[01/08 07:36:30] meddlr INFO: Running in debug mode [01/08 07:36:30] meddlr INFO: Environment info: ------------------- ---------------------------------------------------------------------------------------------- sys.platform linux Python 3.7.11 (default, Jul 27 2021, 14:32:16) [GCC 7.5.0] numpy 1.20.3 PyTorch 1.7.1 @/bmrNAS/people/arjun/miniconda3/envs/meddlr_env/lib/python3.7/site-packages/torch PyTorch debug build False CUDA available False Pillow 8.4.0 torchvision 0.8.2 @/bmrNAS/people/arjun/miniconda3/envs/meddlr_env/lib/python3.7/site-packages/torchvision SLURM_JOB_ID slurm not detected ------------------- ---------------------------------------------------------------------------------------------- PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v1.6.0 (Git Hash 5ef631a030a6f73131c77892041042805a06064f) - OpenMP 201511 (a.k.a. OpenMP 4.5) - NNPACK is enabled - CPU capability usage: AVX - Build settings: BLAS=MKL, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DUSE_VULKAN_WRAPPER -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-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -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, [01/08 07:36:30] meddlr INFO: Command line arguments: Namespace(auto_version=False, config_file='../configs/tests/basic.yaml', debug=True, devices=None, eval_only=False, num_gpus=1, opts=['DATALOADER.NUM_WORKERS', '8', 'SOLVER.MAX_ITER', '1600', 'SOLVER.CHECKPOINT_PERIOD', '200', 'TEST.EVAL_PERIOD', '200'], reproducible=False, restart_iter=False, resume=False) [01/08 07:36:30] meddlr INFO: Contents of args.config_file=../configs/tests/basic.yaml: # Basic testing config # Use this for any testing you may want to do in the future. # The model will be trained for 60 iterations (not epochs) # on the mridata.org 2019 knee dataset. MODEL: UNROLLED: NUM_UNROLLED_STEPS: 8 NUM_RESBLOCKS: 2 NUM_FEATURES: 128 DROPOUT: 0. DATASETS: TRAIN: ("mridata_knee_2019_train",) VAL: ("mridata_knee_2019_val",) TEST: ("mridata_knee_2019_test",) DATALOADER: NUM_WORKERS: 0 # for debugging purposes SOLVER: TRAIN_BATCH_SIZE: 1 TEST_BATCH_SIZE: 2 CHECKPOINT_PERIOD: 20 MAX_ITER: 80 TEST: EVAL_PERIOD: 40 VIS_PERIOD: 20 TIME_SCALE: "iter" OUTPUT_DIR: "results://tests/basic" VERSION: 1 [01/08 07:37:52] meddlr INFO: Running in debug mode [01/08 07:37:54] meddlr INFO: Environment info: ---------------------- ---------------------------------------------------------------------------------------------- sys.platform linux Python 3.7.11 (default, Jul 27 2021, 14:32:16) [GCC 7.5.0] numpy 1.20.3 PyTorch 1.7.1 @/bmrNAS/people/arjun/miniconda3/envs/meddlr_env/lib/python3.7/site-packages/torch PyTorch debug build False CUDA available True GPU 0 GeForce RTX 2080 Ti CUDA_HOME /usr/local/cuda NVCC Cuda compilation tools, release 9.0, V9.0.176 Pillow 8.4.0 torchvision 0.8.2 @/bmrNAS/people/arjun/miniconda3/envs/meddlr_env/lib/python3.7/site-packages/torchvision torchvision arch flags sm_35, sm_50, sm_60, sm_70, sm_75 SLURM_JOB_ID 29543 ---------------------- ---------------------------------------------------------------------------------------------- PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v1.6.0 (Git Hash 5ef631a030a6f73131c77892041042805a06064f) - 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 -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DUSE_VULKAN_WRAPPER -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-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -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, [01/08 07:37:54] meddlr INFO: Command line arguments: Namespace(auto_version=False, config_file='../configs/tests/basic.yaml', debug=True, devices=None, eval_only=False, num_gpus=1, opts=['DATALOADER.NUM_WORKERS', '8', 'SOLVER.MAX_ITER', '1600', 'SOLVER.CHECKPOINT_PERIOD', '200', 'TEST.EVAL_PERIOD', '200'], reproducible=False, restart_iter=False, resume=False) [01/08 07:37:54] meddlr INFO: Contents of args.config_file=../configs/tests/basic.yaml: # Basic testing config # Use this for any testing you may want to do in the future. # The model will be trained for 60 iterations (not epochs) # on the mridata.org 2019 knee dataset. MODEL: UNROLLED: NUM_UNROLLED_STEPS: 8 NUM_RESBLOCKS: 2 NUM_FEATURES: 128 DROPOUT: 0. DATASETS: TRAIN: ("mridata_knee_2019_train",) VAL: ("mridata_knee_2019_val",) TEST: ("mridata_knee_2019_test",) DATALOADER: NUM_WORKERS: 0 # for debugging purposes SOLVER: TRAIN_BATCH_SIZE: 1 TEST_BATCH_SIZE: 2 CHECKPOINT_PERIOD: 20 MAX_ITER: 80 TEST: EVAL_PERIOD: 40 VIS_PERIOD: 20 TIME_SCALE: "iter" OUTPUT_DIR: "results://tests/basic" VERSION: 1 [01/08 07:37:54] meddlr INFO: Running with full config: AUG_TEST: UNDERSAMPLE: ACCELERATIONS: (6,) AUG_TRAIN: MOTION_P: 0.2 MRI_RECON: AUG_SENSITIVITY_MAPS: True SCHEDULER_P: IGNORE: False TRANSFORMS: () NOISE_P: 0.2 UNDERSAMPLE: ACCELERATIONS: (6,) CALIBRATION_SIZE: 20 CENTER_FRACTIONS: () MAX_ATTEMPTS: 30 NAME: PoissonDiskMaskFunc USE_MOTION: False USE_NOISE: False CUDNN_BENCHMARK: False DATALOADER: ALT_SAMPLER: PERIOD_SUPERVISED: 1 PERIOD_UNSUPERVISED: 1 DATA_KEYS: () DROP_LAST: True FILTER: BY: () GROUP_SAMPLER: AS_BATCH_SAMPLER: False BATCH_BY: () NUM_WORKERS: 8 PREFETCH_FACTOR: 2 SAMPLER_TRAIN: SUBSAMPLE_TRAIN: NUM_TOTAL: -1 NUM_TOTAL_BY_GROUP: () NUM_UNDERSAMPLED: 0 NUM_VAL: -1 NUM_VAL_BY_GROUP: () SEED: 1000 DATASETS: TEST: ('mridata_knee_2019_test',) TRAIN: ('mridata_knee_2019_train',) VAL: ('mridata_knee_2019_val',) DESCRIPTION: BRIEF: ENTITY_NAME: ss_recon EXP_NAME: PROJECT_NAME: ss_recon TAGS: () MODEL: A2R: META_ARCHITECTURE: GeneralizedUnrolledCNN USE_SUPERVISED_CONSISTENCY: False CONSISTENCY: AUG: MOTION: RANGE: (0.2, 0.5) SCHEDULER: WARMUP_ITERS: 0 WARMUP_METHOD: MRI_RECON: AUG_SENSITIVITY_MAPS: True SCHEDULER_P: IGNORE: False TRANSFORMS: () NOISE: MASK: RHO: 1.0 SCHEDULER: WARMUP_ITERS: 0 WARMUP_METHOD: STD_DEV: (1,) LATENT_LOSS_NAME: mag_l1 LATENT_LOSS_WEIGHT: 0.1 LOSS_NAME: l1 LOSS_WEIGHT: 0.1 NUM_LATENT_LAYERS: 1 USE_CONSISTENCY: True USE_LATENT: False CS: MAX_ITER: 200 REGULARIZATION: 0.005 DENOISING: META_ARCHITECTURE: GeneralizedUnrolledCNN NOISE: STD_DEV: (1,) USE_FULLY_SAMPLED_TARGET: True USE_FULLY_SAMPLED_TARGET_EVAL: None DEVICE: cuda M2R: META_ARCHITECTURE: GeneralizedUnrolledCNN USE_SUPERVISED_CONSISTENCY: False META_ARCHITECTURE: GeneralizedUnrolledCNN N2R: META_ARCHITECTURE: GeneralizedUnrolledCNN USE_SUPERVISED_CONSISTENCY: False NM2R: META_ARCHITECTURE: GeneralizedUnrolledCNN USE_SUPERVISED_CONSISTENCY: False NORMALIZER: KEYWORDS: () NAME: TopMagnitudeNormalizer RECON_LOSS: NAME: l1 RENORMALIZE_DATA: True SEG: ACTIVATION: sigmoid CLASSES: () INCLUDE_BACKGROUND: False SSDU: MASKER: PARAMS: META_ARCHITECTURE: GeneralizedUnrolledCNN UNET: BLOCK_ORDER: ('conv', 'relu', 'conv', 'relu', 'batchnorm', 'dropout') CHANNELS: 32 DROPOUT: 0.0 IN_CHANNELS: 2 NORMALIZE: False NUM_POOL_LAYERS: 4 OUT_CHANNELS: 2 UNROLLED: BLOCK_ARCHITECTURE: ResNet CONV_BLOCK: ACTIVATION: relu NORM: none NORM_AFFINE: False ORDER: ('norm', 'act', 'drop', 'conv') DROPOUT: 0.0 FIX_STEP_SIZE: False KERNEL_SIZE: (3,) NUM_EMAPS: 1 NUM_FEATURES: 128 NUM_RESBLOCKS: 2 NUM_UNROLLED_STEPS: 8 PADDING: SHARE_WEIGHTS: False WEIGHTS: OUTPUT_DIR: /bmrNAS/people/arjun/results/meddlr/tests/basic SEED: -1 SOLVER: BASE_LR: 0.0001 BIAS_LR_FACTOR: 1.0 CHECKPOINT_PERIOD: 200 GAMMA: 0.1 GRAD_ACCUM_ITERS: 1 LR_SCHEDULER_NAME: WarmupMultiStepLR MAX_ITER: 1600 MOMENTUM: 0.9 OPTIMIZER: Adam STEPS: (30000,) TEST_BATCH_SIZE: 2 TRAIN_BATCH_SIZE: 1 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: EVAL_PERIOD: 200 EXPECTED_RESULTS: [] FLUSH_PERIOD: 0 VAL_AS_TEST: True VAL_METRICS: RECON: () TIME_SCALE: iter VERSION: 1 VIS_PERIOD: 20 [01/08 07:37:54] meddlr INFO: Full config saved to /bmrNAS/people/arjun/results/meddlr/tests/basic/config.yaml [01/08 07:37:54] mr.utils.env INFO: Using a generated random seed 55308770 [01/08 07:37:54] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/train.json takes 0.04 seconds [01/08 07:37:54] mr.data.build INFO: Dropped 0 scans. 14 scans remaining [01/08 07:37:54] mr.data.build INFO: Dropped references for 0/14 scans. 14 scans with reference remaining [01/08 07:37:54] meddlr INFO: Calculated 4480 iterations per epoch [01/08 07:37:54] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/train.json takes 0.00 seconds [01/08 07:37:54] mr.data.build INFO: Dropped 0 scans. 14 scans remaining [01/08 07:37:54] mr.data.build INFO: Dropped references for 0/14 scans. 14 scans with reference remaining [01/08 07:38:01] fvcore.common.checkpoint INFO: No checkpoint found. Initializing model from scratch [01/08 07:38:01] mr.engine.train_loop INFO: Starting training from iteration 0 [01/08 07:38:13] mr.utils.events INFO: eta: 0:05:59 iter: 19 loss: 100222.734 total_loss: 100222.734 time: 0.2279 data_time: 0.2790 lr: 0.000002 max_mem: 2468M [01/08 07:38:18] mr.utils.events INFO: eta: 0:05:55 iter: 39 loss: 31516.157 total_loss: 31516.157 time: 0.2280 data_time: 0.0001 lr: 0.000004 max_mem: 2468M [01/08 07:38:23] mr.utils.events INFO: eta: 0:05:51 iter: 59 loss: 27504.994 total_loss: 27504.994 time: 0.2283 data_time: 0.0001 lr: 0.000006 max_mem: 2468M [01/08 07:38:28] mr.utils.events INFO: eta: 0:05:47 iter: 79 loss: 24592.978 total_loss: 24592.978 time: 0.2288 data_time: 0.0001 lr: 0.000008 max_mem: 2468M [01/08 07:38:34] mr.utils.events INFO: eta: 0:05:43 iter: 99 loss: 24272.967 total_loss: 24272.967 time: 0.2399 data_time: 0.0528 lr: 0.000010 max_mem: 2468M [01/08 07:38:39] mr.utils.events INFO: eta: 0:05:39 iter: 119 loss: 21447.893 total_loss: 21447.893 time: 0.2385 data_time: 0.0001 lr: 0.000012 max_mem: 2468M [01/08 07:38:44] mr.utils.events INFO: eta: 0:05:35 iter: 139 loss: 21284.981 total_loss: 21284.981 time: 0.2376 data_time: 0.0001 lr: 0.000014 max_mem: 2468M [01/08 07:38:49] mr.utils.events INFO: eta: 0:05:31 iter: 159 loss: 22184.024 total_loss: 22184.024 time: 0.2369 data_time: 0.0001 lr: 0.000016 max_mem: 2468M [01/08 07:38:55] mr.utils.events INFO: eta: 0:05:27 iter: 179 loss: 20008.485 total_loss: 20008.485 time: 0.2365 data_time: 0.0001 lr: 0.000018 max_mem: 2468M [01/08 07:39:00] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0000199.pth [01/08 07:39:01] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.07 seconds [01/08 07:39:01] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:39:01] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:39:01] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:39:05] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0258 s / img. ETA=0:00:46 [01/08 07:39:10] mr.evaluation.evaluator INFO: Inference done 44/320. 