inferece with new trained model

[mohamedettebayo]

i trained swin model on my custom data sets
but when iam trying know to inferece i get an error
can any one help me please

from mmcv import Config
cfg = Config.fromfile("/content/mmdetection/configs/swin/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco.py")  ## my config used for instance segmentation

## download weight
!wget https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco_20210906_131725-bacf6f7b.pth

from mmdet.apis import set_random_seed
cfg.device='cuda' # 

# Modify dataset type and path
cfg.dataset_type = 'COCODataset'

cfg.data.test.ann_file = '/content/mmdetection/data/coco/valid/_annotations.coco.json'
cfg.data.test.img_prefix = '/content/mmdetection/data/coco/train/'
cfg.data.test.classes = ('1','10','11','12','13','14','15','16','17','18','19','2','20','21','22','23','24','25','26','27','28','29','3','30','31','32','4','5','6','7','8','9')

cfg.data.train.ann_file = '/content/mmdetection/data/coco/train/_annotations.coco.json'
cfg.data.train.img_prefix = '/content/mmdetection/data/coco/train/'
cfg.data.train.classes = ('1','10','11','12','13','14','15','16','17','18','19','2','20','21','22','23','24','25','26','27','28','29','3','30','31','32','4','5','6','7','8','9')


cfg.data.val.ann_file = '/content/mmdetection/data/coco/valid/_annotations.coco.json'
cfg.data.val.img_prefix = '/content/mmdetection/data/coco/valid/'
cfg.data.val.classes = ('1','10','11','12','13','14','15','16','17','18','19','2','20','21','22','23','24','25','26','27','28','29','3','30','31','32','4','5','6','7','8','9')

#cfg.data.train.seg_prefix = '/content/mmdetection/data/coco/train/'
# modify num classes of the model in box head and mask head
cfg.model.roi_head.bbox_head.num_classes = 32
cfg.model.roi_head.mask_head.num_classes = 32
cfg.runner.max_epochs = 5

# We can still the pre-trained Mask RCNN model to obtain a higher performance
cfg.load_from = '/content/mmdetection/checkpoints/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco_20210906_131725-bacf6f7b.pth'
#cfg.load_from = '/content/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco_20210906_131725-bacf6f7b.pth'

# Set up working dir to save files and logs.
cfg.work_dir = './tutorial_exps'

# The original learning rate (LR) is set for 8-GPU training.
# We divide it by 8 since we only use one GPU.
#cfg.optimizer.lr = 0.0001
#cfg.lr_config.warmup = None
#cfg.log_config.interval = 10
cfg.device='cuda' # 
# We can set the evaluation interval to reduce the evaluation times
#cfg.evaluation.interval = 12
# We can set the checkpoint saving interval to reduce the storage cost
#cfg.checkpoint_config.interval = 12

# Set seed thus the results are more reproducible
cfg.seed = 0
set_random_seed(0, deterministic=False)
cfg.gpu_ids = range(1)

# We can also use tensorboard to log the training process
cfg.log_config.hooks = [
    dict(type='TextLoggerHook'),
    dict(type='TensorboardLoggerHook')]

# We can initialize the logger for training and have a look
# at the final config used for training
print(f'Config:\n{cfg.pretty_text}')

import mmcv ##
from mmdet.datasets import build_dataset
from mmdet.models import build_detector
from mmdet.apis import train_detector




# Build dataset
datasets = [build_dataset(cfg.data.train)]

# Build the detector
model = build_detector(cfg.model)

# Add an attribute for visualization convenience
model.CLASSES = datasets[0].CLASSES

import os.path as osp
# Create work_dir
mmcv.mkdir_or_exist(osp.abspath(cfg.work_dir))
train_detector(model, datasets, cfg, distributed=False, validate=True,)

the output for training was

loading annotations into memory...
Done (t=0.48s)
creating index...
index created!

2022-10-21 08:56:49,601 - mmdet - INFO - Automatic scaling of learning rate (LR) has been disabled.
2022-10-21 08:56:49,636 - mmdet - INFO - load checkpoint from local path: /content/mmdetection/checkpoints/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco_20210906_131725-bacf6f7b.pth
2022-10-21 08:56:49,776 - mmdet - WARNING - The model and loaded state dict do not match exactly

size mismatch for roi_head.bbox_head.fc_cls.weight: copying a param with shape torch.Size([81, 1024]) from checkpoint, the shape in current model is torch.Size([33, 1024]).
size mismatch for roi_head.bbox_head.fc_cls.bias: copying a param with shape torch.Size([81]) from checkpoint, the shape in current model is torch.Size([33]).
size mismatch for roi_head.bbox_head.fc_reg.weight: copying a param with shape torch.Size([320, 1024]) from checkpoint, the shape in current model is torch.Size([128, 1024]).
size mismatch for roi_head.bbox_head.fc_reg.bias: copying a param with shape torch.Size([320]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for roi_head.mask_head.conv_logits.weight: copying a param with shape torch.Size([80, 256, 1, 1]) from checkpoint, the shape in current model is torch.Size([32, 256, 1, 1]).
size mismatch for roi_head.mask_head.conv_logits.bias: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([32]).
2022-10-21 08:56:49,781 - mmdet - INFO - Start running, host: root@4e523559989c, work_dir: /content/mmdetection/tutorial_exps
2022-10-21 08:56:49,784 - mmdet - INFO - Hooks will be executed in the following order:
before_run:
(VERY_HIGH ) StepLrUpdaterHook
(NORMAL ) CheckpointHook
(LOW ) EvalHook
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

