search

facebookresearch / detectron2

Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.
https://detectron2.readthedocs.io/en/latest/
Apache License 2.0
29.92k stars 7.4k forks source link

Train COCO from scratch without using Imagenet weights #348

Closed eleffani closed 4 years ago

eleffani commented 4 years ago

How To Reproduce the Issue

  1. edited file "configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml"

WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl" to WEIGHTS: "" and STEPS: (210000, 250000) MAX_ITER: 270000 to STEPS: (480000, 520000) MAX_ITER: 540000

  1. I run: python3 tools/train_net.py --config-file configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml SOLVER.IMS_PER_BATCH 2 SOLVER.BASE_LR 0.0025
  2. and here is the output: 
    
Command Line Args: Namespace(config_file='configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml', dist_url='tcp://127.0.0.1:50152', eval_only=False, machine_rank=0, num_gpus=1, num_machines=1, opts=['SOLVER.IMS_PER_BATCH', '2', 'SOLVER.BASE_LR', '0.0025'], resume=False)
Config 'configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml' has no VERSION. Assuming it to be compatible with latest v2.
[11/19 15:07:02 detectron2]: Rank of current process: 0. World size: 1
[11/19 15:07:02 detectron2]: Environment info:
------------------------  --------------------------------------------------
sys.platform              linux
Python                    3.6.8 (default, Oct  7 2019, 12:59:55) [GCC 8.3.0]
Numpy                     1.16.3
Detectron2 Compiler       GCC 7.4
Detectron2 CUDA Compiler  10.0
DETECTRON2_ENV_MODULE     <not set>
PyTorch                   1.3.1
PyTorch Debug Build       False
torchvision               0.4.2
CUDA available            True
GPU 0                     TITAN Xp
GPU 1                     Quadro P400
CUDA_HOME                 /usr/local/cuda
NVCC                      Cuda compilation tools, release 10.0, V10.0.166
Pillow                    6.0.0
cv2                       4.1.1
------------------------  --------------------------------------------------
PyTorch built with:
- GCC 7.3
- Intel(R) Math Kernel Library Version 2019.0.4 Product Build 20190411 for Intel(R) 64 architecture applications
- Intel(R) MKL-DNN v0.20.5 (Git Hash 0125f28c61c1f822fd48570b4c1066f96fcb9b2e)
- OpenMP 201511 (a.k.a. OpenMP 4.5)
- NNPACK is enabled
- CUDA Runtime 10.1
- 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 7.6.3
- Magma 2.5.1
- Build settings: BLAS=MKL, BUILD_NAMEDTENSOR=OFF, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -fopenmp -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -O2 -fPIC -Wno-narrowing -Wall -Wextra -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 -Wno-stringop-overflow, DISABLE_NUMA=1, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=True, 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,

