search

open-mmlab / mmdetection

OpenMMLab Detection Toolbox and Benchmark
https://mmdetection.readthedocs.io
Apache License 2.0
29.21k stars 9.4k forks source link

Possible bug: AssertionError: The length of results is not equal to the dataset len #2971

Closed deepaksinghcv closed 4 years ago

deepaksinghcv commented 4 years ago

Describe the bug I'm trying to train MS_RCNN on cityscapes. I converted cityscapes to coco format and created a config file and modified the necessary parameters. The model trains and validates, but while displaying the evaluated result it throws: AssertionError: The length of results is not equal to the dataset len: 493 != 500

Reproduction

  1. What command or script did you run?

    ./tools/dist_train.sh configs/ms_rcnn/ms_rcnn_x101_32x4d_fpn_1x_cityscapes_new.py 4

  2. Did you make any modifications on the code or config? Did you understand what you have modified? I have added a config file in configs/ms_rcnn directory.

  3. What dataset did you use? Cityscapes but in COCO format. Environment

  4. Please run python mmdet/utils/collect_env.py to collect necessary environment infomation and paste it here.

    
sys.platform: linux
Python: 3.7.7 (default, May  7 2020, 21:25:33) [GCC 7.3.0]
CUDA available: True
CUDA_HOME: /usr/local/cuda-10.0
NVCC: Cuda compilation tools, release 10.0, V10.0.130
GPU 0,1,2,3: GeForce GTX 1080 Ti
GCC: gcc (Ubuntu 5.5.0-12ubuntu1~16.04) 5.5.0 20171010
PyTorch: 1.5.0
PyTorch compiling details: PyTorch built with:
- GCC 7.3
- C++ Version: 201402
- Intel(R) Math Kernel Library Version 2020.0.1 Product Build 20200208 for Intel(R) 64 architecture applications
- Intel(R) MKL-DNN v0.21.1 (Git Hash 7d2fd500bc78936d1d648ca713b901012f470dbc)
- OpenMP 201511 (a.k.a. OpenMP 4.5)
- NNPACK is enabled
- CPU capability usage: AVX2
- CUDA Runtime 10.2
- NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_37,code=compute_37
- CuDNN 7.6.5
- Magma 2.5.2
- Build settings: BLAS=MKL, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -fopenmp -DNDEBUG -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DUSE_INTERNAL_THREADPOOL_IMPL -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_STATIC_DISPATCH=OFF,

TorchVision: 0.6.0a0+82fd1c8 OpenCV: 4.2.0 MMCV: 0.5.1 MMDetection: 2.0.0+c802b17 MMDetection Compiler: GCC 5.5 MMDetection CUDA Compiler: 10.2

**Error traceback**
If applicable, paste the error trackback here.

(open-mmlab) dksingh@gnode17:~/sandboxes/msrcnn_coco/mmdetection$ ./tools/dist_train.sh configs/ms_rcnn/ms_rcnn_x101_32x4d_fpn_1x_cityscapes_new.py 4

Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.

2020-06-10 12:20:03,994 - mmdet - INFO - Environment info:

sys.platform: linux Python: 3.7.7 (default, May 7 2020, 21:25:33) [GCC 7.3.0] CUDA available: True CUDA_HOME: /usr/local/cuda-10.0 NVCC: Cuda compilation tools, release 10.0, V10.0.130 GPU 0,1,2,3: GeForce GTX 1080 Ti GCC: gcc (Ubuntu 5.5.0-12ubuntu1~16.04) 5.5.0 20171010 PyTorch: 1.5.0 PyTorch compiling details: PyTorch built with:

TorchVision: 0.6.0a0+82fd1c8 OpenCV: 4.2.0 MMCV: 0.5.1 MMDetection: 2.0.0+c802b17 MMDetection Compiler: GCC 5.5 MMDetection CUDA Compiler: 10.2

