使用mask_rcnn_r50_1x.yml对自定义数据集训练，初始损失值非常小，随着训练次数的增加，loss震荡
Created by: learning-boy
使用mask_rcnn_r50_1x.yml对自定义数据集训练，初始损失值非常小，随着训练次数的增加，loss震荡，自定义数据集数量为97张，77张训练集，20张验证集，目标数量为1，加上背景数，在配置文件中设置为2，学习率设置为0.0001；具体位置文件如下： architecture: MaskRCNN use_gpu: true max_iters: 10000 snapshot_iter: 200 log_smooth_window: 5 save_dir: output pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar metric: COCO weights: output/mask_rcnn_r50_1x/model_final num_classes: 2
MaskRCNN: backbone: ResNet rpn_head: RPNHead roi_extractor: RoIAlign bbox_assigner: BBoxAssigner bbox_head: BBoxHead mask_assigner: MaskAssigner mask_head: MaskHead
ResNet: norm_type: affine_channel norm_decay: 0. depth: 50 feature_maps: 4 freeze_at: 2
ResNetC5: depth: 50 norm_type: affine_channel
RPNHead: anchor_generator: anchor_sizes: [32, 64, 128, 256, 512] aspect_ratios: [0.5, 1.0, 2.0] stride: [16.0, 16.0] variance: [1.0, 1.0, 1.0, 1.0] rpn_target_assign: rpn_batch_size_per_im: 256 rpn_fg_fraction: 0.5 rpn_negative_overlap: 0.3 rpn_positive_overlap: 0.7 rpn_straddle_thresh: 0.0 train_proposal: min_size: 0.0 nms_thresh: 0.7 pre_nms_top_n: 12000 post_nms_top_n: 2000 test_proposal: min_size: 0.0 nms_thresh: 0.7 pre_nms_top_n: 6000 post_nms_top_n: 1000
RoIAlign: resolution: 14 spatial_scale: 0.0625 sampling_ratio: 0
BBoxHead: head: ResNetC5 nms: keep_top_k: 100 nms_threshold: 0.5 normalized: false score_threshold: 0.05
MaskHead: dilation: 1 conv_dim: 256 resolution: 14
BBoxAssigner: batch_size_per_im: 512 bbox_reg_weights: [0.1, 0.1, 0.2, 0.2] bg_thresh_hi: 0.5 bg_thresh_lo: 0.0 fg_fraction: 0.25 fg_thresh: 0.5
MaskAssigner: resolution: 14
LearningRate: base_lr: 0.00001 schedulers:
- !PiecewiseDecay gamma: 0.1 milestones: [120000, 160000]
- !LinearWarmup start_factor: 0.3333333333333333 steps: 500
OptimizerBuilder: optimizer: momentum: 0.9 type: Momentum regularizer: factor: 0.0001 type: L2
READER: 'mask_reader.yml' TrainReader: inputs_def: fields: ['image', 'im_info', 'im_id', 'gt_bbox', 'gt_class', 'is_crowd', 'gt_mask'] dataset: !COCODataSet image_dir: train2017 anno_path: annotations/instances_train.json dataset_dir: dataset/coco1 sample_transforms:
- !DecodeImage to_rgb: true
- !RandomFlipImage prob: 0.5 is_mask_flip: true
- !NormalizeImage is_channel_first: false is_scale: true mean: [0.485,0.456,0.406] std: [0.229, 0.224,0.225]
- !ResizeImage target_size: 800 max_size: 1333 interp: 1 use_cv2: true
- !Permute to_bgr: false channel_first: true batch_size: 1 shuffle: true worker_num: 2 drop_last: false use_process: false
EvalReader: inputs_def: fields: ['image', 'im_info', 'im_id', 'im_shape'] # for voc #fields: ['image', 'im_info', 'im_id', 'gt_bbox', 'gt_class', 'is_difficult'] dataset: !COCODataSet image_dir: val2017 anno_path: annotations/instances_val.json dataset_dir: dataset/coco1 sample_transforms:
- !DecodeImage to_rgb: true
- !NormalizeImage is_channel_first: false is_scale: true mean: [0.485,0.456,0.406] std: [0.229, 0.224,0.225]
- !ResizeImage interp: 1 max_size: 1333 target_size: 800 use_cv2: true
- !Permute channel_first: true to_bgr: false batch_size: 1 shuffle: false drop_last: false drop_empty: false worker_num: 2
TestReader: inputs_def: fields: ['image', 'im_info', 'im_id', 'im_shape'] dataset: !ImageFolder anno_path: annotations/instances_val.json sample_transforms:
- !DecodeImage to_rgb: true with_mixup: false
- !NormalizeImage is_channel_first: false is_scale: true mean: [0.485,0.456,0.406] std: [0.229, 0.224,0.225]
- !ResizeImage interp: 1 max_size: 1333 target_size: 800 use_cv2: true
- !Permute channel_first: true to_bgr: false batch_size: 1 shuffle: false drop_last: false 部分训练打印： /home/aistudio/PaddleDetection-release-0.2 BBoxAssigner: batch_size_per_im: 512 bbox_reg_weights:
- 0.1
- 0.1
- 0.2
- 0.2
bg_thresh_hi: 0.5
bg_thresh_lo: 0.0
fg_fraction: 0.25
fg_thresh: 0.5
num_classes: 81
shuffle_before_sample: true
BBoxHead:
[32mhead[0m: ResNetC5
[32mnms[0m:
keep_top_k: 100
nms_threshold: 0.5
normalized: false
score_threshold: 0.05
bbox_loss:
sigma: 1.0
box_coder:
axis: 1
box_normalized: false
code_type: decode_center_size
prior_box_var:
- 0.1
- 0.1
- 0.2
- 0.2 num_classes: 81 EvalReader: batch_size: 1 dataset: !COCODataSet anno_path: annotations/instances_val.json dataset_dir: dataset/coco1 image_dir: val2017 sample_num: -1 with_background: true drop_empty: false drop_last: false inputs_def: fields:
- image
- im_info
- im_id
- im_shape sample_transforms:
- !DecodeImage to_rgb: true with_mixup: false
- !NormalizeImage
is_channel_first: false
is_scale: true
mean:
- 0.485
- 0.456
- 0.406 std:
- 0.229
- 0.224
- 0.225
- !ResizeImage interp: 1 max_size: 1333 target_size: 800 use_cv2: true
- !Permute channel_first: true to_bgr: false shuffle: false worker_num: 2 LearningRate: [32mbase_lr[0m: 1.0e-05 [32mschedulers[0m:
- !PiecewiseDecay
gamma: 0.1
milestones:
- 120000
- 160000 values: null
- !LinearWarmup
start_factor: 0.3333333333333333
steps: 500
MaskAssigner:
num_classes: 81
resolution: 14
MaskHead:
conv_dim: 256
dilation: 1
norm_type: null
num_classes: 81
num_convs: 0
resolution: 14
MaskRCNN:
[32mbackbone[0m: ResNet
[32mrpn_head[0m: RPNHead
bbox_assigner: BBoxAssigner
bbox_head: BBoxHead
fpn: null
mask_assigner: MaskAssigner
mask_head: MaskHead
roi_extractor: RoIAlign
rpn_only: false
OptimizerBuilder:
optimizer:
momentum: 0.9
type: Momentum
regularizer:
factor: 0.0001
type: L2
RPNHead:
[32mrpn_target_assign[0m:
rpn_batch_size_per_im: 256
rpn_fg_fraction: 0.5
rpn_negative_overlap: 0.3
rpn_positive_overlap: 0.7
rpn_straddle_thresh: 0.0
[32mtest_proposal[0m:
min_size: 0.0
nms_thresh: 0.7
post_nms_top_n: 1000
pre_nms_top_n: 6000
[32mtrain_proposal[0m:
min_size: 0.0
nms_thresh: 0.7
post_nms_top_n: 2000
pre_nms_top_n: 12000
anchor_generator:
anchor_sizes:
- 32
- 64
- 128
- 256
- 512 aspect_ratios:
- 0.5
- 1.0
- 2.0 stride:
- 16.0
- 16.0 variance:
- 1.0
- 1.0
- 1.0
- 1.0 num_classes: 1 ResNet: [32mfeature_maps[0m: 4 [32mnorm_type[0m: affine_channel dcn_v2_stages: [] depth: 50 freeze_at: 2 freeze_norm: true gcb_params: {} gcb_stages: [] nonlocal_stages: [] norm_decay: 0.0 variant: b weight_prefix_name: '' ResNetC5: [32mnorm_type[0m: affine_channel depth: 50 feature_maps:
- 5
freeze_at: 2
freeze_norm: true
norm_decay: 0.0
variant: b
weight_prefix_name: ''
RoIAlign:
[32mresolution[0m: 14
sampling_ratio: 0
spatial_scale: 0.0625
TestReader:
batch_size: 1
dataset: !ImageFolder
anno_path: annotations/instances_val.json
dataset_dir: ''
image_dir: ''
sample_num: -1
use_default_label: null
with_background: true
drop_last: false
inputs_def:
fields:
- image
- im_info
- im_id
- im_shape sample_transforms:
- !DecodeImage to_rgb: true with_mixup: false
- !NormalizeImage
is_channel_first: false
is_scale: true
mean:
- 0.485
- 0.456
- 0.406 std:
- 0.229
- 0.224
- 0.225
- !ResizeImage interp: 1 max_size: 1333 target_size: 800 use_cv2: true
- !Permute
channel_first: true
to_bgr: false
shuffle: false
TrainReader:
batch_size: 1
dataset: !COCODataSet
anno_path: annotations/instances_train.json
dataset_dir: dataset/coco1
image_dir: train2017
sample_num: -1
with_background: true
drop_last: false
inputs_def:
fields:
- image
- im_info
- im_id
- gt_bbox
- gt_class
- is_crowd
- gt_mask sample_transforms:
- !DecodeImage to_rgb: true with_mixup: false
- !RandomFlipImage is_mask_flip: true is_normalized: false prob: 0.5
- !NormalizeImage
is_channel_first: false
is_scale: true
mean:
- 0.485
- 0.456
- 0.406 std:
- 0.229
- 0.224
- 0.225
- !ResizeImage interp: 1 max_size: 1333 target_size: 800 use_cv2: true