before_train_epoch:
(VERY_HIGH ) StepLrUpdaterHook
(NORMAL ) NumClassCheckHook
(LOW ) IterTimerHook
(LOW ) EvalHook
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

before_train_iter:
(VERY_HIGH ) StepLrUpdaterHook
(LOW ) IterTimerHook
(LOW ) EvalHook

after_train_iter:
(ABOVE_NORMAL) OptimizerHook
(NORMAL ) CheckpointHook
(LOW ) IterTimerHook
(LOW ) EvalHook
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

after_train_epoch:
(NORMAL ) CheckpointHook
(LOW ) EvalHook
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

before_val_epoch:
(NORMAL ) NumClassCheckHook
(LOW ) IterTimerHook
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

before_val_iter:
(LOW ) IterTimerHook

after_val_iter:
(LOW ) IterTimerHook

after_val_epoch:
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

after_run:
(VERY_LOW ) TextLoggerHook
(VERY_LOW ) TensorboardLoggerHook

2022-10-21 08:56:49,788 - mmdet - INFO - workflow: [('train', 1)], max: 5 epochs
2022-10-21 08:56:49,792 - mmdet - INFO - Checkpoints will be saved to /content/mmdetection/tutorial_exps by HardDiskBackend.

loading annotations into memory...
Done (t=0.01s)
creating index...
index created!

/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at /pytorch/c10/core/TensorImpl.h:1156.)
return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)
2022-10-21 08:58:04,242 - mmdet - INFO - Epoch [1][50/540] lr: 4.995e-06, eta: 0:59:35, time: 1.349, data_time: 0.344, memory: 8677, loss_rpn_cls: 0.6669, loss_rpn_bbox: 0.1717, loss_cls: 3.0148, acc: 27.3672, loss_bbox: 0.9368, loss_mask: 0.7332, loss: 5.5234
2022-10-21 08:59:10,645 - mmdet - INFO - Epoch [1][100/540] lr: 9.990e-06, eta: 0:58:00, time: 1.328, data_time: 0.316, memory: 9840, loss_rpn_cls: 0.1002, loss_rpn_bbox: 0.1514, loss_cls: 1.9519, acc: 72.4219, loss_bbox: 1.0109, loss_mask: 0.7128, loss: 3.9272
2022-10-21 09:00:12,017 - mmdet - INFO - Epoch [1][150/540] lr: 1.499e-05, eta: 0:55:18, time: 1.227, data_time: 0.284, memory: 9840, loss_rpn_cls: 0.0599, loss_rpn_bbox: 0.1367, loss_cls: 1.3584, acc: 75.0645, loss_bbox: 0.9673, loss_mask: 0.6793, loss: 3.2017
2022-10-21 09:01:14,280 - mmdet - INFO - Epoch [1][200/540] lr: 1.998e-05, eta: 0:53:38, time: 1.245, data_time: 0.281, memory: 9840, loss_rpn_cls: 0.0461, loss_rpn_bbox: 0.1212, loss_cls: 1.2270, acc: 75.1992, loss_bbox: 0.8932, loss_mask: 0.5910, loss: 2.8785
2022-10-21 09:02:19,980 - mmdet - INFO - Epoch [1][250/540] lr: 2.498e-05, eta: 0:52:47, time: 1.314, data_time: 0.322, memory: 9840, loss_rpn_cls: 0.0240, loss_rpn_bbox: 0.1057, loss_cls: 1.1130, acc: 75.4062, loss_bbox: 0.8415, loss_mask: 0.3913, loss: 2.4755
2022-10-21 09:03:25,705 - mmdet - INFO - Epoch [1][300/540] lr: 2.997e-05, eta: 0:51:51, time: 1.315, data_time: 0.318, memory: 9840, loss_rpn_cls: 0.0242, loss_rpn_bbox: 0.1040, loss_cls: 1.0232, acc: 75.3164, loss_bbox: 0.7928, loss_mask: 0.2811, loss: 2.2252
2022-10-21 09:04:31,790 - mmdet - INFO - Epoch [1][350/540] lr: 3.497e-05, eta: 0:50:54, time: 1.322, data_time: 0.346, memory: 9840, loss_rpn_cls: 0.0208, loss_rpn_bbox: 0.1065, loss_cls: 0.9005, acc: 75.8594, loss_bbox: 0.7110, loss_mask: 0.2523, loss: 1.9911
2022-10-21 09:05:33,719 - mmdet - INFO - Epoch [1][400/540] lr: 3.996e-05, eta: 0:49:32, time: 1.239, data_time: 0.291, memory: 9840, loss_rpn_cls: 0.0169, loss_rpn_bbox: 0.0918, loss_cls: 0.7829, acc: 77.2168, loss_bbox: 0.6300, loss_mask: 0.2326, loss: 1.7541
2022-10-21 09:06:35,814 - mmdet - INFO - Epoch [1][450/540] lr: 4.496e-05, eta: 0:48:15, time: 1.242, data_time: 0.284, memory: 9840, loss_rpn_cls: 0.0262, loss_rpn_bbox: 0.0969, loss_cls: 0.7101, acc: 77.6543, loss_bbox: 0.5522, loss_mask: 0.2257, loss: 1.6111
2022-10-21 09:07:40,754 - mmdet - INFO - Epoch [1][500/540] lr: 4.995e-05, eta: 0:47:13, time: 1.299, data_time: 0.329, memory: 9840, loss_rpn_cls: 0.0169, loss_rpn_bbox: 0.0983, loss_cls: 0.6556, acc: 78.5898, loss_bbox: 0.5152, loss_mask: 0.2176, loss: 1.5035
2022-10-21 09:08:29,975 - mmdet - INFO - Saving checkpoint at 1 epochs