2020-06-10 12:20:03,994 - mmdet - INFO - Distributed training: True 2020-06-10 12:20:03,996 - mmdet - INFO - Config: classes=('person', 'rider', 'car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle') model=dict( type='MaskScoringRCNN', pretrained='open-mmlab://resnext101_32x4d', backbone=dict( type='ResNeXt', depth=101, num_stages=4, out_indices=(0, 1, 2, 3), frozen_stages=1, norm_cfg=dict( type='BN', requires_grad=True), norm_eval=True, style='pytorch', groups=32, base_width=4), neck=dict( type='FPN', in_channels=[256, 512, 1024, 2048], out_channels=256, num_outs=5), rpn_head=dict( type='RPNHead', in_channels=256, feat_channels=256, anchor_generator=dict( type='AnchorGenerator', scales=[8], ratios=[0.5, 1.0, 2.0], strides=[4, 8, 16, 32, 64]), bbox_coder=dict( type='DeltaXYWHBBoxCoder', target_means=[0.0, 0.0, 0.0, 0.0], target_stds=[1.0, 1.0, 1.0, 1.0]), loss_cls=dict( type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0), loss_bbox=dict( type='L1Loss', loss_weight=1.0)), roi_head=dict( type='MaskScoringRoIHead', bbox_roi_extractor=dict( type='SingleRoIExtractor', roi_layer=dict( type='RoIAlign', out_size=7, sample_num=0), out_channels=256, featmap_strides=[4, 8, 16, 32]), bbox_head=dict( type='Shared2FCBBoxHead', in_channels=256, fc_out_channels=1024, roi_feat_size=7, num_classes=8, bbox_coder=dict( type='DeltaXYWHBBoxCoder', target_means=[0.0, 0.0, 0.0, 0.0], target_stds=[0.1, 0.1, 0.2, 0.2]), reg_class_agnostic=False, loss_cls=dict( type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0), loss_bbox=dict( type='L1Loss', loss_weight=1.0)), mask_roi_extractor=dict( type='SingleRoIExtractor', roi_layer=dict( type='RoIAlign', out_size=14, sample_num=0), out_channels=256, featmap_strides=[4, 8, 16, 32]), mask_head=dict( type='FCNMaskHead', num_convs=4, in_channels=256, conv_out_channels=256, num_classes=8, loss_mask=dict( type='CrossEntropyLoss', use_mask=True, loss_weight=1.0)), mask_iou_head=dict( type='MaskIoUHead', num_convs=4, num_fcs=2, roi_feat_size=14, in_channels=256, conv_out_channels=256, fc_out_channels=1024, num_classes=8))) train_cfg=dict( rpn=dict( assigner=dict( type='MaxIoUAssigner', pos_iou_thr=0.7, neg_iou_thr=0.3, min_pos_iou=0.3, match_low_quality=True, ignore_iof_thr=-1), sampler=dict( type='RandomSampler', num=256, pos_fraction=0.5, neg_pos_ub=-1, add_gt_as_proposals=False), allowed_border=-1, pos_weight=-1, debug=False), rpn_proposal=dict( nms_across_levels=False, nms_pre=2000, nms_post=1000, max_num=1000, nms_thr=0.7, min_bbox_size=0), rcnn=dict( assigner=dict( type='MaxIoUAssigner', pos_iou_thr=0.5, neg_iou_thr=0.5, min_pos_iou=0.5, match_low_quality=True, ignore_iof_thr=-1), sampler=dict( type='RandomSampler', num=512, pos_fraction=0.25, neg_pos_ub=-1, add_gt_as_proposals=True), mask_size=28, pos_weight=-1, debug=False, mask_thr_binary=0.5)) test_cfg=dict( rpn=dict( nms_across_levels=False, nms_pre=1000, nms_post=1000, max_num=1000, nms_thr=0.7, min_bbox_size=0), rcnn=dict( score_thr=0.05, nms=dict( type='nms', iou_thr=0.5), max_per_img=100, mask_thr_binary=0.5)) dataset_type='CityscapesDataset' data_root='data/cityscapes/' img_norm_cfg=dict( mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) train_pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', with_bbox=True, with_mask=True), dict(type='Resize', img_scale=(1333, 800), keep_ratio=True), dict(type='RandomFlip', flip_ratio=0.5), dict(type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])] test_pipeline=[ dict(type='LoadImageFromFile'), dict(type='MultiScaleFlipAug', img_scale=(1333, 800), flip=False, transforms=[ dict(type='Resize', keep_ratio=True), dict(type='RandomFlip'), dict(type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='ImageToTensor', keys=['img']), dict(type='Collect', keys=['img'])])] data=dict( samples_per_gpu=2, workers_per_gpu=2, train=dict( classes=('person', 'rider', 'car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle'), type='CityscapesDataset', ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_train.json', img_prefix='data/cityscapes/leftImg8bit/train', pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', with_bbox=True, with_mask=True), dict(type='Resize', img_scale=(1333, 800), keep_ratio=True), dict(type='RandomFlip', flip_ratio=0.5), dict(type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])]), val=dict( classes=('person', 'rider', 'car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle'), type='CityscapesDataset', ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json', img_prefix='data/cityscapes/leftImg8bit/val', pipeline=[ dict(type='LoadImageFromFile'), dict(type='MultiScaleFlipAug', img_scale=(1333, 800), flip=False, transforms=[ dict(type='Resize', keep_ratio=True), dict(type='RandomFlip'), dict(type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='ImageToTensor', keys=['img']), dict(type='Collect', keys=['img'])])]), test=dict( classes=('person', 'rider', 'car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle'), type='CityscapesDataset', ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json', img_prefix='data/cityscapes/leftImg8bit/val', pipeline=[ dict(type='LoadImageFromFile'), dict(type='MultiScaleFlipAug', img_scale=(1333, 800), flip=False, transforms=[ dict(type='Resize', keep_ratio=True), dict(type='RandomFlip'), dict(type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='ImageToTensor', keys=['img']), dict(type='Collect', keys=['img'])])])) evaluation=dict( interval=1, metric=['bbox', 'segm']) optimizer=dict( type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001) optimizer_config=dict( grad_clip=None) lr_config=dict( policy='step', warmup='linear', warmup_iters=500, warmup_ratio=0.001, step=[8, 11]) total_epochs=12 checkpoint_config=dict( interval=1) log_config=dict( interval=50, hooks=[ dict(type='TextLoggerHook')]) dist_params=dict( backend='nccl') log_level='INFO' load_from=None resume_from=None workflow=[('train', 1)] work_dir='/ssd_scratch/cvit/dksingh/mmdetection_logs/ms_rcnn_x101_32x4d_fpn_1x_cityscapes_new/' gpu_ids=range(0, 1) 2020-06-10 12:20:07,451 - root - INFO - load model from: open-mmlab://resnext101_32x4d loading annotations into memory... loading annotations into memory... loading annotations into memory... loading annotations into memory... Done (t=1.49s) creating index... index created! Done (t=1.43s) creating index... Done (t=1.43s) creating index... index created! Done (t=1.45s) creating index... index created! index created! loading annotations into memory... loading annotations into memory... Done (t=0.15s) creating index... index created! 2020-06-10 12:20:16,229 - mmdet - INFO - Start running, host: dksingh@gnode17, work_dir: /ssd_scratch/cvit/dksingh/mmdetection_logs/ms_rcnn_x101_32x4d_fpn_1x_cityscapes_new 2020-06-10 12:20:16,229 - mmdet - INFO - workflow: [('train', 1)], max: 12 epochs loading annotations into memory... loading annotations into memory... Done (t=0.19s) creating index... index created! Done (t=0.20s) creating index... index created! Done (t=0.19s) creating index... index created! 2020-06-10 12:23:51,011 - mmdet - INFO - Epoch [1][50/371] lr: 0.00198, eta: 5:15:07, time: 4.295, data_time: 0.936, memory: 6724, loss_rpn_cls: 0.5591, loss_rpn_bbox: 0.1505, loss_cls: 0.5727, acc: 87.2520, loss_bbox: 0.0844, loss_mask: 0.7708, loss_mask_iou: 0.0221, loss: 2.1596 2020-06-10 12:26:56,695 - mmdet - INFO - Epoch [1][100/371] lr: 0.00398, eta: 4:50:25, time: 3.713, data_time: 0.418, memory: 6866, loss_rpn_cls: 0.2295, loss_rpn_bbox: 0.1438, loss_cls: 0.4238, acc: 89.9307, loss_bbox: 0.2964, loss_mask: 0.6281, loss_mask_iou: 0.0134, loss: 1.7350 2020-06-10 12:30:12,380 - mmdet - INFO - Epoch [1][150/371] lr: 0.00597, eta: 4:44:55, time: 3.913, data_time: 0.502, memory: 6866, loss_rpn_cls: 0.1433, loss_rpn_bbox: 0.1500, loss_cls: 0.3553, acc: 90.2085, loss_bbox: 0.2789, loss_mask: 0.6078, loss_mask_iou: 0.0170, loss: 1.5523 2020-06-10 12:33:25,925 - mmdet - INFO - Epoch [1][200/371] lr: 0.00797, eta: 4:39:47, time: 3.871, data_time: 0.461, memory: 6958, loss_rpn_cls: 0.1030, loss_rpn_bbox: 0.1405, loss_cls: 0.4616, acc: 85.2173, loss_bbox: 0.5360, loss_mask: 0.5672, loss_mask_iou: 0.0132, loss: 1.8215 2020-06-10 12:36:48,480 - mmdet - INFO - Epoch [1][250/371] lr: 0.00997, eta: 4:37:56, time: 4.051, data_time: 0.459, memory: 6958, loss_rpn_cls: 0.0828, loss_rpn_bbox: 0.1345, loss_cls: 0.4820, acc: 83.5576, loss_bbox: 0.6226, loss_mask: 0.5180, loss_mask_iou: 0.0144, loss: 1.8544 2020-06-10 12:40:07,226 - mmdet - INFO - Epoch [1][300/371] lr: 0.01197, eta: 4:34:42, time: 3.975, data_time: 0.424, memory: 6958, loss_rpn_cls: 0.0654, loss_rpn_bbox: 0.1189, loss_cls: 0.4370, acc: 84.3989, loss_bbox: 0.6138, loss_mask: 0.4824, loss_mask_iou: 0.0110, loss: 1.7284 2020-06-10 12:43:30,509 - mmdet - INFO - Epoch [1][350/371] lr: 0.01397, eta: 4:32:20, time: 4.065, data_time: 0.480, memory: 6958, loss_rpn_cls: 0.0647, loss_rpn_bbox: 0.1184, loss_cls: 0.4261, acc: 84.9155, loss_bbox: 0.5795, loss_mask: 0.4718, loss_mask_iou: 0.0130, loss: 1.6735 [>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 496/493, 1.5 task/s, elapsed: 336s, ETA: -1s