[11/19 15:07:02 detectron2]: Command line arguments: Namespace(config_file='configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml', dist_url='tcp://127.0.0.1:50152', eval_only=False, machine_rank=0, num_gpus=1, num_machines=1, opts=['SOLVER.IMS_PER_BATCH', '2', 'SOLVER.BASE_LR', '0.0025'], resume=False) [11/19 15:07:02 detectron2]: Contents of args.config_file=configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml: BASE: "../Base-RCNN-FPN.yaml" MODEL: WEIGHTS: "" MASK_ON: False RESNETS: DEPTH: 101 SOLVER: STEPS: (480000, 520000) MAX_ITER: 540000 [11/19 15:07:02 detectron2]: Running with full config: CUDNN_BENCHMARK: False DATALOADER: ASPECT_RATIO_GROUPING: True FILTER_EMPTY_ANNOTATIONS: True NUM_WORKERS: 4 REPEAT_THRESHOLD: 0.0 SAMPLER_TRAIN: TrainingSampler DATASETS: PRECOMPUTED_PROPOSAL_TOPK_TEST: 1000 PRECOMPUTED_PROPOSAL_TOPK_TRAIN: 2000 PROPOSAL_FILES_TEST: () PROPOSAL_FILES_TRAIN: () TEST: ('coco_2017_val',) TRAIN: ('coco_2017_train',) GLOBAL: HACK: 1.0 INPUT: CROP: ENABLED: False SIZE: [0.9, 0.9] TYPE: relative_range FORMAT: BGR MASK_FORMAT: polygon MAX_SIZE_TEST: 1333 MAX_SIZE_TRAIN: 1333 MIN_SIZE_TEST: 800 MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800) MIN_SIZE_TRAIN_SAMPLING: choice MODEL: ANCHOR_GENERATOR: ANGLES: [[-90, 0, 90]] ASPECT_RATIOS: [[0.5, 1.0, 2.0]] NAME: DefaultAnchorGenerator SIZES: [[32], [64], [128], [256], [512]] BACKBONE: FREEZE_AT: 2 NAME: build_resnet_fpn_backbone DEVICE: cuda FPN: FUSE_TYPE: sum IN_FEATURES: ['res2', 'res3', 'res4', 'res5'] NORM: OUT_CHANNELS: 256 KEYPOINT_ON: False LOAD_PROPOSALS: False MASK_ON: False META_ARCHITECTURE: GeneralizedRCNN PANOPTIC_FPN: COMBINE: ENABLED: True INSTANCES_CONFIDENCE_THRESH: 0.5 OVERLAP_THRESH: 0.5 STUFF_AREA_LIMIT: 4096 INSTANCE_LOSS_WEIGHT: 1.0 PIXEL_MEAN: [103.53, 116.28, 123.675] PIXEL_STD: [1.0, 1.0, 1.0] PROPOSAL_GENERATOR: MIN_SIZE: 0 NAME: RPN RESNETS: DEFORM_MODULATED: False DEFORM_NUM_GROUPS: 1 DEFORM_ON_PER_STAGE: [False, False, False, False] DEPTH: 101 NORM: FrozenBN NUM_GROUPS: 1 OUT_FEATURES: ['res2', 'res3', 'res4', 'res5'] RES2_OUT_CHANNELS: 256 RES5_DILATION: 1 STEM_OUT_CHANNELS: 64 STRIDE_IN_1X1: True WIDTH_PER_GROUP: 64 RETINANET: BBOX_REG_WEIGHTS: (1.0, 1.0, 1.0, 1.0) FOCAL_LOSS_ALPHA: 0.25 FOCAL_LOSS_GAMMA: 2.0 IN_FEATURES: ['p3', 'p4', 'p5', 'p6', 'p7'] IOU_LABELS: [0, -1, 1] IOU_THRESHOLDS: [0.4, 0.5] NMS_THRESH_TEST: 0.5 NUM_CLASSES: 80 NUM_CONVS: 4 PRIOR_PROB: 0.01 SCORE_THRESH_TEST: 0.05 SMOOTH_L1_LOSS_BETA: 0.1 TOPK_CANDIDATES_TEST: 1000 ROI_BOX_CASCADE_HEAD: BBOX_REG_WEIGHTS: ((10.0, 10.0, 5.0, 5.0), (20.0, 20.0, 10.0, 10.0), (30.0, 30.0, 15.0, 15.0)) IOUS: (0.5, 0.6, 0.7) ROI_BOX_HEAD: BBOX_REG_WEIGHTS: (10.0, 10.0, 5.0, 5.0) CLS_AGNOSTIC_BBOX_REG: False CONV_DIM: 256 FC_DIM: 1024 NAME: FastRCNNConvFCHead NORM: NUM_CONV: 0 NUM_FC: 2 POOLER_RESOLUTION: 7 POOLER_SAMPLING_RATIO: 0 POOLER_TYPE: ROIAlignV2 SMOOTH_L1_BETA: 0.0 ROI_HEADS: BATCH_SIZE_PER_IMAGE: 512 IN_FEATURES: ['p2', 'p3', 'p4', 'p5'] IOU_LABELS: [0, 1] IOU_THRESHOLDS: [0.5] NAME: StandardROIHeads NMS_THRESH_TEST: 0.5 NUM_CLASSES: 80 POSITIVE_FRACTION: 0.25 PROPOSAL_APPEND_GT: True SCORE_THRESH_TEST: 0.05 ROI_KEYPOINT_HEAD: CONV_DIMS: (512, 512, 512, 512, 512, 512, 512, 512) LOSS_WEIGHT: 1.0 MIN_KEYPOINTS_PER_IMAGE: 1 NAME: KRCNNConvDeconvUpsampleHead NORMALIZE_LOSS_BY_VISIBLE_KEYPOINTS: True NUM_KEYPOINTS: 17 POOLER_RESOLUTION: 14 POOLER_SAMPLING_RATIO: 0 POOLER_TYPE: ROIAlignV2 ROI_MASK_HEAD: CLS_AGNOSTIC_MASK: False CONV_DIM: 256 NAME: MaskRCNNConvUpsampleHead NORM: NUM_CONV: 4 POOLER_RESOLUTION: 14 POOLER_SAMPLING_RATIO: 0 POOLER_TYPE: ROIAlignV2 RPN: BATCH_SIZE_PER_IMAGE: 256 BBOX_REG_WEIGHTS: (1.0, 1.0, 1.0, 1.0) BOUNDARY_THRESH: -1 HEAD_NAME: StandardRPNHead IN_FEATURES: ['p2', 'p3', 'p4', 'p5', 'p6'] IOU_LABELS: [0, -1, 1] IOU_THRESHOLDS: [0.3, 0.7] LOSS_WEIGHT: 1.0 NMS_THRESH: 0.7 POSITIVE_FRACTION: 0.5 POST_NMS_TOPK_TEST: 1000 POST_NMS_TOPK_TRAIN: 1000 PRE_NMS_TOPK_TEST: 1000 PRE_NMS_TOPK_TRAIN: 2000 SMOOTH_L1_BETA: 0.0 SEM_SEG_HEAD: COMMON_STRIDE: 4 CONVS_DIM: 128 IGNORE_VALUE: 255 IN_FEATURES: ['p2', 'p3', 'p4', 'p5'] LOSS_WEIGHT: 1.0 NAME: SemSegFPNHead NORM: GN NUM_CLASSES: 54 WEIGHTS: OUTPUT_DIR: ./output SEED: -1 SOLVER: BASE_LR: 0.0025 BIAS_LR_FACTOR: 1.0 CHECKPOINT_PERIOD: 5000 GAMMA: 0.1 IMS_PER_BATCH: 2 LR_SCHEDULER_NAME: WarmupMultiStepLR MAX_ITER: 540000 MOMENTUM: 0.9 STEPS: (480000, 520000) WARMUP_FACTOR: 0.001 WARMUP_ITERS: 1000 WARMUP_METHOD: linear WEIGHT_DECAY: 0.0001 WEIGHT_DECAY_BIAS: 0.0001 WEIGHT_DECAY_NORM: 0.0 TEST: AUG: ENABLED: False FLIP: True MAX_SIZE: 4000 MIN_SIZES: (400, 500, 600, 700, 800, 900, 1000, 1100, 1200) DETECTIONS_PER_IMAGE: 100 EVAL_PERIOD: 0 EXPECTED_RESULTS: [] KEYPOINT_OKS_SIGMAS: [] PRECISE_BN: ENABLED: False NUM_ITER: 200 VERSION: 2 [11/19 15:07:02 detectron2]: Full config saved to /home/elefan01/od_fr/pytorch/detectron2_repo/output/config.yaml [11/19 15:07:02 d2.utils.env]: Using a generated random seed 2931111