[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 39/39, 0.8 task/s, elapsed: 51s, ETA: 0s

2022-10-21 09:09:26,016 - mmdet - INFO - Evaluating bbox...

Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type bbox
DONE (t=0.55s).
Accumulating evaluation results...

2022-10-21 09:09:26,822 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.250
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.430
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.263
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.154
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.259
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.540
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.540
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.540
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.205
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.562

2022-10-21 09:09:26,823 - mmdet - INFO - Evaluating segm...

DONE (t=0.23s).
Loading and preparing results...
DONE (t=0.04s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type segm

/content/mmdetection/mmdet/datasets/coco.py:474: UserWarning: The key "bbox" is deleted for more accurate mask AP of small/medium/large instances since v2.12.0. This does not change the overall mAP calculation.
UserWarning)

DONE (t=0.64s).
Accumulating evaluation results...

/usr/local/lib/python3.7/dist-packages/pycocotools/cocoeval.py:378: DeprecationWarning: np.float is a deprecated alias for the builtin float. To silence this warning, use float by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use np.float64 here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
tp_sum = np.cumsum(tps, axis=1).astype(dtype=np.float)
2022-10-21 09:09:27,804 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.283
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.430
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.353
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.088
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.319
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.587
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.587
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.587
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.275
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.611

2022-10-21 09:09:27,835 - mmdet - INFO - Epoch(val) [1][39] bbox_mAP: 0.2500, bbox_mAP_50: 0.4300, bbox_mAP_75: 0.2630, bbox_mAP_s: 0.0000, bbox_mAP_m: 0.1540, bbox_mAP_l: 0.2590, bbox_mAP_copypaste: 0.250 0.430 0.263 0.000 0.154 0.259, segm_mAP: 0.2830, segm_mAP_50: 0.4300, segm_mAP_75: 0.3530, segm_mAP_s: 0.0000, segm_mAP_m: 0.0880, segm_mAP_l: 0.3190, segm_mAP_copypaste: 0.283 0.430 0.353 0.000 0.088 0.319

DONE (t=0.24s).

2022-10-21 09:11:05,973 - mmdet - INFO - Epoch [2][50/540] lr: 5.894e-05, eta: 0:44:13, time: 1.962, data_time: 1.012, memory: 9840, loss_rpn_cls: 0.0142, loss_rpn_bbox: 0.0926, loss_cls: 0.6041, acc: 79.4961, loss_bbox: 0.4770, loss_mask: 0.2023, loss: 1.3903
2022-10-21 09:12:56,280 - mmdet - INFO - Epoch [2][100/540] lr: 6.394e-05, eta: 0:45:43, time: 2.206, data_time: 1.199, memory: 9840, loss_rpn_cls: 0.0176, loss_rpn_bbox: 0.0953, loss_cls: 0.5868, acc: 79.8926, loss_bbox: 0.4672, loss_mask: 0.2134, loss: 1.3803
2022-10-21 09:14:43,849 - mmdet - INFO - Epoch [2][150/540] lr: 6.893e-05, eta: 0:46:36, time: 2.151, data_time: 1.190, memory: 9840, loss_rpn_cls: 0.0182, loss_rpn_bbox: 0.0939, loss_cls: 0.5647, acc: 80.0039, loss_bbox: 0.4384, loss_mask: 0.2002, loss: 1.3154
2022-10-21 09:16:30,017 - mmdet - INFO - Epoch [2][200/540] lr: 7.393e-05, eta: 0:47:03, time: 2.123, data_time: 1.147, memory: 9840, loss_rpn_cls: 0.0149, loss_rpn_bbox: 0.0904, loss_cls: 0.5356, acc: 80.7910, loss_bbox: 0.4225, loss_mask: 0.1940, loss: 1.2574
2022-10-21 09:18:36,550 - mmdet - INFO - Epoch [2][250/540] lr: 7.892e-05, eta: 0:48:03, time: 2.531, data_time: 1.525, memory: 9840, loss_rpn_cls: 0.0101, loss_rpn_bbox: 0.0878, loss_cls: 0.5248, acc: 80.8613, loss_bbox: 0.4209, loss_mask: 0.2006, loss: 1.2442
2022-10-21 09:20:43,865 - mmdet - INFO - Epoch [2][300/540] lr: 8.392e-05, eta: 0:48:42, time: 2.546, data_time: 1.531, memory: 9840, loss_rpn_cls: 0.0120, loss_rpn_bbox: 0.0880, loss_cls: 0.4986, acc: 81.7051, loss_bbox: 0.4120, loss_mask: 0.1924, loss: 1.2030
2022-10-21 09:22:34,575 - mmdet - INFO - Epoch [2][350/540] lr: 8.891e-05, eta: 0:48:29, time: 2.214, data_time: 1.238, memory: 9840, loss_rpn_cls: 0.0127, loss_rpn_bbox: 0.0818, loss_cls: 0.4755, acc: 82.0840, loss_bbox: 0.3878, loss_mask: 0.1803, loss: 1.1382
2022-10-21 09:24:12,014 - mmdet - INFO - Epoch [2][400/540] lr: 9.391e-05, eta: 0:47:41, time: 1.949, data_time: 0.990, memory: 9840, loss_rpn_cls: 0.0133, loss_rpn_bbox: 0.0834, loss_cls: 0.4863, acc: 81.9688, loss_bbox: 0.3905, loss_mask: 0.1821, loss: 1.1557
2022-10-21 09:26:08,715 - mmdet - INFO - Epoch [2][450/540] lr: 9.890e-05, eta: 0:47:20, time: 2.334, data_time: 1.361, memory: 9840, loss_rpn_cls: 0.0129, loss_rpn_bbox: 0.0846, loss_cls: 0.4650, acc: 82.0371, loss_bbox: 0.3805, loss_mask: 0.1817, loss: 1.1246
2022-10-21 09:28:14,532 - mmdet - INFO - Epoch [2][500/540] lr: 1.000e-04, eta: 0:47:06, time: 2.516, data_time: 1.523, memory: 9840, loss_rpn_cls: 0.0169, loss_rpn_bbox: 0.0908, loss_cls: 0.4758, acc: 81.7910, loss_bbox: 0.3836, loss_mask: 0.1871, loss: 1.1541
2022-10-21 09:29:37,383 - mmdet - INFO - Saving checkpoint at 2 epochs