- !Permute channel_first: true to_bgr: false shuffle: true use_process: false worker_num: 2 architecture: MaskRCNN log_smooth_window: 5 max_iters: 10000 metric: COCO num_classes: 2 pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar save_dir: output snapshot_iter: 200 use_gpu: true weights: output/mask_rcnn_r50_1x/model_final
W0412 19:00:34.187557 692 device_context.cc:237] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 9.2, Runtime API Version: 9.0 W0412 19:00:34.193058 692 device_context.cc:245] device: 0, cuDNN Version: 7.3. 2020-04-12 19:00:35,977-INFO: Load model and fuse batch norm if have from https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar... 2020-04-12 19:00:35,977-INFO: Found /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained 2020-04-12 19:00:35,983-INFO: Loading parameters from /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained... 2020-04-12 19:00:35,983-WARNING: /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained.pdparams not found, try to load model file saved with [ save_params, save_persistables, save_vars ] 2020-04-12 19:00:35,983-WARNING: /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained.pdparams not found, try to load model file saved with [ save_params, save_persistables, save_vars ] 2020-04-12 19:00:35,989-WARNING: variable file [ /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained/fc_0.b_0 /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained/fc_0.w_0 ] not used 2020-04-12 19:00:35,989-WARNING: variable file [ /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained/fc_0.b_0 /home/aistudio/.cache/paddle/weights/ResNet50_cos_pretrained/fc_0.w_0 ] not used loading annotations into memory... Done (t=0.00s) creating index... index created! 2020-04-12 19:00:36,756-INFO: 77 samples in file dataset/coco1/annotations/instances_train.json 2020-04-12 19:00:36,757-INFO: places would be ommited when DataLoader is not iterable I0412 19:00:36.769461 692 parallel_executor.cc:440] The Program will be executed on CUDA using ParallelExecutor, 1 cards are used, so 1 programs are executed in parallel. I0412 19:00:36.793985 692 build_strategy.cc:365] SeqOnlyAllReduceOps:0, num_trainers:1 I0412 19:00:36.834561 692 parallel_executor.cc:307] Inplace strategy is enabled, when build_strategy.enable_inplace = True I0412 19:00:36.851727 692 parallel_executor.cc:375] Garbage collection strategy is enabled, when FLAGS_eager_delete_tensor_gb = 0 2020-04-12 19:00:37,437-INFO: iter: 0, lr: 0.000003, 'loss_cls': '0.709361', 'loss_bbox': '0.083127', 'loss_rpn_cls': '0.692449', 'loss_rpn_bbox': '0.050421', 'loss_mask': '1.289594', 'loss': '2.824951', time: 0.000, eta: 0:00:00 2020-04-12 19:00:44,129-INFO: iter: 20, lr: 0.000004, 'loss_cls': '0.650372', 'loss_bbox': '0.082469', 'loss_rpn_cls': '0.692441', 'loss_rpn_bbox': '0.058921', 'loss_mask': '0.909254', 'loss': '2.369460', time: 0.330, eta: 0:54:53 2020-04-12 19:00:50,174-INFO: iter: 40, lr: 0.000004, 'loss_cls': '0.556578', 'loss_bbox': '0.057319', 'loss_rpn_cls': '0.692058', 'loss_rpn_bbox': '0.051039', 'loss_mask': '0.784855', 'loss': '2.121035', time: 0.302, eta: 0:50:10 2020-04-12 19:00:56,208-INFO: iter: 60, lr: 0.000004, 'loss_cls': '0.478674', 'loss_bbox': '0.062347', 'loss_rpn_cls': '0.691236', 'loss_rpn_bbox': '0.014977', 'loss_mask': '0.853555', 'loss': '2.091259', time: 0.302, eta: 0:50:01 2020-04-12 19:01:02,312-INFO: iter: 80, lr: 0.000004, 'loss_cls': '0.438932', 'loss_bbox': '0.061302', 'loss_rpn_cls': '0.691199', 'loss_rpn_bbox': '0.024831', 'loss_mask': '0.716690', 'loss': '1.969233', time: 0.309, eta: 0:51:07 2020-04-12 19:01:08,408-INFO: iter: 100, lr: 0.000005, 'loss_cls': '0.403237', 'loss_bbox': '0.073609', 'loss_rpn_cls': '0.691486', 'loss_rpn_bbox': '0.041479', 'loss_mask': '0.898592', 'loss': '2.096893', time: 0.303, eta: 0:49:55 2020-04-12 19:01:14,456-INFO: iter: 120, lr: 0.000005, 'loss_cls': '0.344033', 'loss_bbox': '0.034582', 'loss_rpn_cls': '0.691175', 'loss_rpn_bbox': '0.012103', 'loss_mask': '0.704886', 'loss': '1.813018', time: 0.303, eta: 0:49:50 2020-04-12 19:01:20,527-INFO: iter: 140, lr: 0.000005, 'loss_cls': '0.314110', 'loss_bbox': '0.048092', 'loss_rpn_cls': '0.691119', 'loss_rpn_bbox': '0.033504', 'loss_mask': '0.715242', 'loss': '1.802891', time: 0.304, eta: 0:49:53 2020-04-12 19:01:26,583-INFO: iter: 160, lr: 0.000005, 'loss_cls': '0.319159', 'loss_bbox': '0.066780', 'loss_rpn_cls': '0.691007', 'loss_rpn_bbox': '0.013490', 'loss_mask': '0.735912', 'loss': '1.828875', time: 0.303, eta: 0:49:37 2020-04-12 19:01:32,692-INFO: iter: 180, lr: 0.000006, 'loss_cls': '0.281009', 'loss_bbox': '0.058378', 'loss_rpn_cls': '0.690924', 'loss_rpn_bbox': '0.029440', 'loss_mask': '0.765559', 'loss': '1.865903', time: 0.303, eta: 0:49:37 2020-04-12 19:01:38,764-INFO: iter: 200, lr: 0.000006, 'loss_cls': '0.262157', 'loss_bbox': '0.051979', 'loss_rpn_cls': '0.691355', 'loss_rpn_bbox': '0.026012', 'loss_mask': '0.628796', 'loss': '1.658930', time: 0.304, eta: 0:49:35 2020-04-12 19:01:38,764-INFO: Save model to output/mask_rcnn_r50_1x/200. 2020-04-12 19:01:52,874-INFO: iter: 220, lr: 0.000006, 'loss_cls': '0.213276', 'loss_bbox': '0.023042', 'loss_rpn_cls': '0.691732', 'loss_rpn_bbox': '0.018462', 'loss_mask': '0.571310', 'loss': '1.541541', time: 0.305, eta: 0:49:47 2020-04-12 19:01:59,041-INFO: iter: 240, lr: 0.000007, 'loss_cls': '0.211998', 'loss_bbox': '0.040730', 'loss_rpn_cls': '0.690834', 'loss_rpn_bbox': '0.030524', 'loss_mask': '0.564443', 'loss': '1.514041', time: 0.309, eta: 0:50:14 2020-04-12 19:02:05,150-INFO: iter: 260, lr: 0.000007, 'loss_cls': '0.226521', 'loss_bbox': '0.071110', 'loss_rpn_cls': '0.691036', 'loss_rpn_bbox': '0.026161', 'loss_mask': '0.630226', 'loss': '1.675641', time: 0.307, eta: 0:49:51 2020-04-12 19:02:11,278-INFO: iter: 280, lr: 0.000007, 'loss_cls': '0.192162', 'loss_bbox': '0.045098', 'loss_rpn_cls': '0.690945', 'loss_rpn_bbox': '0.053205', 'loss_mask': '0.646775', 'loss': '1.612253', time: 0.306, eta: 0:49:30 2020-04-12 19:02:17,376-INFO: iter: 300, lr: 0.000007, 'loss_cls': '0.199163', 'loss_bbox': '0.060750', 'loss_rpn_cls': '0.690203', 'loss_rpn_bbox': '0.025999', 'loss_mask': '0.574553', 'loss': '1.579114', time: 0.304, eta: 0:49:10 2020-04-12 19:02:23,495-INFO: iter: 320, lr: 0.000008, 'loss_cls': '0.179281', 'loss_bbox': '0.039314', 'loss_rpn_cls': '0.690511', 'loss_rpn_bbox': '0.014955', 'loss_mask': '0.528969', 'loss': '1.560330', time: 0.307, eta: 0:49:28 2020-04-12 19:02:29,579-INFO: iter: 340, lr: 0.000008, 'loss_cls': '0.185567', 'loss_bbox': '0.061789', 'loss_rpn_cls': '0.690158', 'loss_rpn_bbox': '0.038831', 'loss_mask': '0.568622', 'loss': '1.556766', time: 0.305, eta: 0:49:09 2020-04-12 19:02:35,672-INFO: iter: 360, lr: 0.000008, 'loss_cls': '0.189553', 'loss_bbox': '0.057919', 'loss_rpn_cls': '0.689832', 'loss_rpn_bbox': '0.029010', 'loss_mask': '0.654331', 'loss': '1.622463', time: 0.305, eta: 0:49:00 2020-04-12 19:02:41,764-INFO: iter: 380, lr: 0.000008, 'loss_cls': '0.155318', 'loss_bbox': '0.035328', 'loss_rpn_cls': '0.689991', 'loss_rpn_bbox': '0.022251', 'loss_mask': '0.573573', 'loss': '1.506014', time: 0.304, eta: 0:48:42 2020-04-12 19:02:47,895-INFO: iter: 400, lr: 0.000009, 'loss_cls': '0.188258', 'loss_bbox': '0.089682', 'loss_rpn_cls': '0.689922', 'loss_rpn_bbox': '0.029450', 'loss_mask': '0.532518', 'loss': '1.570914', time: 0.307, eta: 0:49:02 2020-04-12 19:02:47,896-INFO: Save model to output/mask_rcnn_r50_1x/400. 