GeneralizedRCNN( (backbone): FPN( (fpn_lateral2): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1)) (fpn_output2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (fpn_lateral3): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1)) (fpn_output3): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (fpn_lateral4): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1)) (fpn_output4): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (fpn_lateral5): Conv2d(2048, 256, kernel_size=(1, 1), stride=(1, 1)) (fpn_output5): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (top_block): LastLevelMaxPool() (bottom_up): ResNet( (stem): BasicStem( (conv1): Conv2d( 3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) ) (res2): Sequential( (0): BottleneckBlock( (shortcut): Conv2d( 64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv1): Conv2d( 64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) (conv2): Conv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) (conv3): Conv2d( 64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) ) (1): BottleneckBlock( (conv1): Conv2d( 256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) (conv2): Conv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) (conv3): Conv2d( 64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) ) (2): BottleneckBlock( (conv1): Conv2d( 256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) (conv2): Conv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05) ) (conv3): Conv2d( 64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) ) ) (res3): Sequential( (0): BottleneckBlock( (shortcut): Conv2d( 256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv1): Conv2d( 256, 128, kernel_size=(1, 1), stride=(2, 2), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv2): Conv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv3): Conv2d( 128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) ) (1): BottleneckBlock( (conv1): Conv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv2): Conv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv3): Conv2d( 128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) ) (2): BottleneckBlock( (conv1): Conv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv2): Conv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv3): Conv2d( 128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) ) (3): BottleneckBlock( (conv1): Conv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv2): Conv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05) ) (conv3): Conv2d( 128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) ) ) (res4): Sequential( (0): BottleneckBlock( (shortcut): Conv2d( 512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) (conv1): Conv2d( 512, 256, kernel_size=(1, 1), stride=(2, 2), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (1): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (2): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (3): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (4): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (5): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (6): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (7): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (8): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (9): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (10): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (11): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (12): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (13): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (14): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (15): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (16): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (17): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (18): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (19): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (20): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (21): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) (22): BottleneckBlock( (conv1): Conv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv2): Conv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05) ) (conv3): Conv2d( 256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05) ) ) ) (res5): Sequential( (0): BottleneckBlock( (shortcut): Conv2d( 1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05) ) (conv1): Conv2d( 1024, 512, kernel_size=(1, 1), stride=(2, 2), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv2): Conv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv3): Conv2d( 512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05) ) ) (1): BottleneckBlock( (conv1): Conv2d( 2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv2): Conv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv3): Conv2d( 512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05) ) ) (2): BottleneckBlock( (conv1): Conv2d( 2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv2): Conv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05) ) (conv3): Conv2d( 512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05) ) ) ) ) ) (proposal_generator): RPN( (anchor_generator): DefaultAnchorGenerator( (cell_anchors): BufferList() ) (rpn_head): StandardRPNHead( (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (objectness_logits): Conv2d(256, 3, kernel_size=(1, 1), stride=(1, 1)) (anchor_deltas): Conv2d(256, 12, kernel_size=(1, 1), stride=(1, 1)) ) ) (roi_heads): StandardROIHeads( (box_pooler): ROIPooler( (level_poolers): ModuleList( (0): ROIAlign(output_size=(7, 7), spatial_scale=0.25, sampling_ratio=0, aligned=True) (1): ROIAlign(output_size=(7, 7), spatial_scale=0.125, sampling_ratio=0, aligned=True) (2): ROIAlign(output_size=(7, 7), spatial_scale=0.0625, sampling_ratio=0, aligned=True) (3): ROIAlign(output_size=(7, 7), spatial_scale=0.03125, sampling_ratio=0, aligned=True) ) ) (box_head): FastRCNNConvFCHead( (fc1): Linear(in_features=12544, out_features=1024, bias=True) (fc2): Linear(in_features=1024, out_features=1024, bias=True) ) (box_predictor): FastRCNNOutputLayers( (cls_score): Linear(in_features=1024, out_features=81, bias=True) (bbox_pred): Linear(in_features=1024, out_features=320, bias=True) ) ) ) [11/19 15:07:16 d2.data.datasets.coco]: Loading datasets/coco/annotations/instances_train2017.json takes 10.88 seconds. [11/19 15:07:17 d2.data.datasets.coco]: Loaded 118287 images in COCO format from datasets/coco/annotations/instances_train2017.json [11/19 15:07:23 d2.data.build]: Removed 1021 images with no usable annotations. 117266 images left. [11/19 15:07:25 d2.data.build]: Distribution of training instances among all 80 categories: category #instances category #instances category #instances
person 257253 bicycle 7056 car 43533
motorcycle 8654 airplane 5129 bus 6061
train 4570 truck 9970 boat 10576
traffic light 12842 fire hydrant 1865 stop sign 1983
parking meter 1283 bench 9820 bird 10542
cat 4766 dog 5500 horse 6567
sheep 9223 cow 8014 elephant 5484
bear 1294 zebra 5269 giraffe 5128
backpack 8714 umbrella 11265 handbag 12342
tie 6448 suitcase 6112 frisbee 2681
skis 6623 snowboard 2681 sports ball 6299
kite 8802 baseball bat 3273 baseball gl.. 3747
skateboard 5536 surfboard 6095 tennis racket 4807
bottle 24070 wine glass 7839 cup 20574
fork 5474 knife 7760 spoon 6159
bowl 14323 banana 9195 apple 5776
sandwich 4356 orange 6302 broccoli 7261
carrot 7758 hot dog 2884 pizza 5807
donut 7005 cake 6296 chair 38073
couch 5779 potted plant 8631 bed 4192
dining table 15695 toilet 4149 tv 5803
laptop 4960 mouse 2261 remote 5700
keyboard 2854 cell phone 6422 microwave 1672
oven 3334 toaster 225 sink 5609
refrigerator 2634 book 24077 clock 6320
vase 6577 scissors 1464 teddy bear 4729
hair drier 198 toothbrush 1945
total 849949