loading annotations into memory... Done (t=0.16s) creating index... index created! loading annotations into memory... Done (t=0.18s) creating index... index created! Traceback (most recent call last): File "./tools/train.py", line 159, in main() File "./tools/train.py", line 155, in main meta=meta) File "/home/dksingh/sandboxes/msrcnn_coco/mmdetection/mmdet/apis/train.py", line 165, in train_detector runner.run(data_loaders, cfg.workflow, cfg.total_epochs) File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/runner.py", line 383, in run epoch_runner(data_loaders[i], kwargs) File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/runner.py", line 292, in train self.call_hook('after_train_epoch') File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/runner.py", line 245, in call_hook getattr(hook, fn_name)(self) File "/home/dksingh/sandboxes/msrcnn_coco/mmdetection/mmdet/core/evaluation/eval_hooks.py", line 74, in after_train_epoch self.evaluate(runner, results) File "/home/dksingh/sandboxes/msrcnn_coco/mmdetection/mmdet/core/evaluation/eval_hooks.py", line 32, in evaluate results, logger=runner.logger, self.eval_kwargs) File "/home/dksingh/sandboxes/msrcnn_coco/mmdetection/mmdet/datasets/cityscapes.py", line 232, in evaluate classwise, proposal_nums, iou_thrs)) File "/home/dksingh/sandboxes/msrcnn_coco/mmdetection/mmdet/datasets/coco.py", line 349, in evaluate result_files, tmp_dir = self.format_results(results, jsonfile_prefix) File "/home/dksingh/sandboxes/msrcnn_coco/mmdetection/mmdet/datasets/coco.py", line 303, in format_results format(len(results), len(self))) AssertionError: The length of results is not equal to the dataset len: 493 != 500 Traceback (most recent call last): File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/runpy.py", line 193, in _run_module_as_main "main", mod_spec) File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/torch/distributed/launch.py", line 263, in main() File "/home/dksingh/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/torch/distributed/launch.py", line 259, in main cmd=cmd) subprocess.CalledProcessError: Command '['/home/dksingh/anaconda3/envs/open-mmlab/bin/python', '-u', './tools/train.py', '--local_rank=3', 'configs/ms_rcnn/ms_rcnn_x101_32x4d_fpn_1x_cityscapes_new.py', '--launcher', 'pytorch']' returned non-zero exit status 1.