0.0238 s / img. ETA=0:00:41 [01/08 07:39:15] mr.evaluation.evaluator INFO: Inference done 77/320. 0.0229 s / img. ETA=0:00:36 [01/08 07:39:20] mr.evaluation.evaluator INFO: Inference done 110/320. 0.0226 s / img. ETA=0:00:31 [01/08 07:39:25] mr.evaluation.evaluator INFO: Inference done 143/320. 0.0224 s / img. ETA=0:00:26 [01/08 07:39:30] mr.evaluation.evaluator INFO: Inference done 176/320. 0.0223 s / img. ETA=0:00:21 [01/08 07:39:35] mr.evaluation.evaluator INFO: Inference done 209/320. 0.0222 s / img. ETA=0:00:16 [01/08 07:39:40] mr.evaluation.evaluator INFO: Inference done 237/320. 0.0221 s / img. ETA=0:00:12 [01/08 07:39:45] mr.evaluation.evaluator INFO: Inference done 270/320. 0.0221 s / img. ETA=0:00:07 [01/08 07:39:50] mr.evaluation.evaluator INFO: Inference done 303/320. 0.0220 s / img. ETA=0:00:02 [01/08 07:39:53] mr.evaluation.evaluator INFO: Total inference time: 0:00:48.855011 (0.155095 s / batch on 1 devices) [01/08 07:39:53] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.021862 s / batch on 1 devices) [01/08 07:40:03] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:40:03] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:40:06] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.211 (0.055) val_nrmse_mag 0.140 (0.033) val_psnr 31.509 (2.961) val_psnr_mag 35.018 (2.864) val_ssim (Wang) 0.831 (0.090) [01/08 07:40:06] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.131 (0.007) val_nrmse_scan 0.195 (0.012) val_psnr_mag_scan 42.912 (0.363) val_psnr_scan 39.450 (0.290) val_ssim (Wang)_scan 0.957 (0.003) [01/08 07:40:06] mr.evaluation.evaluator INFO: Evaluation Time: 13.612028 s [01/08 07:40:06] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:40:06] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:40:06] mr.evaluation.testing INFO: copypaste: 0.2112,0.1404,31.5090,35.0178,0.8314,0.1310,0.1952,42.9119,39.4501,0.9571 [01/08 07:40:06] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.2112,0.1404,31.5090,35.0178,0.8314,0.1310,0.1952,42.9119,39.4501,0.9571 [01/08 07:40:06] mr.utils.events INFO: eta: 0:13:01 iter: 199 loss: 18439.461 total_loss: 18439.461 time: 0.2363 data_time: 0.0001 lr: 0.000020 max_mem: 4234M [01/08 07:40:12] mr.utils.events INFO: eta: 0:12:57 iter: 219 loss: 17119.530 total_loss: 17119.530 time: 0.2367 data_time: 0.0001 lr: 0.000022 max_mem: 4234M [01/08 07:40:17] mr.utils.events INFO: eta: 0:12:53 iter: 239 loss: 16810.184 total_loss: 16810.184 time: 0.2394 data_time: 0.0279 lr: 0.000024 max_mem: 4234M [01/08 07:40:22] mr.utils.events INFO: eta: 0:12:49 iter: 259 loss: 15697.525 total_loss: 15697.525 time: 0.2395 data_time: 0.0001 lr: 0.000026 max_mem: 4234M [01/08 07:40:28] mr.utils.events INFO: eta: 0:12:45 iter: 279 loss: 16743.501 total_loss: 16743.501 time: 0.2393 data_time: 0.0001 lr: 0.000028 max_mem: 4234M [01/08 07:40:33] mr.utils.events INFO: eta: 0:12:40 iter: 299 loss: 18045.708 total_loss: 18045.708 time: 0.2392 data_time: 0.0013 lr: 0.000030 max_mem: 4234M [01/08 07:40:39] mr.utils.events INFO: eta: 0:12:36 iter: 319 loss: 16170.239 total_loss: 16170.239 time: 0.2392 data_time: 0.0001 lr: 0.000032 max_mem: 4234M [01/08 07:40:45] mr.utils.events INFO: eta: 0:12:32 iter: 339 loss: 16581.456 total_loss: 16581.456 time: 0.2412 data_time: 0.0363 lr: 0.000034 max_mem: 4234M [01/08 07:40:50] mr.utils.events INFO: eta: 0:12:28 iter: 359 loss: 17071.639 total_loss: 17071.639 time: 0.2411 data_time: 0.0001 lr: 0.000036 max_mem: 4234M [01/08 07:40:56] mr.utils.events INFO: eta: 0:12:25 iter: 379 loss: 16226.621 total_loss: 16226.621 time: 0.2411 data_time: 0.0001 lr: 0.000038 max_mem: 4234M [01/08 07:41:01] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0000399.pth [01/08 07:41:02] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:41:02] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:41:02] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:41:02] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:41:06] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0223 s / img. ETA=0:00:47 [01/08 07:41:11] mr.evaluation.evaluator INFO: Inference done 44/320. 0.0222 s / img. ETA=0:00:42 [01/08 07:41:16] mr.evaluation.evaluator INFO: Inference done 77/320. 0.0221 s / img. ETA=0:00:37 [01/08 07:41:21] mr.evaluation.evaluator INFO: Inference done 110/320. 0.0221 s / img. ETA=0:00:32 [01/08 07:41:27] mr.evaluation.evaluator INFO: Inference done 143/320. 0.0221 s / img. ETA=0:00:27 [01/08 07:41:32] mr.evaluation.evaluator INFO: Inference done 176/320. 0.0221 s / img. ETA=0:00:22 [01/08 07:41:37] mr.evaluation.evaluator INFO: Inference done 209/320. 0.0221 s / img. ETA=0:00:17 [01/08 07:41:42] mr.evaluation.evaluator INFO: Inference done 242/320. 0.0221 s / img. ETA=0:00:12 [01/08 07:41:47] mr.evaluation.evaluator INFO: Inference done 275/320. 0.0221 s / img. ETA=0:00:06 [01/08 07:41:52] mr.evaluation.evaluator INFO: Inference done 308/320. 