2020-04-12 19:03:00,357-INFO: iter: 420, lr: 0.000009, 'loss_cls': '0.180680', 'loss_bbox': '0.087848', 'loss_rpn_cls': '0.690892', 'loss_rpn_bbox': '0.036409', 'loss_mask': '0.538045', 'loss': '1.554493', time: 0.306, eta: 0:48:52 2020-04-12 19:03:06,490-INFO: iter: 440, lr: 0.000009, 'loss_cls': '0.172253', 'loss_bbox': '0.069269', 'loss_rpn_cls': '0.689331', 'loss_rpn_bbox': '0.015508', 'loss_mask': '0.540710', 'loss': '1.485177', time: 0.306, eta: 0:48:43 2020-04-12 19:03:12,576-INFO: iter: 460, lr: 0.000009, 'loss_cls': '0.194546', 'loss_bbox': '0.076496', 'loss_rpn_cls': '0.690464', 'loss_rpn_bbox': '0.040063', 'loss_mask': '0.578017', 'loss': '1.556236', time: 0.304, eta: 0:48:23 2020-04-12 19:03:18,700-INFO: iter: 480, lr: 0.000010, 'loss_cls': '0.158979', 'loss_bbox': '0.067991', 'loss_rpn_cls': '0.689469', 'loss_rpn_bbox': '0.024652', 'loss_mask': '0.497855', 'loss': '1.420976', time: 0.305, eta: 0:48:24 2020-04-12 19:03:24,837-INFO: iter: 500, lr: 0.000010, 'loss_cls': '0.171781', 'loss_bbox': '0.073472', 'loss_rpn_cls': '0.689256', 'loss_rpn_bbox': '0.017589', 'loss_mask': '0.529966', 'loss': '1.483658', time: 0.308, eta: 0:48:46 2020-04-12 19:03:31,028-INFO: iter: 520, lr: 0.000010, 'loss_cls': '0.124287', 'loss_bbox': '0.025206', 'loss_rpn_cls': '0.689655', 'loss_rpn_bbox': '0.021209', 'loss_mask': '0.565265', 'loss': '1.387576', time: 0.307, eta: 0:48:31 2020-04-12 19:03:37,201-INFO: iter: 540, lr: 0.000010, 'loss_cls': '0.176836', 'loss_bbox': '0.096081', 'loss_rpn_cls': '0.689362', 'loss_rpn_bbox': '0.027724', 'loss_mask': '0.504007', 'loss': '1.605950', time: 0.310, eta: 0:48:49 2020-04-12 19:03:43,424-INFO: iter: 560, lr: 0.000010, 'loss_cls': '0.137976', 'loss_bbox': '0.061506', 'loss_rpn_cls': '0.689123', 'loss_rpn_bbox': '0.032565', 'loss_mask': '0.488347', 'loss': '1.385690', time: 0.311, eta: 0:48:55 2020-04-12 19:03:49,612-INFO: iter: 580, lr: 0.000010, 'loss_cls': '0.213834', 'loss_bbox': '0.123614', 'loss_rpn_cls': '0.688740', 'loss_rpn_bbox': '0.020710', 'loss_mask': '0.515485', 'loss': '1.621922', time: 0.306, eta: 0:48:01 2020-04-12 19:03:55,788-INFO: iter: 600, lr: 0.000010, 'loss_cls': '0.159223', 'loss_bbox': '0.076401', 'loss_rpn_cls': '0.688530', 'loss_rpn_bbox': '0.025418', 'loss_mask': '0.558597', 'loss': '1.508233', time: 0.307, eta: 0:48:08 2020-04-12 19:03:55,788-INFO: Save model to output/mask_rcnn_r50_1x/600. 2020-04-12 19:04:08,543-INFO: iter: 620, lr: 0.000010, 'loss_cls': '0.112575', 'loss_bbox': '0.032445', 'loss_rpn_cls': '0.688121', 'loss_rpn_bbox': '0.013999', 'loss_mask': '0.495780', 'loss': '1.422422', time: 0.307, eta: 0:47:56 2020-04-12 19:04:14,693-INFO: iter: 640, lr: 0.000010, 'loss_cls': '0.142164', 'loss_bbox': '0.065522', 'loss_rpn_cls': '0.687942', 'loss_rpn_bbox': '0.033242', 'loss_mask': '0.471026', 'loss': '1.422687', time: 0.307, eta: 0:47:56 2020-04-12 19:04:20,835-INFO: iter: 660, lr: 0.000010, 'loss_cls': '0.212521', 'loss_bbox': '0.136588', 'loss_rpn_cls': '0.688202', 'loss_rpn_bbox': '0.039814', 'loss_mask': '0.484226', 'loss': '1.566038', time: 0.308, eta: 0:47:55 2020-04-12 19:04:27,107-INFO: iter: 680, lr: 0.000010, 'loss_cls': '0.147422', 'loss_bbox': '0.069179', 'loss_rpn_cls': '0.688367', 'loss_rpn_bbox': '0.013292', 'loss_mask': '0.528954', 'loss': '1.528994', time: 0.309, eta: 0:47:58 2020-04-12 19:04:33,253-INFO: iter: 700, lr: 0.000010, 'loss_cls': '0.124653', 'loss_bbox': '0.047101', 'loss_rpn_cls': '0.688363', 'loss_rpn_bbox': '0.027703', 'loss_mask': '0.516500', 'loss': '1.440751', time: 0.307, eta: 0:47:33 2020-04-12 19:04:39,407-INFO: iter: 720, lr: 0.000010, 'loss_cls': '0.112647', 'loss_bbox': '0.039291', 'loss_rpn_cls': '0.687909', 'loss_rpn_bbox': '0.013821', 'loss_mask': '0.474246', 'loss': '1.381654', time: 0.307, eta: 0:47:29 2020-04-12 19:04:45,564-INFO: iter: 740, lr: 0.000010, 'loss_cls': '0.177098', 'loss_bbox': '0.108517', 'loss_rpn_cls': '0.687508', 'loss_rpn_bbox': '0.038574', 'loss_mask': '0.493663', 'loss': '1.485019', time: 0.308, eta: 0:47:31 2020-04-12 19:04:51,736-INFO: iter: 760, lr: 0.000010, 'loss_cls': '0.182641', 'loss_bbox': '0.120964', 'loss_rpn_cls': '0.687407', 'loss_rpn_bbox': '0.044008', 'loss_mask': '0.550386', 'loss': '1.616782', time: 0.310, eta: 0:47:45 2020-04-12 19:04:57,915-INFO: iter: 780, lr: 0.000010, 'loss_cls': '0.155265', 'loss_bbox': '0.057853', 'loss_rpn_cls': '0.687158', 'loss_rpn_bbox': '0.021123', 'loss_mask': '0.513816', 'loss': '1.436609', time: 0.307, eta: 0:47:10 2020-04-12 19:05:04,076-INFO: iter: 800, lr: 0.000010, 'loss_cls': '0.192818', 'loss_bbox': '0.103482', 'loss_rpn_cls': '0.687165', 'loss_rpn_bbox': '0.020791', 'loss_mask': '0.513536', 'loss': '1.577556', time: 0.308, eta: 0:47:11 2020-04-12 19:05:04,076-INFO: Save model to output/mask_rcnn_r50_1x/800. 2020-04-12 19:05:16,797-INFO: iter: 820, lr: 0.000010, 'loss_cls': '0.191460', 'loss_bbox': '0.107138', 'loss_rpn_cls': '0.686573', 'loss_rpn_bbox': '0.040939', 'loss_mask': '0.548526', 'loss': '1.539917', time: 0.307, eta: 0:46:55 2020-04-12 19:05:22,924-INFO: iter: 840, lr: 0.000010, 'loss_cls': '0.152466', 'loss_bbox': '0.087949', 'loss_rpn_cls': '0.686791', 'loss_rpn_bbox': '0.024740', 'loss_mask': '0.473760', 'loss': '1.394567', time: 0.307, eta: 0:46:48 2020-04-12 19:05:29,064-INFO: iter: 860, lr: 0.000010, 'loss_cls': '0.178251', 'loss_bbox': '0.129728', 'loss_rpn_cls': '0.685862', 'loss_rpn_bbox': '0.050313', 'loss_mask': '0.570226', 'loss': '1.615928', time: 0.307, eta: 0:46:49 2020-04-12 19:05:35,205-INFO: iter: 880, lr: 0.000010, 'loss_cls': '0.177442', 'loss_bbox': '0.094552', 'loss_rpn_cls': '0.686490', 'loss_rpn_bbox': '0.022188', 'loss_mask': '0.526434', 'loss': '1.579593', time: 0.308, eta: 0:46:45 2020-04-12 19:05:41,360-INFO: iter: 900, lr: 0.000010, 'loss_cls': '0.157121', 'loss_bbox': '0.072273', 'loss_rpn_cls': '0.686740', 'loss_rpn_bbox': '0.045151', 'loss_mask': '0.529014', 'loss': '1.438815', time: 0.309, eta: 0:46:48 2020-04-12 19:05:47,509-INFO: iter: 920, lr: 0.000010, 'loss_cls': '0.186652', 'loss_bbox': '0.121628', 'loss_rpn_cls': '0.686137', 'loss_rpn_bbox': '0.049367', 'loss_mask': '0.478019', 'loss': '1.533623', time: 0.308, eta: 0:46:33 2020-04-12 19:05:53,668-INFO: iter: 940, lr: 0.000010, 'loss_cls': '0.206999', 'loss_bbox': '0.117210', 'loss_rpn_cls': '0.686048', 'loss_rpn_bbox': '0.039984', 'loss_mask': '0.563714', 'loss': '1.551200', time: 0.307, eta: 0:46:21 2020-04-12 19:05:59,840-INFO: iter: 960, lr: 0.000010, 'loss_cls': '0.162035', 'loss_bbox': '0.085512', 'loss_rpn_cls': '0.686338', 'loss_rpn_bbox': '0.026195', 'loss_mask': '0.500405', 'loss': '1.399719', time: 0.309, eta: 0:46:32 2020-04-12 19:06:05,998-INFO: iter: 980, lr: 0.000010, 'loss_cls': '0.180501', 'loss_bbox': '0.093311', 'loss_rpn_cls': '0.684919', 'loss_rpn_bbox': '0.021927', 'loss_mask': '0.479859', 'loss': '1.375473', time: 0.307, eta: 0:46:12 2020-04-12 19:06:12,178-INFO: iter: 1000, lr: 0.000010, 'loss_cls': '0.193095', 'loss_bbox': '0.109644', 'loss_rpn_cls': '0.685483', 'loss_rpn_bbox': '0.050795', 'loss_mask': '0.578834', 'loss': '1.604461', time: 0.309, eta: 0:46:22 2020-04-12 19:06:12,179-INFO: Save model to output/mask_rcnn_r50_1x/1000. 