deepaksinghcv commented 4 years ago

Any suggestions on what should be done to fix this issue.

deepaksinghcv commented 4 years ago

I tried executing the same with v2.0.0,. I'm still facing the same issue.

deepaksinghcv commented 4 years ago

I found the mistake. I have converted cityscapes to coco. But in the config file I have specified dataset_type='CityscapesDataset', when I changed it to dataset_type='CocoDataset' it performed the validation. It worked on v2.0.0

Note: Cityscapes has 500 val images, but the evaluation shows 496/493 images. I'm not closing the issue because the length are still unequal.

yhcao6 commented 4 years ago

Hi, sorry for the late reply, could you show me your config?

deepaksinghcv commented 4 years ago

I wanted to train on cityscapes, so i converted cityscapes to coco format. I have mentioned the path to the annotation files. I have mentioned the classes.

During validation it shows 493 images but evaluates 496.

classes=('person','rider','car','truck','bus','train','motorcycle','bicycle')
# classes='./cityscapes_classes.txt'
model=dict(
    type='MaskScoringRCNN',
    pretrained='open-mmlab://resnext101_32x4d',
    backbone=dict(
        type='ResNeXt',
        depth=101,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        norm_cfg=dict(
            type='BN',
            requires_grad=True),
        norm_eval=True,
        style='pytorch',
        groups=32,
        base_width=4),
    neck=dict(
        type='FPN',
        in_channels=[256, 512, 1024, 2048],
        out_channels=256,
        num_outs=5),
    rpn_head=dict(
        type='RPNHead',
        in_channels=256,
        feat_channels=256,
        anchor_generator=dict(
            type='AnchorGenerator',
            scales=[8],
            ratios=[0.5, 1.0, 2.0],
            strides=[4, 8, 16, 32, 64]),
        bbox_coder=dict(
            type='DeltaXYWHBBoxCoder',
            target_means=[0.0, 0.0, 0.0, 0.0],
            target_stds=[1.0, 1.0, 1.0, 1.0]),
        loss_cls=dict(
            type='CrossEntropyLoss',
            use_sigmoid=True,
            loss_weight=1.0),
        loss_bbox=dict(
            type='L1Loss',
            loss_weight=1.0)),
    roi_head=dict(
        type='MaskScoringRoIHead',
        bbox_roi_extractor=dict(
            type='SingleRoIExtractor',
            roi_layer=dict(
                type='RoIAlign',
                out_size=7,
                sample_num=0),
            out_channels=256,
            featmap_strides=[4, 8, 16, 32]),
        bbox_head=dict(
            type='Shared2FCBBoxHead',
            in_channels=256,
            fc_out_channels=1024,
            roi_feat_size=7,
            num_classes=8,
            bbox_coder=dict(
                type='DeltaXYWHBBoxCoder',
                target_means=[0.0, 0.0, 0.0, 0.0],
                target_stds=[0.1, 0.1, 0.2, 0.2]),
            reg_class_agnostic=False,
            loss_cls=dict(
                type='CrossEntropyLoss',
                use_sigmoid=False,
                loss_weight=1.0),
            loss_bbox=dict(
                type='L1Loss',
                loss_weight=1.0)),
        mask_roi_extractor=dict(
            type='SingleRoIExtractor',
            roi_layer=dict(
                type='RoIAlign',
                out_size=14,
                sample_num=0),
            out_channels=256,
            featmap_strides=[4, 8, 16, 32]),
        mask_head=dict(
            type='FCNMaskHead',
            num_convs=4,
            in_channels=256,
            conv_out_channels=256,
            num_classes=8,
            loss_mask=dict(
                type='CrossEntropyLoss',
                use_mask=True,
                loss_weight=1.0)),
        mask_iou_head=dict(
            type='MaskIoUHead',
            num_convs=4,
            num_fcs=2,
            roi_feat_size=14,
            in_channels=256,
            conv_out_channels=256,
            fc_out_channels=1024,
            num_classes=8)))
train_cfg=dict(
    rpn=dict(
        assigner=dict(
            type='MaxIoUAssigner',
            pos_iou_thr=0.7,
            neg_iou_thr=0.3,
            min_pos_iou=0.3,
            match_low_quality=True,
            ignore_iof_thr=-1),
        sampler=dict(
            type='RandomSampler',
            num=256,
            pos_fraction=0.5,
            neg_pos_ub=-1,
            add_gt_as_proposals=False),
        allowed_border=-1,
        pos_weight=-1,
        debug=False),
    rpn_proposal=dict(
        nms_across_levels=False,
        nms_pre=2000,
        nms_post=1000,
        max_num=1000,
        nms_thr=0.7,
        min_bbox_size=0),