0.0221 s / img. ETA=0:00:01 [01/08 07:41:54] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.053359 (0.155725 s / batch on 1 devices) [01/08 07:41:54] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.021976 s / batch on 1 devices) [01/08 07:42:04] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:42:04] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:42:07] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.176 (0.045) val_nrmse_mag 0.122 (0.030) val_psnr 33.094 (3.019) val_psnr_mag 36.272 (3.018) val_ssim (Wang) 0.860 (0.080) [01/08 07:42:07] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.113 (0.005) val_nrmse_scan 0.163 (0.006) val_psnr_mag_scan 44.208 (0.394) val_psnr_scan 41.032 (0.507) val_ssim (Wang)_scan 0.963 (0.002) [01/08 07:42:07] mr.evaluation.evaluator INFO: Evaluation Time: 13.187915 s [01/08 07:42:07] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:42:07] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:42:07] mr.evaluation.testing INFO: copypaste: 0.1758,0.1218,33.0937,36.2715,0.8602,0.1128,0.1626,44.2078,41.0316,0.9630 [01/08 07:42:07] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.1758,0.1218,33.0937,36.2715,0.8602,0.1128,0.1626,44.2078,41.0316,0.9630 [01/08 07:42:07] mr.utils.events INFO: eta: 0:11:15 iter: 399 loss: 14768.049 total_loss: 14768.049 time: 0.2410 data_time: 0.0001 lr: 0.000040 max_mem: 4235M [01/08 07:42:13] mr.utils.events INFO: eta: 0:11:11 iter: 419 loss: 14830.163 total_loss: 14830.163 time: 0.2410 data_time: 0.0001 lr: 0.000042 max_mem: 4235M [01/08 07:42:18] mr.utils.events INFO: eta: 0:11:06 iter: 439 loss: 16323.925 total_loss: 16323.925 time: 0.2410 data_time: 0.0001 lr: 0.000044 max_mem: 4235M [01/08 07:42:23] mr.utils.events INFO: eta: 0:11:02 iter: 459 loss: 14716.628 total_loss: 14716.628 time: 0.2409 data_time: 0.0001 lr: 0.000046 max_mem: 4235M [01/08 07:42:29] mr.utils.events INFO: eta: 0:10:57 iter: 479 loss: 15069.688 total_loss: 15069.688 time: 0.2409 data_time: 0.0001 lr: 0.000048 max_mem: 4235M [01/08 07:42:34] mr.utils.events INFO: eta: 0:10:53 iter: 499 loss: 14391.693 total_loss: 14391.693 time: 0.2409 data_time: 0.0001 lr: 0.000050 max_mem: 4235M [01/08 07:42:39] mr.utils.events INFO: eta: 0:10:48 iter: 519 loss: 14284.784 total_loss: 14284.784 time: 0.2409 data_time: 0.0001 lr: 0.000052 max_mem: 4235M [01/08 07:42:45] mr.utils.events INFO: eta: 0:10:44 iter: 539 loss: 15812.837 total_loss: 15812.837 time: 0.2409 data_time: 0.0001 lr: 0.000054 max_mem: 4235M [01/08 07:42:51] mr.utils.events INFO: eta: 0:10:39 iter: 559 loss: 15930.277 total_loss: 15930.277 time: 0.2409 data_time: 0.0001 lr: 0.000056 max_mem: 4235M [01/08 07:43:02] mr.utils.events INFO: eta: 0:10:34 iter: 579 loss: 15495.331 total_loss: 15495.331 time: 0.2410 data_time: 0.0001 lr: 0.000058 max_mem: 4235M [01/08 07:43:07] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0000599.pth [01/08 07:43:08] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:43:08] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:43:08] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:43:08] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:43:11] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0221 s / img. ETA=0:00:48 [01/08 07:43:16] mr.evaluation.evaluator INFO: Inference done 44/320. 0.0221 s / img. ETA=0:00:42 [01/08 07:43:21] mr.evaluation.evaluator INFO: Inference done 77/320. 0.0222 s / img. ETA=0:00:37 [01/08 07:43:26] mr.evaluation.evaluator INFO: Inference done 109/320. 0.0222 s / img. ETA=0:00:32 [01/08 07:43:31] mr.evaluation.evaluator INFO: Inference done 141/320. 0.0222 s / img. ETA=0:00:27 [01/08 07:43:37] mr.evaluation.evaluator INFO: Inference done 173/320. 0.0222 s / img. ETA=0:00:22 [01/08 07:43:42] mr.evaluation.evaluator INFO: Inference done 205/320. 0.0222 s / img. ETA=0:00:17 [01/08 07:43:47] mr.evaluation.evaluator INFO: Inference done 237/320. 0.0222 s / img. ETA=0:00:12 [01/08 07:43:52] mr.evaluation.evaluator INFO: Inference done 269/320. 0.0222 s / img. ETA=0:00:07 [01/08 07:43:57] mr.evaluation.evaluator INFO: Inference done 301/320. 0.0222 s / img. ETA=0:00:02 [01/08 07:44:00] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.342107 (0.156642 s / batch on 1 devices) [01/08 07:44:00] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.022043 s / batch on 1 devices) [01/08 07:44:09] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:44:09] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:44:13] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.217 (0.036) val_nrmse_mag 0.169 (0.023) val_psnr 31.138 (2.830) val_psnr_mag 33.292 (2.814) val_ssim (Wang) 0.867 (0.077) [01/08 07:44:13] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.162 (0.003) val_nrmse_scan 0.207 (0.002) val_psnr_mag_scan 41.054 (0.641) val_psnr_scan 38.946 (0.736) val_ssim (Wang)_scan 0.962 (0.000) [01/08 07:44:13] mr.evaluation.evaluator INFO: Evaluation Time: 13.194743 s [01/08 07:44:13] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:44:13] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:44:13] mr.evaluation.testing INFO: copypaste: 0.2174,0.1691,31.1379,33.2924,0.8669,0.1621,0.2066,41.0542,38.9462,0.9623 [01/08 07:44:13] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.2174,0.1691,31.1379,33.2924,0.8669,0.1621,0.2066,41.0542,38.9462,0.9623 [01/08 07:44:13] mr.utils.events INFO: eta: 0:09:24 iter: 599 loss: 13945.590 total_loss: 13945.590 time: 0.2409 data_time: 0.0001 lr: 0.000060 max_mem: 4237M [01/08 07:44:18] mr.utils.events INFO: eta: 0:09:19 iter: 619 loss: 