2020-04-12 19:06:24,669-INFO: iter: 1020, lr: 0.000010, 'loss_cls': '0.189820', 'loss_bbox': '0.121059', 'loss_rpn_cls': '0.685088', 'loss_rpn_bbox': '0.033061', 'loss_mask': '0.480390', 'loss': '1.454112', time: 0.308, eta: 0:46:08 2020-04-12 19:06:30,828-INFO: iter: 1040, lr: 0.000010, 'loss_cls': '0.161560', 'loss_bbox': '0.069314', 'loss_rpn_cls': '0.685109', 'loss_rpn_bbox': '0.012457', 'loss_mask': '0.509523', 'loss': '1.459299', time: 0.307, eta: 0:45:52 2020-04-12 19:06:36,987-INFO: iter: 1060, lr: 0.000010, 'loss_cls': '0.189881', 'loss_bbox': '0.104732', 'loss_rpn_cls': '0.684781', 'loss_rpn_bbox': '0.040619', 'loss_mask': '0.513582', 'loss': '1.525422', time: 0.307, eta: 0:45:47 2020-04-12 19:06:43,181-INFO: iter: 1080, lr: 0.000010, 'loss_cls': '0.133512', 'loss_bbox': '0.083821', 'loss_rpn_cls': '0.684731', 'loss_rpn_bbox': '0.024672', 'loss_mask': '0.451763', 'loss': '1.367279', time: 0.310, eta: 0:46:00 2020-04-12 19:06:49,367-INFO: iter: 1100, lr: 0.000010, 'loss_cls': '0.202809', 'loss_bbox': '0.110874', 'loss_rpn_cls': '0.684618', 'loss_rpn_bbox': '0.036716', 'loss_mask': '0.546747', 'loss': '1.583275', time: 0.308, eta: 0:45:42 2020-04-12 19:06:55,524-INFO: iter: 1120, lr: 0.000010, 'loss_cls': '0.219490', 'loss_bbox': '0.141617', 'loss_rpn_cls': '0.685144', 'loss_rpn_bbox': '0.039648', 'loss_mask': '0.493588', 'loss': '1.636144', time: 0.308, eta: 0:45:33 2020-04-12 19:07:01,689-INFO: iter: 1140, lr: 0.000010, 'loss_cls': '0.171572', 'loss_bbox': '0.074585', 'loss_rpn_cls': '0.683746', 'loss_rpn_bbox': '0.020758', 'loss_mask': '0.442087', 'loss': '1.424493', time: 0.308, eta: 0:45:32 2020-04-12 19:07:07,872-INFO: iter: 1160, lr: 0.000010, 'loss_cls': '0.191832', 'loss_bbox': '0.102839', 'loss_rpn_cls': '0.683925', 'loss_rpn_bbox': '0.014334', 'loss_mask': '0.548521', 'loss': '1.496444', time: 0.308, eta: 0:45:27 2020-04-12 19:07:14,044-INFO: iter: 1180, lr: 0.000010, 'loss_cls': '0.188281', 'loss_bbox': '0.102847', 'loss_rpn_cls': '0.684284', 'loss_rpn_bbox': '0.020482', 'loss_mask': '0.500807', 'loss': '1.510263', time: 0.309, eta: 0:45:24 2020-04-12 19:07:20,221-INFO: iter: 1200, lr: 0.000010, 'loss_cls': '0.198665', 'loss_bbox': '0.124839', 'loss_rpn_cls': '0.683891', 'loss_rpn_bbox': '0.032118', 'loss_mask': '0.461677', 'loss': '1.531209', time: 0.310, eta: 0:45:27 2020-04-12 19:07:20,222-INFO: Save model to output/mask_rcnn_r50_1x/1200. 2020-04-12 19:07:32,662-INFO: iter: 1220, lr: 0.000010, 'loss_cls': '0.162001', 'loss_bbox': '0.091546', 'loss_rpn_cls': '0.683902', 'loss_rpn_bbox': '0.024822', 'loss_mask': '0.486171', 'loss': '1.462982', time: 0.309, eta: 0:45:11 2020-04-12 19:07:38,842-INFO: iter: 1240, lr: 0.000010, 'loss_cls': '0.223047', 'loss_bbox': '0.159888', 'loss_rpn_cls': '0.683615', 'loss_rpn_bbox': '0.038499', 'loss_mask': '0.501161', 'loss': '1.616851', time: 0.310, eta: 0:45:13 2020-04-12 19:07:45,021-INFO: iter: 1260, lr: 0.000010, 'loss_cls': '0.167389', 'loss_bbox': '0.081645', 'loss_rpn_cls': '0.684311', 'loss_rpn_bbox': '0.036812', 'loss_mask': '0.478512', 'loss': '1.445109', time: 0.308, eta: 0:44:50 2020-04-12 19:07:51,195-INFO: iter: 1280, lr: 0.000010, 'loss_cls': '0.197705', 'loss_bbox': '0.126737', 'loss_rpn_cls': '0.683524', 'loss_rpn_bbox': '0.038243', 'loss_mask': '0.464142', 'loss': '1.530859', time: 0.309, eta: 0:44:53 2020-04-12 19:07:57,368-INFO: iter: 1300, lr: 0.000010, 'loss_cls': '0.158003', 'loss_bbox': '0.078831', 'loss_rpn_cls': '0.683364', 'loss_rpn_bbox': '0.038230', 'loss_mask': '0.465716', 'loss': '1.413317', time: 0.308, eta: 0:44:40 2020-04-12 19:08:03,550-INFO: iter: 1320, lr: 0.000010, 'loss_cls': '0.129138', 'loss_bbox': '0.071601', 'loss_rpn_cls': '0.683260', 'loss_rpn_bbox': '0.017819', 'loss_mask': '0.495044', 'loss': '1.377244', time: 0.308, eta: 0:44:37 2020-04-12 19:08:09,752-INFO: iter: 1340, lr: 0.000010, 'loss_cls': '0.200677', 'loss_bbox': '0.112606', 'loss_rpn_cls': '0.681396', 'loss_rpn_bbox': '0.024488', 'loss_mask': '0.426530', 'loss': '1.504689', time: 0.309, eta: 0:44:37 2020-04-12 19:08:15,970-INFO: iter: 1360, lr: 0.000010, 'loss_cls': '0.194279', 'loss_bbox': '0.110581', 'loss_rpn_cls': '0.682089', 'loss_rpn_bbox': '0.029690', 'loss_mask': '0.526065', 'loss': '1.585550', time: 0.314, eta: 0:45:16 2020-04-12 19:08:22,174-INFO: iter: 1380, lr: 0.000010, 'loss_cls': '0.219638', 'loss_bbox': '0.143685', 'loss_rpn_cls': '0.682407', 'loss_rpn_bbox': '0.021669', 'loss_mask': '0.535162', 'loss': '1.554770', time: 0.310, eta: 0:44:34 2020-04-12 19:08:28,392-INFO: iter: 1400, lr: 0.000010, 'loss_cls': '0.209414', 'loss_bbox': '0.119782', 'loss_rpn_cls': '0.682301', 'loss_rpn_bbox': '0.039549', 'loss_mask': '0.506553', 'loss': '1.606503', time: 0.310, eta: 0:44:27 2020-04-12 19:08:28,392-INFO: Save model to output/mask_rcnn_r50_1x/1400. 2020-04-12 19:08:41,736-INFO: iter: 1420, lr: 0.000010, 'loss_cls': '0.197517', 'loss_bbox': '0.140778', 'loss_rpn_cls': '0.683115', 'loss_rpn_bbox': '0.028594', 'loss_mask': '0.450000', 'loss': '1.492373', time: 0.310, eta: 0:44:20 2020-04-12 19:08:47,928-INFO: iter: 1440, lr: 0.000010, 'loss_cls': '0.199652', 'loss_bbox': '0.126807', 'loss_rpn_cls': '0.681883', 'loss_rpn_bbox': '0.029589', 'loss_mask': '0.474785', 'loss': '1.500979', time: 0.310, eta: 0:44:10 2020-04-12 19:08:54,131-INFO: iter: 1460, lr: 0.000010, 'loss_cls': '0.196899', 'loss_bbox': '0.140777', 'loss_rpn_cls': '0.682275', 'loss_rpn_bbox': '0.027317', 'loss_mask': '0.444613', 'loss': '1.425404', time: 0.312, eta: 0:44:28 2020-04-12 19:09:00,328-INFO: iter: 1480, lr: 0.000010, 'loss_cls': '0.204998', 'loss_bbox': '0.130305', 'loss_rpn_cls': '0.682051', 'loss_rpn_bbox': '0.027538', 'loss_mask': '0.485118', 'loss': '1.577770', time: 0.312, eta: 0:44:14 2020-04-12 19:09:06,523-INFO: iter: 1500, lr: 0.000010, 'loss_cls': '0.237758', 'loss_bbox': '0.159909', 'loss_rpn_cls': '0.680939', 'loss_rpn_bbox': '0.039336', 'loss_mask': '0.473339', 'loss': '1.571156', time: 0.310, eta: 0:43:57 2020-04-12 19:09:12,717-INFO: iter: 1520, lr: 0.000010, 'loss_cls': '0.204136', 'loss_bbox': '0.147899', 'loss_rpn_cls': '0.681489', 'loss_rpn_bbox': '0.020549', 'loss_mask': '0.514341', 'loss': '1.520210', time: 0.310, eta: 0:43:44 2020-04-12 19:09:18,903-INFO: iter: 1540, lr: 0.000010, 'loss_cls': '0.115675', 'loss_bbox': '0.038880', 'loss_rpn_cls': '0.680823', 'loss_rpn_bbox': '0.018090', 'loss_mask': '0.533428', 'loss': '1.378499', time: 0.308, eta: 0:43:28 2020-04-12 19:09:25,115-INFO: iter: 1560, lr: 0.000010, 'loss_cls': '0.204536', 'loss_bbox': '0.125220', 'loss_rpn_cls': '0.680893', 'loss_rpn_bbox': '0.024700', 'loss_mask': '0.447253', 'loss': '1.537062', time: 0.311, eta: 0:43:41 2020-04-12 19:09:31,347-INFO: iter: 1580, lr: 0.000010, 'loss_cls': '0.227514', 'loss_bbox': '0.133389', 'loss_rpn_cls': '0.680879', 'loss_rpn_bbox': '0.046361', 'loss_mask': '0.452611', 'loss': '1.624777', time: 0.312, eta: 0:43:47 2020-04-12 19:09:37,607-INFO: iter: 1600, lr: 0.000010, 'loss_cls': '0.161641', 'loss_bbox': '0.064353', 'loss_rpn_cls': '0.680830', 'loss_rpn_bbox': '0.017151', 'loss_mask': '0.463437', 'loss': '1.388914', time: 0.310, eta: 0:43:20 2020-04-12 19:09:37,607-INFO: Save model to output/mask_rcnn_r50_1x/1600. 