    rcnn=dict(
        assigner=dict(
            type='MaxIoUAssigner',
            pos_iou_thr=0.5,
            neg_iou_thr=0.5,
            min_pos_iou=0.5,
            match_low_quality=True,
            ignore_iof_thr=-1),
        sampler=dict(
            type='RandomSampler',
            num=512,
            pos_fraction=0.25,
            neg_pos_ub=-1,
            add_gt_as_proposals=True),
        mask_size=28,
        pos_weight=-1,
        debug=False,
        mask_thr_binary=0.5))
test_cfg=dict(
    rpn=dict(
        nms_across_levels=False,
        nms_pre=1000,
        nms_post=1000,
        max_num=1000,
        nms_thr=0.7,
        min_bbox_size=0),
    rcnn=dict(
        score_thr=0.05,
        nms=dict(
            type='nms',
            iou_thr=0.5),
        max_per_img=100,
        mask_thr_binary=0.5))
dataset_type='CocoDataset'
data_root='data/cityscapes'
img_norm_cfg=dict(
    mean=[123.675, 116.28, 103.53],
    std=[58.395, 57.12, 57.375],
    to_rgb=True)
train_pipeline=[
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations',
        with_bbox=True,
        with_mask=True),
    dict(type='Resize',
        img_scale=(1333, 800),
        keep_ratio=True),
    dict(type='RandomFlip',
        flip_ratio=0.5),
    dict(type='Normalize',
        mean=[123.675, 116.28, 103.53],
        std=[58.395, 57.12, 57.375],
        to_rgb=True),
    dict(type='Pad',
        size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect',
        keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])]
test_pipeline=[
    dict(type='LoadImageFromFile'),
    dict(type='MultiScaleFlipAug',
        img_scale=(1333, 800),
        flip=False,
        transforms=[
            dict(type='Resize',
                keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad',
                size_divisor=32),
            dict(type='ImageToTensor',
                keys=['img']),
            dict(type='Collect',
                keys=['img'])])]
data=dict(
    samples_per_gpu=2,
    workers_per_gpu=2,
    train=dict(
        classes=classes,
        type='CocoDataset',
        ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_train.json',
        img_prefix='data/cityscapes/leftImg8bit/train',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='LoadAnnotations',
                with_bbox=True,
                with_mask=True),
            dict(type='Resize',
                img_scale=(1333, 800),
                keep_ratio=True),
            dict(type='RandomFlip',
                flip_ratio=0.5),
            dict(type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad',
                size_divisor=32),
            dict(type='DefaultFormatBundle'),
            dict(type='Collect',
                keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])]),
    val=dict(
        classes=classes,
        type='CocoDataset',
        ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json',
        img_prefix='data/cityscapes/leftImg8bit/val',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='MultiScaleFlipAug',
                img_scale=(1333, 800),
                flip=False,
                transforms=[
                    dict(type='Resize',
                        keep_ratio=True),
                    dict(type='RandomFlip'),
                    dict(type='Normalize',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12, 57.375],
                        to_rgb=True),
                    dict(type='Pad',
                        size_divisor=32),
                    dict(type='ImageToTensor',
                        keys=['img']),
                    dict(type='Collect',
                        keys=['img'])])]),
    test=dict(
        classes=classes,
        type='CocoDataset',
        ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json',
        img_prefix='data/cityscapes/leftImg8bit/val',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='MultiScaleFlipAug',
                img_scale=(1333, 800),
                flip=False,
                transforms=[
                    dict(type='Resize',
                        keep_ratio=True),
                    dict(type='RandomFlip'),
                    dict(type='Normalize',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12, 57.375],
                        to_rgb=True),
                    dict(type='Pad',
                        size_divisor=32),
                    dict(type='ImageToTensor',
                        keys=['img']),
                    dict(type='Collect',
                        keys=['img'])])]))
evaluation=dict(
    interval=1,
    metric=['bbox', 'segm'])