16922.678 total_loss: 16922.678 time: 0.2410 data_time: 0.0001 lr: 0.000062 max_mem: 4237M [01/08 07:44:23] mr.utils.events INFO: eta: 0:09:15 iter: 639 loss: 14901.438 total_loss: 14901.438 time: 0.2410 data_time: 0.0001 lr: 0.000064 max_mem: 4237M [01/08 07:44:28] mr.utils.events INFO: eta: 0:09:10 iter: 659 loss: 14073.586 total_loss: 14073.586 time: 0.2410 data_time: 0.0001 lr: 0.000066 max_mem: 4237M [01/08 07:44:34] mr.utils.events INFO: eta: 0:09:06 iter: 679 loss: 16618.619 total_loss: 16618.619 time: 0.2410 data_time: 0.0001 lr: 0.000068 max_mem: 4237M [01/08 07:44:39] mr.utils.events INFO: eta: 0:09:01 iter: 699 loss: 17848.932 total_loss: 17848.932 time: 0.2410 data_time: 0.0001 lr: 0.000070 max_mem: 4237M [01/08 07:44:44] mr.utils.events INFO: eta: 0:08:56 iter: 719 loss: 16070.655 total_loss: 16070.655 time: 0.2410 data_time: 0.0001 lr: 0.000072 max_mem: 4237M [01/08 07:44:49] mr.utils.events INFO: eta: 0:08:52 iter: 739 loss: 16163.839 total_loss: 16163.839 time: 0.2411 data_time: 0.0001 lr: 0.000074 max_mem: 4237M [01/08 07:44:54] mr.utils.events INFO: eta: 0:08:47 iter: 759 loss: 15517.578 total_loss: 15517.578 time: 0.2411 data_time: 0.0001 lr: 0.000076 max_mem: 4237M [01/08 07:45:00] mr.utils.events INFO: eta: 0:08:42 iter: 779 loss: 15658.616 total_loss: 15658.616 time: 0.2411 data_time: 0.0001 lr: 0.000078 max_mem: 4237M [01/08 07:45:05] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0000799.pth [01/08 07:45:05] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:45:05] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:45:05] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:45:05] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:45:09] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0251 s / img. ETA=0:00:48 [01/08 07:45:14] mr.evaluation.evaluator INFO: Inference done 43/320. 0.0227 s / img. ETA=0:00:43 [01/08 07:45:19] mr.evaluation.evaluator INFO: Inference done 75/320. 0.0224 s / img. ETA=0:00:38 [01/08 07:45:24] mr.evaluation.evaluator INFO: Inference done 107/320. 0.0224 s / img. ETA=0:00:33 [01/08 07:45:29] mr.evaluation.evaluator INFO: Inference done 139/320. 0.0223 s / img. ETA=0:00:28 [01/08 07:45:34] mr.evaluation.evaluator INFO: Inference done 171/320. 0.0223 s / img. ETA=0:00:23 [01/08 07:45:39] mr.evaluation.evaluator INFO: Inference done 203/320. 0.0223 s / img. ETA=0:00:18 [01/08 07:45:44] mr.evaluation.evaluator INFO: Inference done 235/320. 0.0223 s / img. ETA=0:00:13 [01/08 07:45:49] mr.evaluation.evaluator INFO: Inference done 267/320. 0.0223 s / img. ETA=0:00:08 [01/08 07:45:54] mr.evaluation.evaluator INFO: Inference done 299/320. 0.0222 s / img. ETA=0:00:03 [01/08 07:45:57] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.500946 (0.157146 s / batch on 1 devices) [01/08 07:45:57] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.022107 s / batch on 1 devices) [01/08 07:46:07] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:46:07] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:46:10] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.175 (0.043) val_nrmse_mag 0.120 (0.032) val_psnr 33.099 (2.996) val_psnr_mag 36.426 (3.111) val_ssim (Wang) 0.866 (0.082) [01/08 07:46:10] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.111 (0.006) val_nrmse_scan 0.163 (0.006) val_psnr_mag_scan 44.388 (0.314) val_psnr_scan 41.028 (0.479) val_ssim (Wang)_scan 0.964 (0.002) [01/08 07:46:10] mr.evaluation.evaluator INFO: Evaluation Time: 13.062179 s [01/08 07:46:10] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:46:10] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:46:10] mr.evaluation.testing INFO: copypaste: 0.1754,0.1200,33.0992,36.4262,0.8660,0.1105,0.1627,44.3880,41.0275,0.9637 [01/08 07:46:10] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.1754,0.1200,33.0992,36.4262,0.8660,0.1105,0.1627,44.3880,41.0275,0.9637 [01/08 07:46:10] mr.utils.events INFO: eta: 0:07:32 iter: 799 loss: 15171.014 total_loss: 15171.014 time: 0.2411 data_time: 0.0001 lr: 0.000080 max_mem: 4238M [01/08 07:46:16] mr.utils.events INFO: eta: 0:07:27 iter: 819 loss: 14583.270 total_loss: 14583.270 time: 0.2411 data_time: 0.0001 lr: 0.000082 max_mem: 4238M [01/08 07:46:21] mr.utils.events INFO: eta: 0:07:23 iter: 839 loss: 14240.295 total_loss: 14240.295 time: 0.2411 data_time: 0.0001 lr: 0.000084 max_mem: 4238M [01/08 07:46:26] mr.utils.events INFO: eta: 0:07:18 iter: 859 loss: 15098.030 total_loss: 15098.030 time: 0.2411 data_time: 0.0001 lr: 0.000086 max_mem: 4238M [01/08 07:46:31] mr.utils.events INFO: eta: 0:07:13 iter: 879 loss: 15402.230 total_loss: 15402.230 time: 0.2412 data_time: 0.0001 lr: 0.000088 max_mem: 4238M [01/08 07:46:36] mr.utils.events INFO: eta: 0:07:08 iter: 899 loss: 14225.305 total_loss: 14225.305 time: 0.2412 data_time: 0.0001 lr: 0.000090 max_mem: 4238M [01/08 07:46:43] mr.utils.events INFO: eta: 0:07:04 iter: 919 loss: 13569.448 total_loss: 13569.448 time: 0.2412 data_time: 0.0001 lr: 0.000092 max_mem: 4238M [01/08 07:46:48] mr.utils.events INFO: eta: 0:06:59 iter: 939 loss: 14107.779 total_loss: 14107.779 time: 0.2412 data_time: 0.0001 lr: 0.000094 max_mem: 4238M [01/08 07:46:53] mr.utils.events INFO: eta: 0:06:54 iter: 959 loss: 14675.758 total_loss: 14675.758 time: 0.2413 data_time: 0.0001 lr: 0.000096 max_mem: 4238M [01/08 07:46:58] mr.utils.events INFO: eta: 0:06:50 iter: 979 loss: 14544.494 total_loss: 14544.494 time: 0.2413 data_time: 0.0001 lr: 0.000098 max_mem: 4238M [01/08 07:47:03] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0000999.pth [01/08 07:47:04] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:47:04] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:47:04] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:47:04] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:47:08] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0250 s / img. ETA=0:00:48 [01/08 07:47:13] mr.evaluation.evaluator INFO: Inference done 43/320. 