[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 39/39, 0.2 task/s, elapsed: 158s, ETA: 0s

2022-10-21 09:32:20,428 - mmdet - INFO - Evaluating bbox...

Loading and preparing results...
DONE (t=0.01s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type bbox
DONE (t=0.52s).
Accumulating evaluation results...

2022-10-21 09:32:21,219 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.405
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.618
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.503
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.248
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.420
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.684
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.684
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.684
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.256
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.713

2022-10-21 09:32:21,220 - mmdet - INFO - Evaluating segm...

DONE (t=0.24s).

/content/mmdetection/mmdet/datasets/coco.py:474: UserWarning: The key "bbox" is deleted for more accurate mask AP of small/medium/large instances since v2.12.0. This does not change the overall mAP calculation.
UserWarning)

Loading and preparing results...
DONE (t=0.04s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type segm
DONE (t=0.58s).
Accumulating evaluation results...

/usr/local/lib/python3.7/dist-packages/pycocotools/cocoeval.py:378: DeprecationWarning: np.float is a deprecated alias for the builtin float. To silence this warning, use float by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use np.float64 here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
tp_sum = np.cumsum(tps, axis=1).astype(dtype=np.float)
2022-10-21 09:32:22,558 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.442
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.619
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.553
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.062
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.485
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.727
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.727
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.727
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.314
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.756

2022-10-21 09:32:22,570 - mmdet - INFO - Epoch(val) [2][39] bbox_mAP: 0.4050, bbox_mAP_50: 0.6180, bbox_mAP_75: 0.5030, bbox_mAP_s: 0.0000, bbox_mAP_m: 0.2480, bbox_mAP_l: 0.4200, bbox_mAP_copypaste: 0.405 0.618 0.503 0.000 0.248 0.420, segm_mAP: 0.4420, segm_mAP_50: 0.6190, segm_mAP_75: 0.5530, segm_mAP_s: 0.0000, segm_mAP_m: 0.0620, segm_mAP_l: 0.4850, segm_mAP_copypaste: 0.442 0.619 0.553 0.000 0.062 0.485

DONE (t=0.24s).