optimizer=dict(
    type='SGD',
    lr=0.02,
    momentum=0.9,
    weight_decay=0.0001)
optimizer_config=dict(
    grad_clip=None)
lr_config=dict(
    policy='step',
    warmup='linear',
    warmup_iters=500,
    warmup_ratio=0.001,
    step=[8, 11])
total_epochs=12
checkpoint_config=dict(
    interval=1)
log_config=dict(
    interval=50,
    hooks=[
        dict(type='TextLoggerHook')])
dist_params=dict(
    backend='nccl')
model=dict(
    type='MaskScoringRCNN',
    pretrained='open-mmlab://resnext101_32x4d',
    backbone=dict(
        type='ResNeXt',
        depth=101,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        norm_cfg=dict(
            type='BN',
            requires_grad=True),
        norm_eval=True,
        style='pytorch',
        groups=32,
        base_width=4),
    neck=dict(
        type='FPN',
        in_channels=[256, 512, 1024, 2048],
        out_channels=256,
        num_outs=5),
    rpn_head=dict(
        type='RPNHead',
        in_channels=256,
        feat_channels=256,
        anchor_generator=dict(
            type='AnchorGenerator',
            scales=[8],
            ratios=[0.5, 1.0, 2.0],
            strides=[4, 8, 16, 32, 64]),
        bbox_coder=dict(
            type='DeltaXYWHBBoxCoder',
            target_means=[0.0, 0.0, 0.0, 0.0],
            target_stds=[1.0, 1.0, 1.0, 1.0]),
        loss_cls=dict(
            type='CrossEntropyLoss',
            use_sigmoid=True,
            loss_weight=1.0),
        loss_bbox=dict(
            type='L1Loss',
            loss_weight=1.0)),
    roi_head=dict(
        type='MaskScoringRoIHead',
        bbox_roi_extractor=dict(
            type='SingleRoIExtractor',
            roi_layer=dict(
                type='RoIAlign',
                out_size=7,
                sample_num=0),
            out_channels=256,
            featmap_strides=[4, 8, 16, 32]),
        bbox_head=dict(
            type='Shared2FCBBoxHead',
            in_channels=256,
            fc_out_channels=1024,
            roi_feat_size=7,
            num_classes=8,
            bbox_coder=dict(
                type='DeltaXYWHBBoxCoder',
                target_means=[0.0, 0.0, 0.0, 0.0],
                target_stds=[0.1, 0.1, 0.2, 0.2]),
            reg_class_agnostic=False,
            loss_cls=dict(
                type='CrossEntropyLoss',
                use_sigmoid=False,
                loss_weight=1.0),
            loss_bbox=dict(
                type='L1Loss',
                loss_weight=1.0)),
        mask_roi_extractor=dict(
            type='SingleRoIExtractor',
            roi_layer=dict(
                type='RoIAlign',
                out_size=14,
                sample_num=0),
            out_channels=256,
            featmap_strides=[4, 8, 16, 32]),
        mask_head=dict(
            type='FCNMaskHead',
            num_convs=4,
            in_channels=256,
            conv_out_channels=256,
            num_classes=8,
            loss_mask=dict(
                type='CrossEntropyLoss',
                use_mask=True,
                loss_weight=1.0)),
        mask_iou_head=dict(
            type='MaskIoUHead',
            num_convs=4,
            num_fcs=2,
            roi_feat_size=14,
            in_channels=256,
            conv_out_channels=256,
            fc_out_channels=1024,
            num_classes=8)))
train_cfg=dict(
    rpn=dict(
        assigner=dict(
            type='MaxIoUAssigner',
            pos_iou_thr=0.7,
            neg_iou_thr=0.3,
            min_pos_iou=0.3,
            match_low_quality=True,
            ignore_iof_thr=-1),
        sampler=dict(
            type='RandomSampler',
            num=256,
            pos_fraction=0.5,
            neg_pos_ub=-1,
            add_gt_as_proposals=False),
        allowed_border=-1,
        pos_weight=-1,
        debug=False),
    rpn_proposal=dict(
        nms_across_levels=False,
        nms_pre=2000,
        nms_post=1000,
        max_num=1000,
        nms_thr=0.7,
        min_bbox_size=0),
    rcnn=dict(
        assigner=dict(
            type='MaxIoUAssigner',
            pos_iou_thr=0.5,
            neg_iou_thr=0.5,
            min_pos_iou=0.5,
            match_low_quality=True,
            ignore_iof_thr=-1),
        sampler=dict(
            type='RandomSampler',
            num=512,
            pos_fraction=0.25,
            neg_pos_ub=-1,
            add_gt_as_proposals=True),
        mask_size=28,
        pos_weight=-1,
        debug=False,
        mask_thr_binary=0.5))
test_cfg=dict(
    rpn=dict(
        nms_across_levels=False,
        nms_pre=1000,
        nms_post=1000,
        max_num=1000,