[11/19 15:07:25 d2.data.detection_utils]: TransformGens used in training: [ResizeShortestEdge(short_edge_length=(640, 672, 704, 736, 768, 800), max_size=1333, sample_style='choice'), RandomFlip()] [11/19 15:07:25 d2.data.build]: Using training sampler TrainingSampler [11/19 15:07:25 fvcore.common.checkpoint]: No checkpoint found. Initializing model from scratch [11/19 15:07:25 d2.engine.train_loop]: Starting training from iteration 0 [11/19 15:07:26 d2.engine.hooks]: Total training time: 0:00:00 (0:00:00 on hooks) Traceback (most recent call last): File "tools/train_net.py", line 161, in args=(args,), File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/engine/launch.py", line 52, in launch main_func(args) File "tools/train_net.py", line 149, in main return trainer.train() File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/engine/defaults.py", line 354, in train super().train(self.start_iter, self.max_iter) File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/engine/train_loop.py", line 132, in train self.run_step() File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/engine/train_loop.py", line 212, in run_step loss_dict = self.model(data) File "/home/elefan01/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in call result = self.forward(input, *kwargs) File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/modeling/meta_arch/rcnn.py", line 82, in forward proposals, proposal_losses = self.proposal_generator(images, features, gt_instances) File "/home/elefan01/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in call result = self.forward(input, **kwargs) File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/modeling/proposal_generator/rpn.py", line 172, in forward outputs.predict_proposals(), File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/modeling/proposal_generator/rpn_outputs.py", line 416, in predict_proposals pred_anchor_deltas_i, anchors_i.tensor File "/home/elefan01/od_fr/pytorch/detectron2_repo/detectron2/modeling/box_regression.py", line 79, in apply_deltas assert torch.isfinite(deltas).all().item(), "Box regression deltas become infinite or NaN!" AssertionError: Box regression deltas become infinite or NaN!