2022-10-21 09:33:27,848 - mmdet - INFO - Epoch [3][50/540] lr: 1.000e-04, eta: 0:42:30, time: 1.305, data_time: 0.345, memory: 9840, loss_rpn_cls: 0.0103, loss_rpn_bbox: 0.0875, loss_cls: 0.4443, acc: 82.8320, loss_bbox: 0.3652, loss_mask: 0.1728, loss: 1.0801
2022-10-21 09:34:29,910 - mmdet - INFO - Epoch [3][100/540] lr: 1.000e-04, eta: 0:40:44, time: 1.241, data_time: 0.253, memory: 9868, loss_rpn_cls: 0.0128, loss_rpn_bbox: 0.0846, loss_cls: 0.4569, acc: 82.2129, loss_bbox: 0.3679, loss_mask: 0.1851, loss: 1.1072
2022-10-21 09:35:28,615 - mmdet - INFO - Epoch [3][150/540] lr: 1.000e-04, eta: 0:38:58, time: 1.174, data_time: 0.241, memory: 9868, loss_rpn_cls: 0.0123, loss_rpn_bbox: 0.0801, loss_cls: 0.4343, acc: 82.6992, loss_bbox: 0.3523, loss_mask: 0.1746, loss: 1.0536
2022-10-21 09:36:30,567 - mmdet - INFO - Epoch [3][200/540] lr: 1.000e-04, eta: 0:37:19, time: 1.239, data_time: 0.269, memory: 9868, loss_rpn_cls: 0.0103, loss_rpn_bbox: 0.0837, loss_cls: 0.4286, acc: 83.0215, loss_bbox: 0.3625, loss_mask: 0.1826, loss: 1.0678
2022-10-21 09:37:32,214 - mmdet - INFO - Epoch [3][250/540] lr: 1.000e-04, eta: 0:35:42, time: 1.233, data_time: 0.248, memory: 9868, loss_rpn_cls: 0.0067, loss_rpn_bbox: 0.0782, loss_cls: 0.4401, acc: 82.4102, loss_bbox: 0.3513, loss_mask: 0.1699, loss: 1.0461
2022-10-21 09:38:36,074 - mmdet - INFO - Epoch [3][300/540] lr: 1.000e-04, eta: 0:34:10, time: 1.277, data_time: 0.281, memory: 9868, loss_rpn_cls: 0.0095, loss_rpn_bbox: 0.0843, loss_cls: 0.4147, acc: 83.6055, loss_bbox: 0.3449, loss_mask: 0.1732, loss: 1.0265
2022-10-21 09:39:39,328 - mmdet - INFO - Epoch [3][350/540] lr: 1.000e-04, eta: 0:32:40, time: 1.265, data_time: 0.289, memory: 9868, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0874, loss_cls: 0.4185, acc: 83.3770, loss_bbox: 0.3535, loss_mask: 0.1764, loss: 1.0460
2022-10-21 09:40:38,513 - mmdet - INFO - Epoch [3][400/540] lr: 1.000e-04, eta: 0:31:08, time: 1.184, data_time: 0.237, memory: 9868, loss_rpn_cls: 0.0145, loss_rpn_bbox: 0.0814, loss_cls: 0.3938, acc: 84.2070, loss_bbox: 0.3486, loss_mask: 0.1767, loss: 1.0151
2022-10-21 09:41:38,928 - mmdet - INFO - Epoch [3][450/540] lr: 1.000e-04, eta: 0:29:39, time: 1.208, data_time: 0.242, memory: 9868, loss_rpn_cls: 0.0105, loss_rpn_bbox: 0.0908, loss_cls: 0.4349, acc: 82.8066, loss_bbox: 0.3627, loss_mask: 0.1989, loss: 1.0978
2022-10-21 09:42:40,873 - mmdet - INFO - Epoch [3][500/540] lr: 1.000e-04, eta: 0:28:13, time: 1.239, data_time: 0.260, memory: 9868, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0828, loss_cls: 0.4099, acc: 83.5078, loss_bbox: 0.3441, loss_mask: 0.1810, loss: 1.0282
2022-10-21 09:43:27,018 - mmdet - INFO - Saving checkpoint at 3 epochs

[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 39/39, 0.4 task/s, elapsed: 100s, ETA: 0s

2022-10-21 09:45:11,708 - mmdet - INFO - Evaluating bbox...

Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type bbox
DONE (t=0.54s).
Accumulating evaluation results...

2022-10-21 09:45:12,524 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.493
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.704
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.624
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.242
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.511
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.723
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.723
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.723
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.298
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.752

2022-10-21 09:45:12,525 - mmdet - INFO - Evaluating segm...

DONE (t=0.25s).
Loading and preparing results...
DONE (t=0.04s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type segm

/content/mmdetection/mmdet/datasets/coco.py:474: UserWarning: The key "bbox" is deleted for more accurate mask AP of small/medium/large instances since v2.12.0. This does not change the overall mAP calculation.
UserWarning)

DONE (t=0.60s).
Accumulating evaluation results...

/usr/local/lib/python3.7/dist-packages/pycocotools/cocoeval.py:378: DeprecationWarning: np.float is a deprecated alias for the builtin float. To silence this warning, use float by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use np.float64 here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
tp_sum = np.cumsum(tps, axis=1).astype(dtype=np.float)
2022-10-21 09:45:13,441 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.506
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.703
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.631
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.076
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.542
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.729
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.729
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.729
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.304
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.758

2022-10-21 09:45:13,453 - mmdet - INFO - Epoch(val) [3][39] bbox_mAP: 0.4930, bbox_mAP_50: 0.7040, bbox_mAP_75: 0.6240, bbox_mAP_s: 0.0000, bbox_mAP_m: 0.2420, bbox_mAP_l: 0.5110, bbox_mAP_copypaste: 0.493 0.704 0.624 0.000 0.242 0.511, segm_mAP: 0.5060, segm_mAP_50: 0.7030, segm_mAP_75: 0.6310, segm_mAP_s: 0.0000, segm_mAP_m: 0.0760, segm_mAP_l: 0.5420, segm_mAP_copypaste: 0.506 0.703 0.631 0.000 0.076 0.542

DONE (t=0.24s).