2020-04-12 19:09:51,346-INFO: iter: 1620, lr: 0.000010, 'loss_cls': '0.202798', 'loss_bbox': '0.115121', 'loss_rpn_cls': '0.680751', 'loss_rpn_bbox': '0.038178', 'loss_mask': '0.454965', 'loss': '1.493991', time: 0.309, eta: 0:43:13 2020-04-12 19:09:57,542-INFO: iter: 1640, lr: 0.000010, 'loss_cls': '0.150487', 'loss_bbox': '0.069327', 'loss_rpn_cls': '0.681516', 'loss_rpn_bbox': '0.051813', 'loss_mask': '0.465364', 'loss': '1.489097', time: 0.309, eta: 0:43:01 2020-04-12 19:10:03,819-INFO: iter: 1660, lr: 0.000010, 'loss_cls': '0.226892', 'loss_bbox': '0.139751', 'loss_rpn_cls': '0.680934', 'loss_rpn_bbox': '0.028844', 'loss_mask': '0.436827', 'loss': '1.621590', time: 0.316, eta: 0:43:51 2020-04-12 19:10:10,081-INFO: iter: 1680, lr: 0.000010, 'loss_cls': '0.161868', 'loss_bbox': '0.109561', 'loss_rpn_cls': '0.680030', 'loss_rpn_bbox': '0.024972', 'loss_mask': '0.430990', 'loss': '1.349245', time: 0.312, eta: 0:43:15 2020-04-12 19:10:16,386-INFO: iter: 1700, lr: 0.000010, 'loss_cls': '0.202876', 'loss_bbox': '0.126174', 'loss_rpn_cls': '0.679840', 'loss_rpn_bbox': '0.028835', 'loss_mask': '0.477904', 'loss': '1.498930', time: 0.317, eta: 0:43:49 2020-04-12 19:10:22,708-INFO: iter: 1720, lr: 0.000010, 'loss_cls': '0.244006', 'loss_bbox': '0.150203', 'loss_rpn_cls': '0.679759', 'loss_rpn_bbox': '0.038487', 'loss_mask': '0.510047', 'loss': '1.626436', time: 0.313, eta: 0:43:09 2020-04-12 19:10:29,057-INFO: iter: 1740, lr: 0.000010, 'loss_cls': '0.218859', 'loss_bbox': '0.137107', 'loss_rpn_cls': '0.678940', 'loss_rpn_bbox': '0.021903', 'loss_mask': '0.430260', 'loss': '1.595463', time: 0.312, eta: 0:42:55 2020-04-12 19:10:35,289-INFO: iter: 1760, lr: 0.000010, 'loss_cls': '0.130746', 'loss_bbox': '0.067939', 'loss_rpn_cls': '0.678889', 'loss_rpn_bbox': '0.022606', 'loss_mask': '0.421495', 'loss': '1.354213', time: 0.313, eta: 0:43:02 2020-04-12 19:10:41,584-INFO: iter: 1780, lr: 0.000010, 'loss_cls': '0.262385', 'loss_bbox': '0.169043', 'loss_rpn_cls': '0.679589', 'loss_rpn_bbox': '0.031818', 'loss_mask': '0.487277', 'loss': '1.607844', time: 0.312, eta: 0:42:43 2020-04-12 19:10:47,807-INFO: iter: 1800, lr: 0.000010, 'loss_cls': '0.226841', 'loss_bbox': '0.141067', 'loss_rpn_cls': '0.679890', 'loss_rpn_bbox': '0.041424', 'loss_mask': '0.496764', 'loss': '1.617090', time: 0.313, eta: 0:42:46 2020-04-12 19:10:47,807-INFO: Save model to output/mask_rcnn_r50_1x/1800. 2020-04-12 19:11:02,719-INFO: iter: 1820, lr: 0.000010, 'loss_cls': '0.169506', 'loss_bbox': '0.068839', 'loss_rpn_cls': '0.678894', 'loss_rpn_bbox': '0.025192', 'loss_mask': '0.479178', 'loss': '1.408392', time: 0.310, eta: 0:42:16 2020-04-12 19:11:08,918-INFO: iter: 1840, lr: 0.000010, 'loss_cls': '0.117834', 'loss_bbox': '0.069967', 'loss_rpn_cls': '0.678962', 'loss_rpn_bbox': '0.029704', 'loss_mask': '0.480582', 'loss': '1.398610', time: 0.311, eta: 0:42:13 2020-04-12 19:11:15,134-INFO: iter: 1860, lr: 0.000010, 'loss_cls': '0.122838', 'loss_bbox': '0.045424', 'loss_rpn_cls': '0.679158', 'loss_rpn_bbox': '0.030094', 'loss_mask': '0.459384', 'loss': '1.409732', time: 0.311, eta: 0:42:15 2020-04-12 19:11:21,347-INFO: iter: 1880, lr: 0.000010, 'loss_cls': '0.192062', 'loss_bbox': '0.127198', 'loss_rpn_cls': '0.678984', 'loss_rpn_bbox': '0.024144', 'loss_mask': '0.445525', 'loss': '1.510110', time: 0.311, eta: 0:42:08 2020-04-12 19:11:27,592-INFO: iter: 1900, lr: 0.000010, 'loss_cls': '0.188529', 'loss_bbox': '0.089532', 'loss_rpn_cls': '0.678246', 'loss_rpn_bbox': '0.036887', 'loss_mask': '0.498453', 'loss': '1.480202', time: 0.316, eta: 0:42:35 2020-04-12 19:11:33,811-INFO: iter: 1920, lr: 0.000010, 'loss_cls': '0.216560', 'loss_bbox': '0.127347', 'loss_rpn_cls': '0.678729', 'loss_rpn_bbox': '0.020469', 'loss_mask': '0.464754', 'loss': '1.497090', time: 0.311, eta: 0:41:51 2020-04-12 19:11:40,122-INFO: iter: 1940, lr: 0.000010, 'loss_cls': '0.169251', 'loss_bbox': '0.090620', 'loss_rpn_cls': '0.677745', 'loss_rpn_bbox': '0.023998', 'loss_mask': '0.480123', 'loss': '1.482071', time: 0.316, eta: 0:42:25 2020-04-12 19:11:46,426-INFO: iter: 1960, lr: 0.000010, 'loss_cls': '0.186032', 'loss_bbox': '0.125796', 'loss_rpn_cls': '0.678326', 'loss_rpn_bbox': '0.026806', 'loss_mask': '0.436666', 'loss': '1.465089', time: 0.318, eta: 0:42:37 2020-04-12 19:11:52,711-INFO: iter: 1980, lr: 0.000010, 'loss_cls': '0.231929', 'loss_bbox': '0.140800', 'loss_rpn_cls': '0.678110', 'loss_rpn_bbox': '0.045696', 'loss_mask': '0.503554', 'loss': '1.621533', time: 0.313, eta: 0:41:46 2020-04-12 19:11:59,022-INFO: iter: 2000, lr: 0.000010, 'loss_cls': '0.237458', 'loss_bbox': '0.159110', 'loss_rpn_cls': '0.678751', 'loss_rpn_bbox': '0.020447', 'loss_mask': '0.519607', 'loss': '1.624557', time: 0.318, eta: 0:42:20 2020-04-12 19:11:59,022-INFO: Save model to output/mask_rcnn_r50_1x/2000. 2020-04-12 19:12:15,515-INFO: iter: 2020, lr: 0.000010, 'loss_cls': '0.213213', 'loss_bbox': '0.157893', 'loss_rpn_cls': '0.676772', 'loss_rpn_bbox': '0.037170', 'loss_mask': '0.479328', 'loss': '1.563199', time: 0.313, eta: 0:41:37 2020-04-12 19:12:21,826-INFO: iter: 2040, lr: 0.000010, 'loss_cls': '0.178655', 'loss_bbox': '0.091364', 'loss_rpn_cls': '0.677610', 'loss_rpn_bbox': '0.022278', 'loss_mask': '0.459008', 'loss': '1.478283', time: 0.321, eta: 0:42:33 2020-04-12 19:12:28,142-INFO: iter: 2060, lr: 0.000010, 'loss_cls': '0.142857', 'loss_bbox': '0.083172', 'loss_rpn_cls': '0.676745', 'loss_rpn_bbox': '0.025520', 'loss_mask': '0.512292', 'loss': '1.396975', time: 0.313, eta: 0:41:26 2020-04-12 19:12:34,469-INFO: iter: 2080, lr: 0.000010, 'loss_cls': '0.214538', 'loss_bbox': '0.141871', 'loss_rpn_cls': '0.677379', 'loss_rpn_bbox': '0.026793', 'loss_mask': '0.478271', 'loss': '1.523065', time: 0.321, eta: 0:42:19 2020-04-12 19:12:40,747-INFO: iter: 2100, lr: 0.000010, 'loss_cls': '0.174576', 'loss_bbox': '0.115565', 'loss_rpn_cls': '0.676781', 'loss_rpn_bbox': '0.029479', 'loss_mask': '0.454077', 'loss': '1.558492', time: 0.317, eta: 0:41:42 2020-04-12 19:12:46,997-INFO: iter: 2120, lr: 0.000010, 'loss_cls': '0.239293', 'loss_bbox': '0.166249', 'loss_rpn_cls': '0.676117', 'loss_rpn_bbox': '0.031448', 'loss_mask': '0.497078', 'loss': '1.624302', time: 0.312, eta: 0:40:59 2020-04-12 19:12:53,227-INFO: iter: 2140, lr: 0.000010, 'loss_cls': '0.190785', 'loss_bbox': '0.125103', 'loss_rpn_cls': '0.676561', 'loss_rpn_bbox': '0.019331', 'loss_mask': '0.440254', 'loss': '1.455096', time: 0.312, eta: 0:40:56 2020-04-12 19:12:59,471-INFO: iter: 2160, lr: 0.000010, 'loss_cls': '0.199821', 'loss_bbox': '0.134545', 'loss_rpn_cls': '0.676524', 'loss_rpn_bbox': '0.049181', 'loss_mask': '0.485303', 'loss': '1.548986', time: 0.312, eta: 0:40:44 2020-04-12 19:13:05,716-INFO: iter: 2180, lr: 0.000010, 'loss_cls': '0.152130', 'loss_bbox': '0.067805', 'loss_rpn_cls': '0.676190', 'loss_rpn_bbox': '0.014714', 'loss_mask': '0.450225', 'loss': '1.360856', time: 0.311, eta: 0:40:35 2020-04-12 19:13:11,961-INFO: iter: 2200, lr: 0.000010, 'loss_cls': '0.237398', 'loss_bbox': '0.147018', 'loss_rpn_cls': '0.676997', 'loss_rpn_bbox': '0.032779', 'loss_mask': '0.469115', 'loss': '1.537581', time: 0.312, eta: 0:40:32 2020-04-12 19:13:11,961-INFO: Save model to output/mask_rcnn_r50_1x/2200. 