25:35 d2.evaluation.evaluator]: Total inference pure compute time: 0:00:01 (0.143658 s / iter per device, on 1 devices)
[07/14 07:25:35 d2.evaluation.coco_evaluation]: Preparing results for COCO format ...
[07/14 07:25:35 d2.evaluation.coco_evaluation]: Saving results to ./output/coco_instances_results.json
[07/14 07:25:35 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API...
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:25:35 d2.evaluation.fast_eval_api]: Evaluate annotation type *bbox*
[07/14 07:25:35 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:25:35 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:25:35 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.663
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.843
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.754
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.245
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.562
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.790
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.228
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.712
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.758
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.567
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.647
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.840
[07/14 07:25:35 d2.evaluation.coco_evaluation]: Evaluation results for bbox:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 66.307 | 84.257 | 75.431 | 24.466 | 56.175 | 79.035 |
Loading and preparing results...
DONE (t=0.01s)
creating index...
index created!
[07/14 07:25:35 d2.evaluation.fast_eval_api]: Evaluate annotation type *segm*
[07/14 07:25:35 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.02 seconds.
[07/14 07:25:35 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:25:35 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.756
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.839
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.833
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.135
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.581
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.915
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.248
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.788
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.836
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.600
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.676
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.950
[07/14 07:25:35 d2.evaluation.coco_evaluation]: Evaluation results for segm:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 75.579 | 83.916 | 83.342 | 13.466 | 58.113 | 91.479 |
[07/14 07:25:40 d2.utils.events]: eta: 0:01:07 iter: 159 total_loss: 0.845 loss_cls: 0.1613 loss_box_reg: 0.5442 loss_mask: 0.1211 loss_rpn_cls: 0.01358 loss_rpn_loc: 0.006381 time: 0.4768 data_time: 0.0110 lr: 0.00013262 max_mem: 5425M
[07/14 07:25:49 d2.utils.events]: eta: 0:00:58 iter: 179 total_loss: 0.7381 loss_cls: 0.1207 loss_box_reg: 0.4569 loss_mask: 0.1153 loss_rpn_cls: 0.01103 loss_rpn_loc: 0.005893 time: 0.4782 data_time: 0.0098 lr: 0.00014927 max_mem: 5425M
[07/14 07:25:59 d2.utils.events]: eta: 0:00:48 iter: 199 total_loss: 0.5811 loss_cls: 0.108 loss_box_reg: 0.3294 loss_mask: 0.09868 loss_rpn_cls: 0.01414 loss_rpn_loc: 0.008676 time: 0.4783 data_time: 0.0101 lr: 0.00016592 max_mem: 5425M
[07/14 07:25:59 d2.evaluation.evaluator]: Start inference on 13 batches
[07/14 07:26:05 d2.evaluation.evaluator]: Inference done 11/13. Dataloading: 0.0017 s/iter. Inference: 0.1317 s/iter. Eval: 0.0985 s/iter. Total: 0.2319 s/iter. ETA=0:00:00
[07/14 07:26:05 d2.evaluation.evaluator]: Total inference time: 0:00:01.788219 (0.223527 s / iter per device, on 1 devices)
[07/14 07:26:05 d2.evaluation.evaluator]: Total inference pure compute time: 0:00:01 (0.127455 s / iter per device, on 1 devices)
[07/14 07:26:05 d2.evaluation.coco_evaluation]: Preparing results for COCO format ...
[07/14 07:26:05 d2.evaluation.coco_evaluation]: Saving results to ./output/coco_instances_results.json
[07/14 07:26:05 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API...
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:26:05 d2.evaluation.fast_eval_api]: Evaluate annotation type *bbox*
[07/14 07:26:05 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:26:05 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:26:05 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.01 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.728
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.894
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.858
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.303
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.571
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.848
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.218
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.742
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.790
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.500
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.688
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.877
[07/14 07:26:05 d2.evaluation.coco_evaluation]: Evaluation results for bbox:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 72.797 | 89.384 | 85.752 | 30.301 | 57.057 | 84.812 |
Loading and preparing results...
DONE (t=0.01s)
creating index...
index created!
[07/14 07:26:05 d2.evaluation.fast_eval_api]: Evaluate annotation type *segm*
[07/14 07:26:05 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:26:05 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:26:05 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.805
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.885
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.880
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.252
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.617
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.950
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.250
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.808
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.860
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.567
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.735
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.960
[07/14 07:26:05 d2.evaluation.coco_evaluation]: Evaluation results for segm:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 80.490 | 88.547 | 87.953 | 25.206 | 61.723 | 94.959 |
[07/14 07:26:15 d2.utils.events]: eta: 0:00:38 iter: 219 total_loss: 0.4771 loss_cls: 0.08176 loss_box_reg: 0.2226 loss_mask: 0.09229 loss_rpn_cls: 0.01647 loss_rpn_loc: 0.009867 time: 0.4789 data_time: 0.0132 lr: 0.00018257 max_mem: 5425M
[07/14 07:26:25 d2.utils.events]: eta: 0:00:28 iter: 239 total_loss: 0.366 loss_cls: 0.07189 loss_box_reg: 0.1961 loss_mask: 0.08049 loss_rpn_cls: 0.01413 loss_rpn_loc: 0.006811 time: 0.4785 data_time: 0.0122 lr: 0.00019922 max_mem: 5425M
[07/14 07:26:29 d2.evaluation.evaluator]: Start inference on 13 batches
[07/14 07:26:34 d2.evaluation.evaluator]: Inference done 11/13. Dataloading: 0.0015 s/iter. Inference: 0.1195 s/iter. Eval: 0.0502 s/iter. Total: 0.1711 s/iter. ETA=0:00:00
[07/14 07:26:34 d2.evaluation.evaluator]: Total inference time: 0:00:01.375643 (0.171955 s / iter per device, on 1 devices)
[07/14 07:26:34 d2.evaluation.evaluator]: Total inference pure compute time: 0:00:00 (0.117491 s / iter per device, on 1 devices)
[07/14 07:26:34 d2.evaluation.coco_evaluation]: Preparing results for COCO format ...