2022-10-21 09:46:16,877 - mmdet - INFO - Epoch [4][50/540] lr: 1.000e-04, eta: 0:25:12, time: 1.268, data_time: 0.323, memory: 9868, loss_rpn_cls: 0.0104, loss_rpn_bbox: 0.0835, loss_cls: 0.3932, acc: 84.1406, loss_bbox: 0.3254, loss_mask: 0.1633, loss: 0.9758
2022-10-21 09:47:19,378 - mmdet - INFO - Epoch [4][100/540] lr: 1.000e-04, eta: 0:23:52, time: 1.250, data_time: 0.259, memory: 9868, loss_rpn_cls: 0.0111, loss_rpn_bbox: 0.0813, loss_cls: 0.3986, acc: 83.4023, loss_bbox: 0.3316, loss_mask: 0.1797, loss: 1.0024
2022-10-21 09:48:23,727 - mmdet - INFO - Epoch [4][150/540] lr: 1.000e-04, eta: 0:22:35, time: 1.287, data_time: 0.307, memory: 9868, loss_rpn_cls: 0.0144, loss_rpn_bbox: 0.0822, loss_cls: 0.3767, acc: 84.8184, loss_bbox: 0.3258, loss_mask: 0.1710, loss: 0.9701
2022-10-21 09:49:26,360 - mmdet - INFO - Epoch [4][200/540] lr: 1.000e-04, eta: 0:21:17, time: 1.252, data_time: 0.272, memory: 9868, loss_rpn_cls: 0.0088, loss_rpn_bbox: 0.0848, loss_cls: 0.3917, acc: 84.2871, loss_bbox: 0.3330, loss_mask: 0.1689, loss: 0.9872
2022-10-21 09:50:28,763 - mmdet - INFO - Epoch [4][250/540] lr: 1.000e-04, eta: 0:20:00, time: 1.248, data_time: 0.262, memory: 9868, loss_rpn_cls: 0.0067, loss_rpn_bbox: 0.0771, loss_cls: 0.3915, acc: 83.6738, loss_bbox: 0.3132, loss_mask: 0.1613, loss: 0.9498
2022-10-21 09:51:32,891 - mmdet - INFO - Epoch [4][300/540] lr: 1.000e-04, eta: 0:18:44, time: 1.283, data_time: 0.272, memory: 9868, loss_rpn_cls: 0.0092, loss_rpn_bbox: 0.0798, loss_cls: 0.4066, acc: 83.9375, loss_bbox: 0.3299, loss_mask: 0.1737, loss: 0.9992
2022-10-21 09:52:33,752 - mmdet - INFO - Epoch [4][350/540] lr: 1.000e-04, eta: 0:17:28, time: 1.217, data_time: 0.255, memory: 9868, loss_rpn_cls: 0.0096, loss_rpn_bbox: 0.0815, loss_cls: 0.3980, acc: 83.9902, loss_bbox: 0.3337, loss_mask: 0.1868, loss: 1.0096
2022-10-21 09:53:32,541 - mmdet - INFO - Epoch [4][400/540] lr: 1.000e-04, eta: 0:16:12, time: 1.176, data_time: 0.228, memory: 9868, loss_rpn_cls: 0.0109, loss_rpn_bbox: 0.0769, loss_cls: 0.3674, acc: 84.5742, loss_bbox: 0.3106, loss_mask: 0.1557, loss: 0.9215
2022-10-21 09:54:33,726 - mmdet - INFO - Epoch [4][450/540] lr: 1.000e-04, eta: 0:14:57, time: 1.224, data_time: 0.259, memory: 9868, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0782, loss_cls: 0.3749, acc: 84.6094, loss_bbox: 0.3187, loss_mask: 0.1645, loss: 0.9464
2022-10-21 09:55:33,595 - mmdet - INFO - Epoch [4][500/540] lr: 1.000e-04, eta: 0:13:43, time: 1.197, data_time: 0.239, memory: 9868, loss_rpn_cls: 0.0079, loss_rpn_bbox: 0.0840, loss_cls: 0.3820, acc: 84.1113, loss_bbox: 0.3341, loss_mask: 0.1751, loss: 0.9830
2022-10-21 09:56:19,460 - mmdet - INFO - Saving checkpoint at 4 epochs

[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 39/39, 0.5 task/s, elapsed: 74s, ETA: 0s

2022-10-21 09:57:37,994 - mmdet - INFO - Evaluating bbox...

Loading and preparing results...
DONE (t=0.00s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type bbox
DONE (t=0.47s).
Accumulating evaluation results...

2022-10-21 09:57:38,722 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.507
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.706
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.622
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.232
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.526
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.738
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.738
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.738
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.308
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.767

2022-10-21 09:57:38,723 - mmdet - INFO - Evaluating segm...

DONE (t=0.24s).
Loading and preparing results...
DONE (t=0.03s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type segm

/content/mmdetection/mmdet/datasets/coco.py:474: UserWarning: The key "bbox" is deleted for more accurate mask AP of small/medium/large instances since v2.12.0. This does not change the overall mAP calculation.
UserWarning)

DONE (t=0.58s).
Accumulating evaluation results...