2020-04-12 19:13:28,734-INFO: iter: 2220, lr: 0.000010, 'loss_cls': '0.247591', 'loss_bbox': '0.151754', 'loss_rpn_cls': '0.675307', 'loss_rpn_bbox': '0.027869', 'loss_mask': '0.489234', 'loss': '1.621261', time: 0.312, eta: 0:40:25 2020-04-12 19:13:34,943-INFO: iter: 2240, lr: 0.000010, 'loss_cls': '0.220039', 'loss_bbox': '0.134291', 'loss_rpn_cls': '0.676409', 'loss_rpn_bbox': '0.028589', 'loss_mask': '0.456556', 'loss': '1.518595', time: 0.309, eta: 0:40:00 2020-04-12 19:13:41,171-INFO: iter: 2260, lr: 0.000010, 'loss_cls': '0.173373', 'loss_bbox': '0.097001', 'loss_rpn_cls': '0.675442', 'loss_rpn_bbox': '0.021549', 'loss_mask': '0.461916', 'loss': '1.551025', time: 0.311, eta: 0:40:06 2020-04-12 19:13:47,397-INFO: iter: 2280, lr: 0.000010, 'loss_cls': '0.222645', 'loss_bbox': '0.139721', 'loss_rpn_cls': '0.675607', 'loss_rpn_bbox': '0.037248', 'loss_mask': '0.483418', 'loss': '1.611545', time: 0.311, eta: 0:40:03 2020-04-12 19:13:53,623-INFO: iter: 2300, lr: 0.000010, 'loss_cls': '0.176801', 'loss_bbox': '0.109012', 'loss_rpn_cls': '0.675436', 'loss_rpn_bbox': '0.022128', 'loss_mask': '0.471628', 'loss': '1.441870', time: 0.312, eta: 0:40:01 2020-04-12 19:13:59,857-INFO: iter: 2320, lr: 0.000010, 'loss_cls': '0.228998', 'loss_bbox': '0.149026', 'loss_rpn_cls': '0.676013', 'loss_rpn_bbox': '0.039883', 'loss_mask': '0.421429', 'loss': '1.554473', time: 0.313, eta: 0:40:04 2020-04-12 19:14:06,089-INFO: iter: 2340, lr: 0.000010, 'loss_cls': '0.210004', 'loss_bbox': '0.145116', 'loss_rpn_cls': '0.675229', 'loss_rpn_bbox': '0.031689', 'loss_mask': '0.476567', 'loss': '1.508428', time: 0.312, eta: 0:39:52 2020-04-12 19:14:12,325-INFO: iter: 2360, lr: 0.000010, 'loss_cls': '0.144831', 'loss_bbox': '0.071569', 'loss_rpn_cls': '0.675304', 'loss_rpn_bbox': '0.019925', 'loss_mask': '0.459881', 'loss': '1.353447', time: 0.313, eta: 0:39:52 2020-04-12 19:14:18,554-INFO: iter: 2380, lr: 0.000010, 'loss_cls': '0.170630', 'loss_bbox': '0.070795', 'loss_rpn_cls': '0.674430', 'loss_rpn_bbox': '0.014631', 'loss_mask': '0.470059', 'loss': '1.398263', time: 0.312, eta: 0:39:40 2020-04-12 19:14:24,853-INFO: iter: 2400, lr: 0.000010, 'loss_cls': '0.225324', 'loss_bbox': '0.147949', 'loss_rpn_cls': '0.675097', 'loss_rpn_bbox': '0.021562', 'loss_mask': '0.433219', 'loss': '1.498723', time: 0.311, eta: 0:39:25 2020-04-12 19:14:24,853-INFO: Save model to output/mask_rcnn_r50_1x/2400. 2020-04-12 19:14:40,921-INFO: iter: 2420, lr: 0.000010, 'loss_cls': '0.210051', 'loss_bbox': '0.145115', 'loss_rpn_cls': '0.674619', 'loss_rpn_bbox': '0.032646', 'loss_mask': '0.476620', 'loss': '1.502943', time: 0.312, eta: 0:39:22 2020-04-12 19:14:47,158-INFO: iter: 2440, lr: 0.000010, 'loss_cls': '0.219640', 'loss_bbox': '0.154375', 'loss_rpn_cls': '0.674558', 'loss_rpn_bbox': '0.037298', 'loss_mask': '0.432885', 'loss': '1.505762', time: 0.311, eta: 0:39:14 2020-04-12 19:14:53,400-INFO: iter: 2460, lr: 0.000010, 'loss_cls': '0.202968', 'loss_bbox': '0.115576', 'loss_rpn_cls': '0.673774', 'loss_rpn_bbox': '0.037199', 'loss_mask': '0.482314', 'loss': '1.546787', time: 0.312, eta: 0:39:12 2020-04-12 19:14:59,633-INFO: iter: 2480, lr: 0.000010, 'loss_cls': '0.209148', 'loss_bbox': '0.126762', 'loss_rpn_cls': '0.673283', 'loss_rpn_bbox': '0.026134', 'loss_mask': '0.381481', 'loss': '1.582597', time: 0.312, eta: 0:39:02 2020-04-12 19:15:05,872-INFO: iter: 2500, lr: 0.000010, 'loss_cls': '0.178465', 'loss_bbox': '0.070605', 'loss_rpn_cls': '0.674192', 'loss_rpn_bbox': '0.031799', 'loss_mask': '0.485181', 'loss': '1.453874', time: 0.311, eta: 0:38:52 2020-04-12 19:15:12,122-INFO: iter: 2520, lr: 0.000010, 'loss_cls': '0.152894', 'loss_bbox': '0.094793', 'loss_rpn_cls': '0.673193', 'loss_rpn_bbox': '0.029279', 'loss_mask': '0.415488', 'loss': '1.379925', time: 0.314, eta: 0:39:09 2020-04-12 19:15:18,374-INFO: iter: 2540, lr: 0.000010, 'loss_cls': '0.199300', 'loss_bbox': '0.109602', 'loss_rpn_cls': '0.674683', 'loss_rpn_bbox': '0.020928', 'loss_mask': '0.476535', 'loss': '1.422234', time: 0.314, eta: 0:39:01 2020-04-12 19:15:24,656-INFO: iter: 2560, lr: 0.000010, 'loss_cls': '0.223392', 'loss_bbox': '0.156912', 'loss_rpn_cls': '0.673052', 'loss_rpn_bbox': '0.038830', 'loss_mask': '0.489915', 'loss': '1.609865', time: 0.320, eta: 0:39:37 2020-04-12 19:15:30,904-INFO: iter: 2580, lr: 0.000010, 'loss_cls': '0.181098', 'loss_bbox': '0.110433', 'loss_rpn_cls': '0.672873', 'loss_rpn_bbox': '0.031388', 'loss_mask': '0.438347', 'loss': '1.469130', time: 0.313, eta: 0:38:41 2020-04-12 19:15:37,150-INFO: iter: 2600, lr: 0.000010, 'loss_cls': '0.206945', 'loss_bbox': '0.139660', 'loss_rpn_cls': '0.672801', 'loss_rpn_bbox': '0.026736', 'loss_mask': '0.424634', 'loss': '1.458906', time: 0.313, eta: 0:38:34 2020-04-12 19:15:37,150-INFO: Save model to output/mask_rcnn_r50_1x/2600. 2020-04-12 19:15:53,112-INFO: iter: 2620, lr: 0.000010, 'loss_cls': '0.242985', 'loss_bbox': '0.146154', 'loss_rpn_cls': '0.673803', 'loss_rpn_bbox': '0.048899', 'loss_mask': '0.514158', 'loss': '1.664126', time: 0.312, eta: 0:38:24 2020-04-12 19:15:59,347-INFO: iter: 2640, lr: 0.000010, 'loss_cls': '0.247756', 'loss_bbox': '0.134982', 'loss_rpn_cls': '0.672127', 'loss_rpn_bbox': '0.019858', 'loss_mask': '0.449085', 'loss': '1.558025', time: 0.312, eta: 0:38:18 2020-04-12 19:16:05,609-INFO: iter: 2660, lr: 0.000010, 'loss_cls': '0.251123', 'loss_bbox': '0.141473', 'loss_rpn_cls': '0.672773', 'loss_rpn_bbox': '0.019846', 'loss_mask': '0.444554', 'loss': '1.527069', time: 0.314, eta: 0:38:28 2020-04-12 19:16:11,860-INFO: iter: 2680, lr: 0.000010, 'loss_cls': '0.273818', 'loss_bbox': '0.175250', 'loss_rpn_cls': '0.672206', 'loss_rpn_bbox': '0.031371', 'loss_mask': '0.433439', 'loss': '1.582957', time: 0.312, eta: 0:38:00 2020-04-12 19:16:18,109-INFO: iter: 2700, lr: 0.000010, 'loss_cls': '0.196931', 'loss_bbox': '0.108791', 'loss_rpn_cls': '0.673146', 'loss_rpn_bbox': '0.028698', 'loss_mask': '0.449363', 'loss': '1.449443', time: 0.312, eta: 0:37:59 2020-04-12 19:16:24,363-INFO: iter: 2720, lr: 0.000010, 'loss_cls': '0.226481', 'loss_bbox': '0.148690', 'loss_rpn_cls': '0.671854', 'loss_rpn_bbox': '0.023539', 'loss_mask': '0.457820', 'loss': '1.553706', time: 0.313, eta: 0:37:56 2020-04-12 19:16:30,605-INFO: iter: 2740, lr: 0.000010, 'loss_cls': '0.098426', 'loss_bbox': '0.030988', 'loss_rpn_cls': '0.671802', 'loss_rpn_bbox': '0.017874', 'loss_mask': '0.466082', 'loss': '1.293685', time: 0.312, eta: 0:37:43 2020-04-12 19:16:36,870-INFO: iter: 2760, lr: 0.000010, 'loss_cls': '0.201074', 'loss_bbox': '0.127970', 'loss_rpn_cls': '0.673485', 'loss_rpn_bbox': '0.032176', 'loss_mask': '0.432475', 'loss': '1.523252', time: 0.312, eta: 0:37:39 2020-04-12 19:16:43,116-INFO: iter: 2780, lr: 0.000010, 'loss_cls': '0.238042', 'loss_bbox': '0.167331', 'loss_rpn_cls': '0.671684', 'loss_rpn_bbox': '0.020402', 'loss_mask': '0.443180', 'loss': '1.533122', time: 0.313, eta: 0:37:37 2020-04-12 19:16:49,372-INFO: iter: 2800, lr: 0.000010, 'loss_cls': '0.186766', 'loss_bbox': '0.093643', 'loss_rpn_cls': '0.671307', 'loss_rpn_bbox': '0.017021', 'loss_mask': '0.439003', 'loss': '1.481909', time: 0.314, eta: 0:37:41 2020-04-12 19:16:49,373-INFO: Save model to output/mask_rcnn_r50_1x/2800. 