0.0227 s / img. ETA=0:00:43 [01/08 07:47:18] mr.evaluation.evaluator INFO: Inference done 75/320. 0.0225 s / img. ETA=0:00:38 [01/08 07:47:23] mr.evaluation.evaluator INFO: Inference done 107/320. 0.0225 s / img. ETA=0:00:33 [01/08 07:47:28] mr.evaluation.evaluator INFO: Inference done 139/320. 0.0224 s / img. ETA=0:00:28 [01/08 07:47:33] mr.evaluation.evaluator INFO: Inference done 171/320. 0.0224 s / img. ETA=0:00:23 [01/08 07:47:38] mr.evaluation.evaluator INFO: Inference done 203/320. 0.0223 s / img. ETA=0:00:18 [01/08 07:47:43] mr.evaluation.evaluator INFO: Inference done 235/320. 0.0223 s / img. ETA=0:00:13 [01/08 07:47:48] mr.evaluation.evaluator INFO: Inference done 267/320. 0.0223 s / img. ETA=0:00:08 [01/08 07:47:53] mr.evaluation.evaluator INFO: Inference done 299/320. 0.0223 s / img. ETA=0:00:03 [01/08 07:47:56] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.544317 (0.157284 s / batch on 1 devices) [01/08 07:47:56] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.022172 s / batch on 1 devices) [01/08 07:48:06] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:48:06] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:48:09] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.175 (0.042) val_nrmse_mag 0.138 (0.030) val_psnr 33.089 (3.029) val_psnr_mag 35.133 (2.998) val_ssim (Wang) 0.810 (0.075) [01/08 07:48:09] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.129 (0.006) val_nrmse_scan 0.163 (0.005) val_psnr_mag_scan 43.045 (0.422) val_psnr_scan 41.020 (0.545) val_ssim (Wang)_scan 0.932 (0.011) [01/08 07:48:09] mr.evaluation.evaluator INFO: Evaluation Time: 13.296574 s [01/08 07:48:09] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:48:09] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:48:09] mr.evaluation.testing INFO: copypaste: 0.1755,0.1383,33.0890,35.1330,0.8102,0.1290,0.1628,43.0453,41.0200,0.9319 [01/08 07:48:09] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.1755,0.1383,33.0890,35.1330,0.8102,0.1290,0.1628,43.0453,41.0200,0.9319 [01/08 07:48:09] mr.utils.events INFO: eta: 0:05:40 iter: 999 loss: 14674.882 total_loss: 14674.882 time: 0.2413 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:15] mr.utils.events INFO: eta: 0:05:35 iter: 1019 loss: 14348.898 total_loss: 14348.898 time: 0.2413 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:20] mr.utils.events INFO: eta: 0:05:31 iter: 1039 loss: 14988.102 total_loss: 14988.102 time: 0.2413 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:25] mr.utils.events INFO: eta: 0:05:26 iter: 1059 loss: 17364.371 total_loss: 17364.371 time: 0.2413 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:30] mr.utils.events INFO: eta: 0:05:21 iter: 1079 loss: 14381.812 total_loss: 14381.812 time: 0.2413 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:35] mr.utils.events INFO: eta: 0:05:16 iter: 1099 loss: 14260.652 total_loss: 14260.652 time: 0.2413 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:41] mr.utils.events INFO: eta: 0:05:12 iter: 1119 loss: 14124.476 total_loss: 14124.476 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:46] mr.utils.events INFO: eta: 0:05:07 iter: 1139 loss: 13414.080 total_loss: 13414.080 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:51] mr.utils.events INFO: eta: 0:05:02 iter: 1159 loss: 14458.338 total_loss: 14458.338 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:48:56] mr.utils.events INFO: eta: 0:04:57 iter: 1179 loss: 13799.362 total_loss: 13799.362 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4240M [01/08 07:49:01] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0001199.pth [01/08 07:49:02] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:49:02] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:49:02] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:49:02] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:49:06] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0246 s / img. ETA=0:00:48 [01/08 07:49:11] mr.evaluation.evaluator INFO: Inference done 43/320. 0.0227 s / img. ETA=0:00:43 [01/08 07:49:16] mr.evaluation.evaluator INFO: Inference done 75/320. 0.0224 s / img. ETA=0:00:38 [01/08 07:49:21] mr.evaluation.evaluator INFO: Inference done 107/320. 0.0224 s / img. ETA=0:00:33 [01/08 07:49:26] mr.evaluation.evaluator INFO: Inference done 139/320. 0.0224 s / img. ETA=0:00:28 [01/08 07:49:31] mr.evaluation.evaluator INFO: Inference done 171/320. 0.0223 s / img. ETA=0:00:23 [01/08 07:49:36] mr.evaluation.evaluator INFO: Inference done 203/320. 