/usr/local/lib/python3.7/dist-packages/pycocotools/cocoeval.py:378: DeprecationWarning: np.float is a deprecated alias for the builtin float. To silence this warning, use float by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use np.float64 here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
tp_sum = np.cumsum(tps, axis=1).astype(dtype=np.float)
2022-10-21 09:57:39,627 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.512
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.706
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.631
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.102
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.547
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.738
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.738
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.738
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.327
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.767

2022-10-21 09:57:39,638 - mmdet - INFO - Epoch(val) [4][39] bbox_mAP: 0.5070, bbox_mAP_50: 0.7060, bbox_mAP_75: 0.6220, bbox_mAP_s: 0.0000, bbox_mAP_m: 0.2320, bbox_mAP_l: 0.5260, bbox_mAP_copypaste: 0.507 0.706 0.622 0.000 0.232 0.526, segm_mAP: 0.5120, segm_mAP_50: 0.7060, segm_mAP_75: 0.6310, segm_mAP_s: 0.0000, segm_mAP_m: 0.1020, segm_mAP_l: 0.5470, segm_mAP_copypaste: 0.512 0.706 0.631 0.000 0.102 0.547

DONE (t=0.24s).

2022-10-21 09:58:41,154 - mmdet - INFO - Epoch [5][50/540] lr: 1.000e-04, eta: 0:11:20, time: 1.230, data_time: 0.292, memory: 9868, loss_rpn_cls: 0.0148, loss_rpn_bbox: 0.0812, loss_cls: 0.3666, acc: 85.0605, loss_bbox: 0.3215, loss_mask: 0.1744, loss: 0.9585
2022-10-21 09:59:43,189 - mmdet - INFO - Epoch [5][100/540] lr: 1.000e-04, eta: 0:10:09, time: 1.241, data_time: 0.247, memory: 9868, loss_rpn_cls: 0.0089, loss_rpn_bbox: 0.0750, loss_cls: 0.3616, acc: 85.2188, loss_bbox: 0.3060, loss_mask: 0.1589, loss: 0.9104
2022-10-21 10:00:43,342 - mmdet - INFO - Epoch [5][150/540] lr: 1.000e-04, eta: 0:08:59, time: 1.203, data_time: 0.253, memory: 9868, loss_rpn_cls: 0.0075, loss_rpn_bbox: 0.0793, loss_cls: 0.3630, acc: 85.2324, loss_bbox: 0.3026, loss_mask: 0.1612, loss: 0.9136
2022-10-21 10:01:43,791 - mmdet - INFO - Epoch [5][200/540] lr: 1.000e-04, eta: 0:07:48, time: 1.209, data_time: 0.248, memory: 9868, loss_rpn_cls: 0.0087, loss_rpn_bbox: 0.0769, loss_cls: 0.3667, acc: 84.8496, loss_bbox: 0.3085, loss_mask: 0.1606, loss: 0.9214
2022-10-21 10:02:46,438 - mmdet - INFO - Epoch [5][250/540] lr: 1.000e-04, eta: 0:06:38, time: 1.253, data_time: 0.266, memory: 9868, loss_rpn_cls: 0.0081, loss_rpn_bbox: 0.0754, loss_cls: 0.3682, acc: 84.7617, loss_bbox: 0.2944, loss_mask: 0.1578, loss: 0.9040
2022-10-21 10:03:50,247 - mmdet - INFO - Epoch [5][300/540] lr: 1.000e-04, eta: 0:05:29, time: 1.276, data_time: 0.268, memory: 9868, loss_rpn_cls: 0.0084, loss_rpn_bbox: 0.0738, loss_cls: 0.3601, acc: 85.1758, loss_bbox: 0.3061, loss_mask: 0.1623, loss: 0.9107
2022-10-21 10:04:50,288 - mmdet - INFO - Epoch [5][350/540] lr: 1.000e-04, eta: 0:04:20, time: 1.201, data_time: 0.240, memory: 9868, loss_rpn_cls: 0.0085, loss_rpn_bbox: 0.0787, loss_cls: 0.3484, acc: 85.3242, loss_bbox: 0.3028, loss_mask: 0.1595, loss: 0.8980
2022-10-21 10:05:48,099 - mmdet - INFO - Epoch [5][400/540] lr: 1.000e-04, eta: 0:03:11, time: 1.156, data_time: 0.213, memory: 9868, loss_rpn_cls: 0.0121, loss_rpn_bbox: 0.0787, loss_cls: 0.3552, acc: 85.7227, loss_bbox: 0.3119, loss_mask: 0.1785, loss: 0.9364
2022-10-21 10:06:48,258 - mmdet - INFO - Epoch [5][450/540] lr: 1.000e-04, eta: 0:02:02, time: 1.203, data_time: 0.239, memory: 9868, loss_rpn_cls: 0.0088, loss_rpn_bbox: 0.0766, loss_cls: 0.3537, acc: 85.3828, loss_bbox: 0.3096, loss_mask: 0.1691, loss: 0.9178
2022-10-21 10:07:48,570 - mmdet - INFO - Epoch [5][500/540] lr: 1.000e-04, eta: 0:00:54, time: 1.206, data_time: 0.245, memory: 9868, loss_rpn_cls: 0.0091, loss_rpn_bbox: 0.0859, loss_cls: 0.3546, acc: 85.6035, loss_bbox: 0.3183, loss_mask: 0.1790, loss: 0.9469
2022-10-21 10:08:34,535 - mmdet - INFO - Saving checkpoint at 5 epochs

[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 39/39, 0.6 task/s, elapsed: 68s, ETA: 0s

2022-10-21 10:09:47,469 - mmdet - INFO - Evaluating bbox...