2020-04-12 19:17:04,875-INFO: iter: 2820, lr: 0.000010, 'loss_cls': '0.156072', 'loss_bbox': '0.066318', 'loss_rpn_cls': '0.672713', 'loss_rpn_bbox': '0.017551', 'loss_mask': '0.437765', 'loss': '1.393559', time: 0.311, eta: 0:37:16 2020-04-12 19:17:11,146-INFO: iter: 2840, lr: 0.000010, 'loss_cls': '0.137040', 'loss_bbox': '0.058395', 'loss_rpn_cls': '0.670958', 'loss_rpn_bbox': '0.015044', 'loss_mask': '0.451881', 'loss': '1.394194', time: 0.312, eta: 0:37:11 2020-04-12 19:17:17,393-INFO: iter: 2860, lr: 0.000010, 'loss_cls': '0.193442', 'loss_bbox': '0.128927', 'loss_rpn_cls': '0.671292', 'loss_rpn_bbox': '0.012919', 'loss_mask': '0.482171', 'loss': '1.474029', time: 0.313, eta: 0:37:13 2020-04-12 19:17:23,638-INFO: iter: 2880, lr: 0.000010, 'loss_cls': '0.223344', 'loss_bbox': '0.172523', 'loss_rpn_cls': '0.670703', 'loss_rpn_bbox': '0.043653', 'loss_mask': '0.404981', 'loss': '1.506812', time: 0.313, eta: 0:37:05 2020-04-12 19:17:29,906-INFO: iter: 2900, lr: 0.000010, 'loss_cls': '0.181952', 'loss_bbox': '0.080318', 'loss_rpn_cls': '0.672194', 'loss_rpn_bbox': '0.051024', 'loss_mask': '0.467330', 'loss': '1.408267', time: 0.314, eta: 0:37:12 2020-04-12 19:17:36,185-INFO: iter: 2920, lr: 0.000010, 'loss_cls': '0.259455', 'loss_bbox': '0.144422', 'loss_rpn_cls': '0.669862', 'loss_rpn_bbox': '0.019747', 'loss_mask': '0.470121', 'loss': '1.564603', time: 0.313, eta: 0:36:57 2020-04-12 19:17:42,466-INFO: iter: 2940, lr: 0.000010, 'loss_cls': '0.153529', 'loss_bbox': '0.071788', 'loss_rpn_cls': '0.671111', 'loss_rpn_bbox': '0.028569', 'loss_mask': '0.446193', 'loss': '1.444543', time: 0.314, eta: 0:36:56 2020-04-12 19:17:48,719-INFO: iter: 2960, lr: 0.000010, 'loss_cls': '0.156661', 'loss_bbox': '0.068588', 'loss_rpn_cls': '0.671329', 'loss_rpn_bbox': '0.031354', 'loss_mask': '0.381010', 'loss': '1.450279', time: 0.314, eta: 0:36:49 2020-04-12 19:17:55,008-INFO: iter: 2980, lr: 0.000010, 'loss_cls': '0.231161', 'loss_bbox': '0.158397', 'loss_rpn_cls': '0.670491', 'loss_rpn_bbox': '0.019284', 'loss_mask': '0.485176', 'loss': '1.505694', time: 0.316, eta: 0:36:56 2020-04-12 19:18:01,282-INFO: iter: 3000, lr: 0.000010, 'loss_cls': '0.192047', 'loss_bbox': '0.111483', 'loss_rpn_cls': '0.670459', 'loss_rpn_bbox': '0.049830', 'loss_mask': '0.473955', 'loss': '1.562895', time: 0.315, eta: 0:36:42 2020-04-12 19:18:01,283-INFO: Save model to output/mask_rcnn_r50_1x/3000. 2020-04-12 19:18:15,576-INFO: iter: 3020, lr: 0.000010, 'loss_cls': '0.236650', 'loss_bbox': '0.150820', 'loss_rpn_cls': '0.668947', 'loss_rpn_bbox': '0.032482', 'loss_mask': '0.419164', 'loss': '1.525955', time: 0.313, eta: 0:36:24 2020-04-12 19:18:21,845-INFO: iter: 3040, lr: 0.000010, 'loss_cls': '0.131159', 'loss_bbox': '0.064954', 'loss_rpn_cls': '0.669428', 'loss_rpn_bbox': '0.014129', 'loss_mask': '0.467175', 'loss': '1.360088', time: 0.313, eta: 0:36:16 2020-04-12 19:18:28,105-INFO: iter: 3060, lr: 0.000010, 'loss_cls': '0.172071', 'loss_bbox': '0.078265', 'loss_rpn_cls': '0.669719', 'loss_rpn_bbox': '0.032297', 'loss_mask': '0.478122', 'loss': '1.439916', time: 0.311, eta: 0:36:00 2020-04-12 19:18:34,414-INFO: iter: 3080, lr: 0.000010, 'loss_cls': '0.183305', 'loss_bbox': '0.102732', 'loss_rpn_cls': '0.671658', 'loss_rpn_bbox': '0.045485', 'loss_mask': '0.447850', 'loss': '1.520272', time: 0.314, eta: 0:36:13 2020-04-12 19:18:40,666-INFO: iter: 3100, lr: 0.000010, 'loss_cls': '0.229421', 'loss_bbox': '0.154446', 'loss_rpn_cls': '0.669505', 'loss_rpn_bbox': '0.039948', 'loss_mask': '0.419655', 'loss': '1.555703', time: 0.313, eta: 0:35:59 2020-04-12 19:18:46,930-INFO: iter: 3120, lr: 0.000010, 'loss_cls': '0.203260', 'loss_bbox': '0.140440', 'loss_rpn_cls': '0.667932', 'loss_rpn_bbox': '0.048598', 'loss_mask': '0.455281', 'loss': '1.521403', time: 0.315, eta: 0:36:08 2020-04-12 19:18:53,195-INFO: iter: 3140, lr: 0.000010, 'loss_cls': '0.143494', 'loss_bbox': '0.081887', 'loss_rpn_cls': '0.669260', 'loss_rpn_bbox': '0.013749', 'loss_mask': '0.390585', 'loss': '1.290240', time: 0.312, eta: 0:35:42 2020-04-12 19:18:59,470-INFO: iter: 3160, lr: 0.000010, 'loss_cls': '0.158234', 'loss_bbox': '0.078481', 'loss_rpn_cls': '0.669253', 'loss_rpn_bbox': '0.020251', 'loss_mask': '0.446877', 'loss': '1.449142', time: 0.314, eta: 0:35:47 2020-04-12 19:19:05,748-INFO: iter: 3180, lr: 0.000010, 'loss_cls': '0.189619', 'loss_bbox': '0.094126', 'loss_rpn_cls': '0.668860', 'loss_rpn_bbox': '0.026734', 'loss_mask': '0.439686', 'loss': '1.464205', time: 0.313, eta: 0:35:35 2020-04-12 19:19:12,042-INFO: iter: 3200, lr: 0.000010, 'loss_cls': '0.149705', 'loss_bbox': '0.068039', 'loss_rpn_cls': '0.668737', 'loss_rpn_bbox': '0.049870', 'loss_mask': '0.519142', 'loss': '1.415281', time: 0.316, eta: 0:35:47 2020-04-12 19:19:12,042-INFO: Save model to output/mask_rcnn_r50_1x/3200. 2020-04-12 19:19:25,801-INFO: iter: 3220, lr: 0.000010, 'loss_cls': '0.135449', 'loss_bbox': '0.064716', 'loss_rpn_cls': '0.669397', 'loss_rpn_bbox': '0.017670', 'loss_mask': '0.447311', 'loss': '1.342127', time: 0.311, eta: 0:35:11 2020-04-12 19:19:32,070-INFO: iter: 3240, lr: 0.000010, 'loss_cls': '0.199373', 'loss_bbox': '0.120718', 'loss_rpn_cls': '0.668347', 'loss_rpn_bbox': '0.032206', 'loss_mask': '0.457151', 'loss': '1.471775', time: 0.313, eta: 0:35:13 2020-04-12 19:19:38,313-INFO: iter: 3260, lr: 0.000010, 'loss_cls': '0.157522', 'loss_bbox': '0.080736', 'loss_rpn_cls': '0.667660', 'loss_rpn_bbox': '0.024746', 'loss_mask': '0.338962', 'loss': '1.352480', time: 0.311, eta: 0:34:58 2020-04-12 19:19:44,580-INFO: iter: 3280, lr: 0.000010, 'loss_cls': '0.195963', 'loss_bbox': '0.133155', 'loss_rpn_cls': '0.668406', 'loss_rpn_bbox': '0.029029', 'loss_mask': '0.447330', 'loss': '1.456097', time: 0.315, eta: 0:35:18 2020-04-12 19:19:50,847-INFO: iter: 3300, lr: 0.000010, 'loss_cls': '0.178845', 'loss_bbox': '0.103699', 'loss_rpn_cls': '0.667426', 'loss_rpn_bbox': '0.026891', 'loss_mask': '0.448315', 'loss': '1.420633', time: 0.313, eta: 0:34:54 2020-04-12 19:19:57,140-INFO: iter: 3320, lr: 0.000010, 'loss_cls': '0.231082', 'loss_bbox': '0.137732', 'loss_rpn_cls': '0.666830', 'loss_rpn_bbox': '0.020257', 'loss_mask': '0.423753', 'loss': '1.474663', time: 0.315, eta: 0:35:07 2020-04-12 19:20:03,444-INFO: iter: 3340, lr: 0.000010, 'loss_cls': '0.198445', 'loss_bbox': '0.110278', 'loss_rpn_cls': '0.668683', 'loss_rpn_bbox': '0.030330', 'loss_mask': '0.483810', 'loss': '1.531009', time: 0.317, eta: 0:35:10 2020-04-12 19:20:09,746-INFO: iter: 3360, lr: 0.000010, 'loss_cls': '0.189412', 'loss_bbox': '0.129265', 'loss_rpn_cls': '0.666723', 'loss_rpn_bbox': '0.041751', 'loss_mask': '0.404535', 'loss': '1.525172', time: 0.315, eta: 0:34:53 2020-04-12 19:20:16,026-INFO: iter: 3380, lr: 0.000010, 'loss_cls': '0.225269', 'loss_bbox': '0.141508', 'loss_rpn_cls': '0.668767', 'loss_rpn_bbox': '0.023126', 'loss_mask': '0.426346', 'loss': '1.453345', time: 0.314, eta: 0:34:41 2020-04-12 19:20:22,299-INFO: iter: 3400, lr: 0.000010, 'loss_cls': '0.156870', 'loss_bbox': '0.078222', 'loss_rpn_cls': '0.666369', 'loss_rpn_bbox': '0.031256', 'loss_mask': '0.468440', 'loss': '1.425375', time: 0.312, eta: 0:34:21 2020-04-12 19:20:22,299-INFO: Save model to output/mask_rcnn_r50_1x/3400. 2020-04-12 19:20:38,185-INFO: iter: 3420, lr: 0.000010, 'loss_cls': '0.171823', 'loss_bbox': '0.110642', 'loss_rpn_cls': '0.666645', 'loss_rpn_bbox': '0.037166', 'loss_mask': '0.488310', 'loss': '1.574501', time: 0.314, eta: 0:34:23 2020-04-12 19:20:44,429-INFO: iter: 3440, lr: 0.000010, 'loss_cls': '0.156654', 'loss_bbox': '0.078122', 'loss_rpn_cls': '0.666414', 'loss_rpn_bbox': '0.028579', 'loss_mask': '0.474610', 'loss': '1.431127', time: 0.312, eta: 0:34:06 2020-04-12 19:20:50,688-INFO: iter: 3460, lr: 0.000010, 'loss_cls': '0.176988', 'loss_bbox': '0.111378', 'loss_rpn_cls': '0.665635', 'loss_rpn_bbox': '0.023896', 'loss_mask': '0.447071', 'loss': '1.428934', time: 0.314, eta: 0:34:12 2020-04-12 19:20:56,952-INFO: iter: 3480, lr: 0.000010, 'loss_cls': '0.290263', 'loss_bbox': '0.204163', 'loss_rpn_cls': '0.666004', 'loss_rpn_bbox': '0.031194', 'loss_mask': '0.466734', 'loss': '1.697813', time: 0.315, eta: 0:34:13 2020-04-12 19:21:03,253-INFO: iter: 3500, lr: 0.000010, 'loss_cls': '0.203343', 'loss_bbox': '0.121325', 'loss_rpn_cls': '0.665510', 'loss_rpn_bbox': '0.025138', 'loss_mask': '0.462976', 'loss': '1.488306', time: 0.315, eta: 0:34:08 2020-04-12 