0.0223 s / img. ETA=0:00:18 [01/08 07:49:41] mr.evaluation.evaluator INFO: Inference done 235/320. 0.0223 s / img. ETA=0:00:13 [01/08 07:49:46] mr.evaluation.evaluator INFO: Inference done 267/320. 0.0223 s / img. ETA=0:00:08 [01/08 07:49:51] mr.evaluation.evaluator INFO: Inference done 299/320. 0.0223 s / img. ETA=0:00:03 [01/08 07:49:54] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.547863 (0.157295 s / batch on 1 devices) [01/08 07:49:54] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.022115 s / batch on 1 devices) [01/08 07:50:04] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:50:04] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:50:08] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.182 (0.041) val_nrmse_mag 0.135 (0.029) val_psnr 32.762 (2.957) val_psnr_mag 35.311 (2.952) val_ssim (Wang) 0.863 (0.078) [01/08 07:50:08] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.127 (0.005) val_nrmse_scan 0.170 (0.005) val_psnr_mag_scan 43.202 (0.427) val_psnr_scan 40.667 (0.561) val_ssim (Wang)_scan 0.961 (0.005) [01/08 07:50:08] mr.evaluation.evaluator INFO: Evaluation Time: 13.242378 s [01/08 07:50:08] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:50:08] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:50:08] mr.evaluation.testing INFO: copypaste: 0.1817,0.1353,32.7618,35.3113,0.8626,0.1267,0.1695,43.2024,40.6668,0.9606 [01/08 07:50:08] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.1817,0.1353,32.7618,35.3113,0.8626,0.1267,0.1695,43.2024,40.6668,0.9606 [01/08 07:50:08] mr.utils.events INFO: eta: 0:03:48 iter: 1199 loss: 14033.812 total_loss: 14033.812 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:13] mr.utils.events INFO: eta: 0:03:43 iter: 1219 loss: 15478.925 total_loss: 15478.925 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:18] mr.utils.events INFO: eta: 0:03:38 iter: 1239 loss: 14415.258 total_loss: 14415.258 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:23] mr.utils.events INFO: eta: 0:03:33 iter: 1259 loss: 14165.776 total_loss: 14165.776 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:28] mr.utils.events INFO: eta: 0:03:28 iter: 1279 loss: 14173.212 total_loss: 14173.212 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:34] mr.utils.events INFO: eta: 0:03:24 iter: 1299 loss: 13085.434 total_loss: 13085.434 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:39] mr.utils.events INFO: eta: 0:03:19 iter: 1319 loss: 12723.986 total_loss: 12723.986 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:45] mr.utils.events INFO: eta: 0:03:14 iter: 1339 loss: 15378.066 total_loss: 15378.066 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:50] mr.utils.events INFO: eta: 0:03:09 iter: 1359 loss: 14171.920 total_loss: 14171.920 time: 0.2414 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:50:55] mr.utils.events INFO: eta: 0:03:04 iter: 1379 loss: 14608.114 total_loss: 14608.114 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4241M [01/08 07:51:00] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0001399.pth [01/08 07:51:01] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:51:01] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:51:01] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:51:01] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:51:05] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0218 s / img. ETA=0:00:48 [01/08 07:51:10] mr.evaluation.evaluator INFO: Inference done 43/320. 0.0218 s / img. ETA=0:00:43 [01/08 07:51:15] mr.evaluation.evaluator INFO: Inference done 75/320. 0.0218 s / img. ETA=0:00:38 [01/08 07:51:20] mr.evaluation.evaluator INFO: Inference done 107/320. 0.0219 s / img. ETA=0:00:33 [01/08 07:51:25] mr.evaluation.evaluator INFO: Inference done 139/320. 0.0219 s / img. ETA=0:00:28 [01/08 07:51:30] mr.evaluation.evaluator INFO: Inference done 171/320. 0.0218 s / img. ETA=0:00:23 [01/08 07:51:35] mr.evaluation.evaluator INFO: Inference done 203/320. 0.0218 s / img. ETA=0:00:18 [01/08 07:51:40] mr.evaluation.evaluator INFO: Inference done 235/320. 0.0218 s / img. ETA=0:00:13 [01/08 07:51:45] mr.evaluation.evaluator INFO: Inference done 267/320. 0.0218 s / img. ETA=0:00:08 [01/08 07:51:50] mr.evaluation.evaluator INFO: Inference done 299/320. 0.0218 s / img. ETA=0:00:03 [01/08 07:51:53] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.483551 (0.157091 s / batch on 1 devices) [01/08 07:51:53] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:06 (0.021677 s / batch on 1 devices) [01/08 07:52:03] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:52:03] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:52:06] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.164 (0.045) val_nrmse_mag 0.120 (0.034) val_psnr 33.694 (3.086) val_psnr_mag 36.480 (3.132) val_ssim (Wang) 0.833 (0.078) [01/08 07:52:06] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.110 (0.008) val_nrmse_scan 0.151 (0.006) val_psnr_mag_scan 44.465 (0.208) val_psnr_scan 41.657 (0.464) val_ssim (Wang)_scan 0.944 (0.010) [01/08 07:52:06] mr.evaluation.evaluator INFO: Evaluation Time: 13.212838 s [01/08 07:52:06] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:52:06] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:52:06] mr.evaluation.testing INFO: copypaste: 