[07/14 07:26:34 d2.evaluation.coco_evaluation]: Saving results to ./output/coco_instances_results.json
[07/14 07:26:34 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API...
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:26:34 d2.evaluation.fast_eval_api]: Evaluate annotation type *bbox*
[07/14 07:26:34 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:26:34 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:26:34 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.779
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.916
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.878
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.350
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.615
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.896
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.234
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.800
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.826
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.467
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.718
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.923
[07/14 07:26:34 d2.evaluation.coco_evaluation]: Evaluation results for bbox:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 77.888 | 91.606 | 87.774 | 34.965 | 61.497 | 89.576 |
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:26:34 d2.evaluation.fast_eval_api]: Evaluate annotation type *segm*
[07/14 07:26:34 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:26:34 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:26:34 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.823
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.894
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.891
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.248
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.624
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.967
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.254
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.832
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.858
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.367
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.741
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.973
[07/14 07:26:34 d2.evaluation.coco_evaluation]: Evaluation results for segm:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 82.323 | 89.379 | 89.068 | 24.752 | 62.427 | 96.691 |
[07/14 07:26:39 d2.utils.events]: eta: 0:00:19 iter: 259 total_loss: 0.2651 loss_cls: 0.05436 loss_box_reg: 0.1442 loss_mask: 0.06249 loss_rpn_cls: 0.005261 loss_rpn_loc: 0.00489 time: 0.4781 data_time: 0.0123 lr: 0.00021587 max_mem: 5425M
[07/14 07:26:49 d2.utils.events]: eta: 0:00:09 iter: 279 total_loss: 0.4224 loss_cls: 0.07591 loss_box_reg: 0.1941 loss_mask: 0.09489 loss_rpn_cls: 0.009817 loss_rpn_loc: 0.008633 time: 0.4777 data_time: 0.0109 lr: 0.00023252 max_mem: 5425M
[07/14 07:26:59 d2.utils.events]: eta: 0:00:00 iter: 299 total_loss: 0.3534 loss_cls: 0.07829 loss_box_reg: 0.1646 loss_mask: 0.08058 loss_rpn_cls: 0.01157 loss_rpn_loc: 0.006635 time: 0.4779 data_time: 0.0120 lr: 0.00024917 max_mem: 5425M
[07/14 07:27:00 d2.engine.hooks]: Overall training speed: 298 iterations in 0:02:22 (0.4779 s / it)
[07/14 07:27:00 d2.engine.hooks]: Total training time: 0:03:06 (0:00:43 on hooks)
[07/14 07:27:00 d2.evaluation.evaluator]: Start inference on 13 batches
[07/14 07:27:04 d2.evaluation.evaluator]: Inference done 11/13. Dataloading: 0.0015 s/iter. Inference: 0.1155 s/iter. Eval: 0.0340 s/iter. Total: 0.1510 s/iter. ETA=0:00:00
[07/14 07:27:04 d2.evaluation.evaluator]: Total inference time: 0:00:01.238510 (0.154814 s / iter per device, on 1 devices)
[07/14 07:27:04 d2.evaluation.evaluator]: Total inference pure compute time: 0:00:00 (0.114618 s / iter per device, on 1 devices)
[07/14 07:27:04 d2.evaluation.coco_evaluation]: Preparing results for COCO format ...
[07/14 07:27:04 d2.evaluation.coco_evaluation]: Saving results to ./output/coco_instances_results.json
[07/14 07:27:04 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API...
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:27:04 d2.evaluation.fast_eval_api]: Evaluate annotation type *bbox*
[07/14 07:27:04 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:27:04 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:27:04 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.762
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.927
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.859
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.310
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.640
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.864
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.236
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.788
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.814
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.433
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.724
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.903
[07/14 07:27:04 d2.evaluation.coco_evaluation]: Evaluation results for bbox:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 76.245 | 92.732 | 85.874 | 31.015 | 63.981 | 86.418 |
Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
[07/14 07:27:04 d2.evaluation.fast_eval_api]: Evaluate annotation type *segm*
[07/14 07:27:04 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 0.01 seconds.
[07/14 07:27:04 d2.evaluation.fast_eval_api]: Accumulating evaluation results...
[07/14 07:27:04 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 0.00 seconds.
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.818
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.902
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.899
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.253
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.632
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.956
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.252
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.828
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.856
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.400
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.735
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.970
[07/14 07:27:04 d2.evaluation.coco_evaluation]: Evaluation results for segm:
| AP | AP50 | AP75 | APs | APm | APl |
|:------:|:------:|:------:|:------:|:------:|:------:|
| 81.780 | 90.213 | 89.900 | 25.284 | 63.179 | 95.585 |