19:21:09,537-INFO: iter: 3520, lr: 0.000010, 'loss_cls': '0.127693', 'loss_bbox': '0.068771', 'loss_rpn_cls': '0.666651', 'loss_rpn_bbox': '0.027291', 'loss_mask': '0.391470', 'loss': '1.267895', time: 0.314, eta: 0:33:53 2020-04-12 19:21:15,820-INFO: iter: 3540, lr: 0.000010, 'loss_cls': '0.144239', 'loss_bbox': '0.076892', 'loss_rpn_cls': '0.665802', 'loss_rpn_bbox': '0.032333', 'loss_mask': '0.451535', 'loss': '1.458961', time: 0.316, eta: 0:34:04 2020-04-12 19:21:22,191-INFO: iter: 3560, lr: 0.000010, 'loss_cls': '0.156625', 'loss_bbox': '0.078049', 'loss_rpn_cls': '0.665125', 'loss_rpn_bbox': '0.023223', 'loss_mask': '0.465593', 'loss': '1.436662', time: 0.320, eta: 0:34:19 2020-04-12 19:21:28,550-INFO: iter: 3580, lr: 0.000010, 'loss_cls': '0.189123', 'loss_bbox': '0.117465', 'loss_rpn_cls': '0.665995', 'loss_rpn_bbox': '0.020830', 'loss_mask': '0.451084', 'loss': '1.519307', time: 0.315, eta: 0:33:43 2020-04-12 19:21:34,816-INFO: iter: 3600, lr: 0.000010, 'loss_cls': '0.170753', 'loss_bbox': '0.096753', 'loss_rpn_cls': '0.664982', 'loss_rpn_bbox': '0.032039', 'loss_mask': '0.466287', 'loss': '1.449429', time: 0.314, eta: 0:33:27 2020-04-12 19:21:34,816-INFO: Save model to output/mask_rcnn_r50_1x/3600. 2020-04-12 19:21:47,875-INFO: iter: 3620, lr: 0.000010, 'loss_cls': '0.201908', 'loss_bbox': '0.121842', 'loss_rpn_cls': '0.666509', 'loss_rpn_bbox': '0.037458', 'loss_mask': '0.432538', 'loss': '1.488950', time: 0.315, eta: 0:33:30 2020-04-12 19:21:54,126-INFO: iter: 3640, lr: 0.000010, 'loss_cls': '0.189705', 'loss_bbox': '0.084598', 'loss_rpn_cls': '0.664819', 'loss_rpn_bbox': '0.022228', 'loss_mask': '0.464062', 'loss': '1.509298', time: 0.313, eta: 0:33:10 2020-04-12 19:22:00,388-INFO: iter: 3660, lr: 0.000010, 'loss_cls': '0.227702', 'loss_bbox': '0.142128', 'loss_rpn_cls': '0.664515', 'loss_rpn_bbox': '0.050485', 'loss_mask': '0.468259', 'loss': '1.564367', time: 0.312, eta: 0:33:00 2020-04-12 19:22:06,645-INFO: iter: 3680, lr: 0.000010, 'loss_cls': '0.142362', 'loss_bbox': '0.048927', 'loss_rpn_cls': '0.667248', 'loss_rpn_bbox': '0.025105', 'loss_mask': '0.476718', 'loss': '1.347363', time: 0.313, eta: 0:32:55 2020-04-12 19:22:12,902-INFO: iter: 3700, lr: 0.000010, 'loss_cls': '0.192876', 'loss_bbox': '0.126948', 'loss_rpn_cls': '0.665755', 'loss_rpn_bbox': '0.018563', 'loss_mask': '0.469767', 'loss': '1.434391', time: 0.313, eta: 0:32:54 2020-04-12 19:22:19,168-INFO: iter: 3720, lr: 0.000010, 'loss_cls': '0.177404', 'loss_bbox': '0.105137', 'loss_rpn_cls': '0.663700', 'loss_rpn_bbox': '0.028097', 'loss_mask': '0.483050', 'loss': '1.484103', time: 0.313, eta: 0:32:47 2020-04-12 19:22:25,450-INFO: iter: 3740, lr: 0.000010, 'loss_cls': '0.117043', 'loss_bbox': '0.044262', 'loss_rpn_cls': '0.664751', 'loss_rpn_bbox': '0.035887', 'loss_mask': '0.462314', 'loss': '1.324282', time: 0.312, eta: 0:32:32 2020-04-12 19:22:31,722-INFO: iter: 3760, lr: 0.000010, 'loss_cls': '0.131780', 'loss_bbox': '0.076443', 'loss_rpn_cls': '0.665389', 'loss_rpn_bbox': '0.027291', 'loss_mask': '0.434567', 'loss': '1.396293', time: 0.313, eta: 0:32:34 2020-04-12 19:22:37,985-INFO: iter: 3780, lr: 0.000010, 'loss_cls': '0.191903', 'loss_bbox': '0.110940', 'loss_rpn_cls': '0.662837', 'loss_rpn_bbox': '0.038824', 'loss_mask': '0.459523', 'loss': '1.501657', time: 0.315, eta: 0:32:38 2020-04-12 19:22:44,232-INFO: iter: 3800, lr: 0.000010, 'loss_cls': '0.196803', 'loss_bbox': '0.137776', 'loss_rpn_cls': '0.664525', 'loss_rpn_bbox': '0.033289', 'loss_mask': '0.407999', 'loss': '1.494284', time: 0.312, eta: 0:32:15 2020-04-12 19:22:44,233-INFO: Save model to output/mask_rcnn_r50_1x/3800. 2020-04-12 19:22:57,129-INFO: iter: 3820, lr: 0.000010, 'loss_cls': '0.148278', 'loss_bbox': '0.068216', 'loss_rpn_cls': '0.662876', 'loss_rpn_bbox': '0.029901', 'loss_mask': '0.503529', 'loss': '1.393205', time: 0.313, eta: 0:32:16 2020-04-12 19:23:03,403-INFO: iter: 3840, lr: 0.000010, 'loss_cls': '0.146724', 'loss_bbox': '0.093352', 'loss_rpn_cls': '0.662710', 'loss_rpn_bbox': '0.032662', 'loss_mask': '0.379453', 'loss': '1.423286', time: 0.314, eta: 0:32:13 2020-04-12 19:23:09,674-INFO: iter: 3860, lr: 0.000010, 'loss_cls': '0.132706', 'loss_bbox': '0.061472', 'loss_rpn_cls': '0.662994', 'loss_rpn_bbox': '0.014324', 'loss_mask': '0.423745', 'loss': '1.296894', time: 0.314, eta: 0:32:06 2020-04-12 19:23:15,930-INFO: iter: 3880, lr: 0.000010, 'loss_cls': '0.134696', 'loss_bbox': '0.070502', 'loss_rpn_cls': '0.665815', 'loss_rpn_bbox': '0.056177', 'loss_mask': '0.460671', 'loss': '1.426115', time: 0.312, eta: 0:31:49 2020-04-12 19:23:22,202-INFO: iter: 3900, lr: 0.000010, 'loss_cls': '0.183502', 'loss_bbox': '0.118884', 'loss_rpn_cls': '0.663020', 'loss_rpn_bbox': '0.052802', 'loss_mask': '0.374709', 'loss': '1.402453', time: 0.312, eta: 0:31:44 2020-04-12 19:23:28,539-INFO: iter: 3920, lr: 0.000010, 'loss_cls': '0.209540', 'loss_bbox': '0.136324', 'loss_rpn_cls': '0.661844', 'loss_rpn_bbox': '0.039212', 'loss_mask': '0.446309', 'loss': '1.492800', time: 0.318, eta: 0:32:12 2020-04-12 19:23:34,881-INFO: iter: 3940, lr: 0.000010, 'loss_cls': '0.135176', 'loss_bbox': '0.067890', 'loss_rpn_cls': '0.660461', 'loss_rpn_bbox': '0.026780', 'loss_mask': '0.384685', 'loss': '1.238222', time: 0.316, eta: 0:31:57 2020-04-12 19:23:41,171-INFO: iter: 3960, lr: 0.000010, 'loss_cls': '0.188297', 'loss_bbox': '0.092907', 'loss_rpn_cls': '0.661603', 'loss_rpn_bbox': '0.039935', 'loss_mask': '0.468302', 'loss': '1.429802', time: 0.312, eta: 0:31:24 2020-04-12 19:23:47,446-INFO: iter: 3980, lr: 0.000010, 'loss_cls': '0.161779', 'loss_bbox': '0.112419', 'loss_rpn_cls': '0.663066', 'loss_rpn_bbox': '0.023498', 'loss_mask': '0.353826', 'loss': '1.348194', time: 0.314, eta: 0:31:28 2020-04-12 19:23:53,710-INFO: iter: 4000, lr: 0.000010, 'loss_cls': '0.116867', 'loss_bbox': '0.071873', 'loss_rpn_cls': '0.661855', 'loss_rpn_bbox': '0.037850', 'loss_mask': '0.423442', 'loss': '1.362958', time: 0.315, eta: 0:31:27 2020-04-12 19:23:53,710-INFO: Save model to output/mask_rcnn_r50_1x/4000. 2020-04-12 19:24:06,860-INFO: iter: 4020, lr: 0.000010, 'loss_cls': '0.149067', 'loss_bbox': '0.083953', 'loss_rpn_cls': '0.661271', 'loss_rpn_bbox': '0.048802', 'loss_mask': '0.468825', 'loss': '1.523677', time: 0.314, eta: 0:31:19 2020-04-12 19:24:13,129-INFO: iter: 4040, lr: 0.000010, 'loss_cls': '0.122440', 'loss_bbox': '0.063576', 'loss_rpn_cls': '0.661424', 'loss_rpn_bbox': '0.028700', 'loss_mask': '0.434025', 'loss': '1.353182', time: 0.314, eta: 0:31:12 2020-04-12 19:24:19,410-INFO: iter: 4060, lr: 0.000010, 'loss_cls': '0.185411', 'loss_bbox': '0.105037', 'loss_rpn_cls': '0.661409', 'loss_rpn_bbox': '0.018866', 'loss_mask': '0.433659', 'loss': '1.444223', time: 0.314, eta: 0:31:05 2020-04-12 19:24:25,679-INFO: iter: 4080, lr: 0.000010, 'loss_cls': '0.211064', 'loss_bbox': '0.114713', 'loss_rpn_cls': '0.661286', 'loss_rpn_bbox': '0.020472', 'loss_mask': '0.447985', 'loss': '1.529290', time: 0.314, eta: 0:30:56 2020-04-12 19:24:31,932-INFO: iter: 4100, lr: 0.000010, 'loss_cls': '0.155372', 'loss_bbox': '0.108112', 'loss_rpn_cls': '0.662345', 'loss_rpn_bbox': '0.029405', 'loss_mask': '0.430589', 'loss': '1.333557', time: 0.312, eta: 0:30:38 2020-04-12 19:24:38,183-INFO: iter: 4120, lr: 0.000010, 'loss_cls': '0.244481', 'loss_bbox': '0.165366', 'loss_rpn_cls': '0.660865', 'loss_rpn_bbox': '0.031774', 'loss_mask': '0.445879', 'loss': '1.541602', time: 0.314, eta: 0:30:44 请问这是什么原因造成的，怎么修改，谢谢