0.1645,0.1196,33.6935,36.4799,0.8328,0.1096,0.1513,44.4653,41.6570,0.9443 [01/08 07:52:06] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.1645,0.1196,33.6935,36.4799,0.8328,0.1096,0.1513,44.4653,41.6570,0.9443 [01/08 07:52:06] mr.utils.events INFO: eta: 0:01:54 iter: 1399 loss: 14015.138 total_loss: 14015.138 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:12] mr.utils.events INFO: eta: 0:01:49 iter: 1419 loss: 14504.874 total_loss: 14504.874 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:17] mr.utils.events INFO: eta: 0:01:44 iter: 1439 loss: 14570.553 total_loss: 14570.553 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:22] mr.utils.events INFO: eta: 0:01:39 iter: 1459 loss: 14148.630 total_loss: 14148.630 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:27] mr.utils.events INFO: eta: 0:01:34 iter: 1479 loss: 14428.591 total_loss: 14428.591 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:32] mr.utils.events INFO: eta: 0:01:30 iter: 1499 loss: 13502.969 total_loss: 13502.969 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:38] mr.utils.events INFO: eta: 0:01:25 iter: 1519 loss: 16368.226 total_loss: 16368.226 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:43] mr.utils.events INFO: eta: 0:01:20 iter: 1539 loss: 14534.019 total_loss: 14534.019 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:48] mr.utils.events INFO: eta: 0:01:15 iter: 1559 loss: 14640.928 total_loss: 14640.928 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:53] mr.utils.events INFO: eta: 0:01:10 iter: 1579 loss: 13991.085 total_loss: 13991.085 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4243M [01/08 07:52:58] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_0001599.pth [01/08 07:52:59] fvcore.common.checkpoint INFO: Saving checkpoint to /bmrNAS/people/arjun/results/meddlr/tests/basic/model_final.pth [01/08 07:52:59] mr.data.datasets.register_mrco INFO: Loading /bmrNAS/people/arjun/code/meddlr/annotations/mridata_knee_2019/val.json takes 0.00 seconds [01/08 07:52:59] mr.data.build INFO: Dropped 0 scans. 2 scans remaining [01/08 07:52:59] mr.data.build INFO: Dropped references for 0/2 scans. 2 scans with reference remaining [01/08 07:52:59] mr.evaluation.evaluator INFO: Start inference on 320 batches [01/08 07:53:03] mr.evaluation.evaluator INFO: Inference done 11/320. 0.0260 s / img. ETA=0:00:48 [01/08 07:53:08] mr.evaluation.evaluator INFO: Inference done 43/320. 0.0230 s / img. ETA=0:00:43 [01/08 07:53:13] mr.evaluation.evaluator INFO: Inference done 75/320. 0.0228 s / img. ETA=0:00:38 [01/08 07:53:18] mr.evaluation.evaluator INFO: Inference done 107/320. 0.0227 s / img. ETA=0:00:33 [01/08 07:53:23] mr.evaluation.evaluator INFO: Inference done 139/320. 0.0227 s / img. ETA=0:00:28 [01/08 07:53:28] mr.evaluation.evaluator INFO: Inference done 171/320. 0.0226 s / img. ETA=0:00:23 [01/08 07:53:33] mr.evaluation.evaluator INFO: Inference done 203/320. 0.0226 s / img. ETA=0:00:18 [01/08 07:53:38] mr.evaluation.evaluator INFO: Inference done 235/320. 0.0226 s / img. ETA=0:00:13 [01/08 07:53:43] mr.evaluation.evaluator INFO: Inference done 267/320. 0.0225 s / img. ETA=0:00:08 [01/08 07:53:48] mr.evaluation.evaluator INFO: Inference done 299/320. 0.0225 s / img. ETA=0:00:03 [01/08 07:53:52] mr.evaluation.evaluator INFO: Total inference time: 0:00:49.495082 (0.157127 s / batch on 1 devices) [01/08 07:53:52] mr.evaluation.evaluator INFO: Total inference pure compute time: 0:00:07 (0.022384 s / batch on 1 devices) [01/08 07:54:01] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:54:01] mr.evaluation.scan_evaluator INFO: Structuring slices into volumes... [01/08 07:54:05] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Slice metrics summary: channel_0 --------------- -------------- val_nrmse 0.189 (0.041) val_nrmse_mag 0.117 (0.032) val_psnr 32.410 (2.903) val_psnr_mag 36.686 (3.132) val_ssim (Wang) 0.858 (0.081) [01/08 07:54:05] mr.evaluation.recon_evaluation INFO: [ReconEvaluator] Scan metrics summary: channel_0 -------------------- -------------- val_nrmse_mag_scan 0.107 (0.005) val_nrmse_scan 0.177 (0.004) val_psnr_mag_scan 44.659 (0.393) val_psnr_scan 40.292 (0.600) val_ssim (Wang)_scan 0.959 (0.005) [01/08 07:54:05] mr.evaluation.evaluator INFO: Evaluation Time: 13.055382 s [01/08 07:54:05] mr.engine.trainer INFO: Evaluation results for mridata_knee_2019_val in csv format: [01/08 07:54:05] mr.evaluation.testing INFO: copypaste: val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan [01/08 07:54:05] mr.evaluation.testing INFO: copypaste: 0.1889,0.1167,32.4100,36.6860,0.8582,0.1071,0.1770,44.6590,40.2923,0.9595 [01/08 07:54:05] mr.evaluation.testing INFO: Metrics (comma delimited): val_nrmse,val_nrmse_mag,val_psnr,val_psnr_mag,val_ssim (Wang),val_nrmse_mag_scan,val_nrmse_scan,val_psnr_mag_scan,val_psnr_scan,val_ssim (Wang)_scan 0.1889,0.1167,32.4100,36.6860,0.8582,0.1071,0.1770,44.6590,40.2923,0.9595 [01/08 07:54:05] mr.utils.events INFO: eta: 0:00:00 iter: 1599 loss: 13840.166 total_loss: 13840.166 time: 0.2415 data_time: 0.0001 lr: 0.000100 max_mem: 4245M [01/08 07:54:05] mr.engine.hooks INFO: Overall training speed: 1597 iterations in 0:06:25 (0.2417 s / it) [01/08 07:54:05] mr.engine.hooks INFO: Total training time: 0:15:56 (0:09:30 on hooks)