Loading and preparing results...
DONE (t=0.39s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type bbox
DONE (t=0.49s).
Accumulating evaluation results...

2022-10-21 10:09:48,585 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.542
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.742
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.675
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.254
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.561
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.748
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.748
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.748
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.302
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.778

2022-10-21 10:09:48,586 - mmdet - INFO - Evaluating segm...

DONE (t=0.22s).
Loading and preparing results...
DONE (t=0.03s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type segm

/content/mmdetection/mmdet/datasets/coco.py:474: UserWarning: The key "bbox" is deleted for more accurate mask AP of small/medium/large instances since v2.12.0. This does not change the overall mAP calculation.
UserWarning)

DONE (t=0.54s).
Accumulating evaluation results...

/usr/local/lib/python3.7/dist-packages/pycocotools/cocoeval.py:378: DeprecationWarning: np.float is a deprecated alias for the builtin float. To silence this warning, use float by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use np.float64 here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
tp_sum = np.cumsum(tps, axis=1).astype(dtype=np.float)
2022-10-21 10:09:49,440 - mmdet - INFO -
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.548
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=1000 ] = 0.745
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=1000 ] = 0.678
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.093
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.583
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.749
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=300 ] = 0.749
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=1000 ] = 0.749
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.331
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.778

2022-10-21 10:09:49,452 - mmdet - INFO - Epoch(val) [5][39] bbox_mAP: 0.5420, bbox_mAP_50: 0.7420, bbox_mAP_75: 0.6750, bbox_mAP_s: 0.0000, bbox_mAP_m: 0.2540, bbox_mAP_l: 0.5610, bbox_mAP_copypaste: 0.542 0.742 0.675 0.000 0.254 0.561, segm_mAP: 0.5480, segm_mAP_50: 0.7450, segm_mAP_75: 0.6780, segm_mAP_s: 0.0000, segm_mAP_m: 0.0930, segm_mAP_l: 0.5830, segm_mAP_copypaste: 0.548 0.745 0.678 0.000 0.093 0.583

DONE (t=0.24s).

from mmdet.apis import init_detector, inference_detector, show_result_pyplot
import mmcv


config_file = '/content/mmdetection/configs/swin/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco.py'
checkpoint_file = '/content/mmdetection/tutorial_exps/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco.pth'

# build the model from a config file and a checkpoint file
model = init_detector(config_file, checkpoint_file, device='cuda:0')

# test a single image
img = '/content/mm.jpg'
result = inference_detector(model, img)

# show the results
show_result_pyplot(model, img, result)

l

oad checkpoint from local path: /content/mmdetection/tutorial_exps/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco.pth
The model and loaded state dict do not match exactly

size mismatch for roi_head.bbox_head.fc_cls.weight: copying a param with shape torch.Size([33, 1024]) from checkpoint, the shape in current model is torch.Size([81, 1024]).
size mismatch for roi_head.bbox_head.fc_cls.bias: copying a param with shape torch.Size([33]) from checkpoint, the shape in current model is torch.Size([81]).
size mismatch for roi_head.bbox_head.fc_reg.weight: copying a param with shape torch.Size([128, 1024]) from checkpoint, the shape in current model is torch.Size([320, 1024]).
size mismatch for roi_head.bbox_head.fc_reg.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([320]).
size mismatch for roi_head.mask_head.conv_logits.weight: copying a param with shape torch.Size([32, 256, 1, 1]) from checkpoint, the shape in current model is torch.Size([80, 256, 1, 1]).
size mismatch for roi_head.mask_head.conv_logits.bias: copying a param with shape torch.Size([32]) from checkpoint, the shape in current model is torch.Size([80]).

/usr/local/lib/python3.7/dist-packages/mmdet-2.25.2-py3.7.egg/mmdet/datasets/utils.py:70: UserWarning: "ImageToTensor" pipeline is replaced by "DefaultFormatBundle" for batch inference. It is recommended to manually replace it in the test data pipeline in your config file.
'data pipeline in your config file.', UserWarning)

---

IndexError Traceback (most recent call last)

in
14
15 # show the results
---> 16 show_result_pyplot(model, img, result)
17

3 frames

/usr/local/lib/python3.7/dist-packages/mmdet-2.25.2-py3.7.egg/mmdet/core/visualization/image.py in draw_labels(ax, labels, positions, scores, class_names, color, font_size, scales, horizontal_alignment)
140 for i, (pos, label) in enumerate(zip(positions, labels)):
141 label_text = class_names[
--> 142 label] if class_names is not None else f'class {label}'
143 if scores is not None:
144 label_text += f'|{scores[i]:.02f}'

IndexError: tuple index out of range