## Expected behavior
Able to train from scratch without providing Imagenet weights

## Environment
------------------------  --------------------------------------------------
sys.platform              linux
Python                    3.6.8 (default, Oct  7 2019, 12:59:55) [GCC 8.3.0]
Numpy                     1.16.3
Detectron2 Compiler       GCC 7.4
Detectron2 CUDA Compiler  10.0
DETECTRON2_ENV_MODULE     <not set>
PyTorch                   1.3.1
PyTorch Debug Build       False
torchvision               0.4.2
CUDA available            True
GPU 0                     TITAN Xp
GPU 1                     Quadro P400
CUDA_HOME                 /usr/local/cuda
NVCC                      Cuda compilation tools, release 10.0, V10.0.166
Pillow                    6.0.0
cv2                       4.1.1
------------------------  --------------------------------------------------
PyTorch built with:
  - GCC 7.3
  - Intel(R) Math Kernel Library Version 2019.0.4 Product Build 20190411 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v0.20.5 (Git Hash 0125f28c61c1f822fd48570b4c1066f96fcb9b2e)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - NNPACK is enabled
  - CUDA Runtime 10.1
  - 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 7.6.3
  - Magma 2.5.1
  - Build settings: BLAS=MKL, BUILD_NAMEDTENSOR=OFF, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -fopenmp -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -O2 -fPIC -Wno-narrowing -Wall -Wextra -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 -Wno-stringop-overflow, DISABLE_NUMA=1, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=True, 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,
ppwwyyxx commented 4 years ago

It is simply not the correct way to train from scratch. We've provided from-scratch configs in the model zoo and please use them as a reference. We do not help users tune their parameters for anything different from the model zoo.

You may want to read https://arxiv.org/abs/1811.08883