        nms_thr=0.7,
        min_bbox_size=0),
    rcnn=dict(
        score_thr=0.05,
        nms=dict(
            type='nms',
            iou_thr=0.5),
        max_per_img=100,
        mask_thr_binary=0.5))
dataset_type='CocoDataset'
data_root='data/cityscapes/'
img_norm_cfg=dict(
    mean=[123.675, 116.28, 103.53],
    std=[58.395, 57.12, 57.375],
    to_rgb=True)
train_pipeline=[
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations',
        with_bbox=True,
        with_mask=True),
    dict(type='Resize',
        img_scale=(1333, 800),
        keep_ratio=True),
    dict(type='RandomFlip',
        flip_ratio=0.5),
    dict(type='Normalize',
        mean=[123.675, 116.28, 103.53],
        std=[58.395, 57.12, 57.375],
        to_rgb=True),
    dict(type='Pad',
        size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect',
        keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])]
test_pipeline=[
    dict(type='LoadImageFromFile'),
    dict(type='MultiScaleFlipAug',
        img_scale=(1333, 800),
        flip=False,
        transforms=[
            dict(type='Resize',
                keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad',
                size_divisor=32),
            dict(type='ImageToTensor',
                keys=['img']),
            dict(type='Collect',
                keys=['img'])])]
data=dict(
    samples_per_gpu=2,
    workers_per_gpu=2,
    train=dict(
        classes=classes,
        type='CocoDataset',
        ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_train.json',
        img_prefix='data/cityscapes/leftImg8bit/train',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='LoadAnnotations',
                with_bbox=True,
                with_mask=True),
            dict(type='Resize',
                img_scale=(1333, 800),
                keep_ratio=True),
            dict(type='RandomFlip',
                flip_ratio=0.5),
            dict(type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad',
                size_divisor=32),
            dict(type='DefaultFormatBundle'),
            dict(type='Collect',
                keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])]),
    val=dict(
        classes=classes,
        type='CocoDataset',
        ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json',
        img_prefix='data/cityscapes/leftImg8bit/val',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='MultiScaleFlipAug',
                img_scale=(1333, 800),
                flip=False,
                transforms=[
                    dict(type='Resize',
                        keep_ratio=True),
                    dict(type='RandomFlip'),
                    dict(type='Normalize',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12, 57.375],
                        to_rgb=True),
                    dict(type='Pad',
                        size_divisor=32),
                    dict(type='ImageToTensor',
                        keys=['img']),
                    dict(type='Collect',
                        keys=['img'])])]),
    test=dict(
        classes=classes,
        type='CocoDataset',
        ann_file='data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json',
        img_prefix='data/cityscapes/leftImg8bit/val',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='MultiScaleFlipAug',
                img_scale=(1333, 800),
                flip=False,
                transforms=[
                    dict(type='Resize',
                        keep_ratio=True),
                    dict(type='RandomFlip'),
                    dict(type='Normalize',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12, 57.375],
                        to_rgb=True),
                    dict(type='Pad',
                        size_divisor=32),
                    dict(type='ImageToTensor',
                        keys=['img']),
                    dict(type='Collect',
                        keys=['img'])])]))
evaluation=dict(
    interval=1,
    metric=['bbox', 'segm'])
optimizer=dict(
    type='SGD',
    lr=0.02,
    momentum=0.9,
    weight_decay=0.0001)
optimizer_config=dict(
    grad_clip=None)
lr_config=dict(
    policy='step',
    warmup='linear',
    warmup_iters=500,
    warmup_ratio=0.001,
    step=[8, 11])
total_epochs=12
checkpoint_config=dict(
    interval=1)
log_config=dict(
    interval=50,
    hooks=[
        dict(type='TextLoggerHook')])
dist_params=dict(
    backend='nccl')

log_level='INFO'
load_from=None
resume_from=None
workflow=[('train', 1)]
work_dir='/ssd_scratch/cvit/dksingh/mmdetection_logs/ms_rcnn_x101_32x4d_fpn_1x_cityscapes_new/'
gpu_ids=range(0, 4)