Created by: lxk767363331

环境：Python 3.7 PaddlePaddle 1.6.2

aistudio@jupyter-61916-232190:~/PaddleDetection$ python tools/eval.py -c configs/faster_rcnn_r50_1x.yml loading annotations into memory... Done (t=0.00s) creating index... index created! 2019-12-29 21:32:29,197-INFO: 180 samples in file dataset/coco/annotations/instances_val2017.json 2019-12-29 21:32:29,198-INFO: places would be ommited when DataLoader is not iterable W1229 21:32:30.190289 1413 device_context.cc:236] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 9.2, Runtime API Version: 9.0 W1229 21:32:30.194048 1413 device_context.cc:244] device: 0, cuDNN Version: 7.3. Traceback (most recent call last): File "tools/eval.py", line 188, in main() File "tools/eval.py", line 108, in main checkpoint.load_params(exe, eval_prog, cfg.weights) File "/home/aistudio/PaddleDetection/ppdet/utils/checkpoint.py", line 98, in load_params "exists.".format(path)) ValueError: Model pretrain path output/faster_rcnn_r50_1x/model_final does not exists. aistudio@jupyter-61916-232190:~/PaddleDetection$ python tools/eval.py -c configs/faster_rcnn_r50_1x.yml loading annotations into memory... Done (t=0.00s) creating index... index created! 2019-12-29 21:40:25,572-INFO: 180 samples in file dataset/coco/annotations/instances_val2017.json 2019-12-29 21:40:25,573-INFO: places would be ommited when DataLoader is not iterable W1229 21:40:26.565089 1507 device_context.cc:236] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 9.2, Runtime API Version: 9.0 W1229 21:40:26.568838 1507 device_context.cc:244] device: 0, cuDNN Version: 7.3. 2019-12-29 21:40:27,966-INFO: Loading parameters from output/faster_rcnn_r50_1x/model_final... I1229 21:40:28.223676 1507 parallel_executor.cc:421] The number of CUDAPlace, which is used in ParallelExecutor, is 1. And the Program will be copied 1 copies I1229 21:40:28.229634 1507 build_strategy.cc:363] SeqOnlyAllReduceOps:0, num_trainers:1 I1229 21:40:28.234683 1507 parallel_executor.cc:285] Inplace strategy is enabled, when build_strategy.enable_inplace = True I1229 21:40:28.238572 1507 parallel_executor.cc:368] Garbage collection strategy is enabled, when FLAGS_eager_delete_tensor_gb = 0 2019-12-29 21:40:28,435-INFO: Test iter 0 2019-12-29 21:40:33,650-INFO: Test iter 100 2019-12-29 21:40:37,655-INFO: Test finish iter 180 2019-12-29 21:40:37,655-INFO: Total number of images: 180, inference time: 19.07627173697089 fps. loading annotations into memory... Done (t=0.00s) creating index... index created! 2019-12-29 21:40:37,702-INFO: Start evaluate... Loading and preparing results... Traceback (most recent call last): File "tools/eval.py", line 188, in main() File "tools/eval.py", line 165, in main is_bbox_normalized, FLAGS.output_eval, map_type) File "/home/aistudio/PaddleDetection/ppdet/utils/eval_utils.py", line 205, in eval_results is_bbox_normalized=is_bbox_normalized) File "/home/aistudio/PaddleDetection/ppdet/utils/coco_eval.py", line 96, in bbox_eval map_stats = cocoapi_eval(outfile, 'bbox', coco_gt=coco_gt) File "/home/aistudio/PaddleDetection/ppdet/utils/coco_eval.py", line 143, in cocoapi_eval coco_dt = coco_gt.loadRes(jsonfile) File "/home/aistudio/.local/lib/python3.7/site-packages/pycocotools-2.0-py3.7-linux-x86_64.egg/pycocotools/coco.py", line 317, in loadRes 'Results do not correspond to current coco set' AssertionError: Results do not correspond to current coco set