search

PaddlePaddle / PaddleDetection

Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.
Apache License 2.0
12.78k stars 2.89k forks source link

EfficientNet b7 测试时image ResizeAndPad target_dim不是512就报异常 #2833

Open bbbxixixixi opened 3 years ago

bbbxixixixi commented 3 years ago

试过1024，800，608，480，都报错

architecture: EfficientDet max_iters: 14000 use_gpu: true pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/EfficientNetB7_pretrained.tar weights: output/efficientdet_d7/model_final log_iter: 20 snapshot_iter: 280 metric: COCO save_dir: output num_classes: 5 use_ema: true ema_decay: 0.9998

EfficientDet: backbone: EfficientNet fpn: BiFPN efficient_head: EfficientHead anchor_grid: AnchorGrid box_loss_weight: 50.

EfficientNet:

norm_type: sync_bn

TODO

norm_type: bn scale: b7 use_se: true

BiFPN: num_chan: 64 repeat: 3 levels: 5

EfficientHead: repeat: 3 num_chan: 64 prior_prob: 0.01 num_anchors: 9 gamma: 1.5 alpha: 0.25 delta: 0.1 output_decoder: score_thresh: 0.05 # originally 0. nms_thresh: 0.5 pre_nms_top_n: 1000 # originally 5000 detections_per_im: 100 nms_eta: 1.0

AnchorGrid: anchor_base_scale: 4 num_scales: 3 aspect_ratios: [[1, 1], [1.4, 0.7], [0.7, 1.4]]

LearningRate: base_lr: 0.001 schedulers:

OptimizerBuilder: clip_grad_by_norm: 10. optimizer: momentum: 0.9 type: Momentum regularizer: factor: 0.00004 type: L2

TrainReader: inputs_def: fields: ['image', 'im_id', 'fg_num', 'gt_label', 'gt_target'] dataset: !COCODataSet dataset_dir: /home/aistudio/data/wd anno_path: annotations/train_c_f5_1.json image_dir: images sample_transforms:

EvalReader: inputs_def: fields: ['image', 'im_info', 'im_id'] dataset: !COCODataSet dataset_dir: /home/aistudio/data/wd anno_path: annotations/eval_c_f5_1.json image_dir: images sample_transforms:

TestReader: inputs_def: fields: ['image', 'im_info', 'im_id'] image_shape: [3, 512, 512] dataset: !ImageFolder anno_path: /home/aistudio/data/wd/annotations/eval_c_f5_1.json sample_transforms:

Traceback (most recent call last): File "/home/aistudio/PaddleDetection-master/tools/train.py", line 405, in main() File "/home/aistudio/PaddleDetection-master/tools/train.py", line 322, in main resolution=resolution) File "/home/aistudio/PaddleDetection-master/ppdet/utils/eval_utils.py", line 148, in eval_run return_numpy=False) File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1110, in run six.reraise(*sys.exc_info()) File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/six.py", line 703, in reraise raise value File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1108, in run return_merged=return_merged) File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1252, in _run_impl return_merged=return_merged) File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 913, in _run_parallel tensors = exe.run(fetch_var_names, return_merged)._move_to_list() ValueError: In user code:

File "/home/aistudio/PaddleDetection-master/tools/train.py", line 405, in <module>
  main()
File "/home/aistudio/PaddleDetection-master/tools/train.py", line 170, in main
  fetches = model.eval(feed_vars)
File "/home/aistudio/PaddleDetection-master/ppdet/modeling/architectures/efficientdet.py", line 147, in eval
  return self.build(feed_vars, 'test')
File "/home/aistudio/PaddleDetection-master/ppdet/modeling/architectures/efficientdet.py", line 86, in build
  im_info)
File "/home/aistudio/PaddleDetection-master/ppdet/modeling/anchor_heads/efficient_head.py", line 162, in get_prediction
  im_info=im_info)
File "/home/aistudio/PaddleDetection-master/ppdet/core/workspace.py", line 179, in partial_apply
  return op(*args, **kwargs_)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/layers/detection.py", line 3257, in retinanet_detection_output
  outputs={'Out': output})
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/layer_helper.py", line 43, in append_op
  return self.main_program.current_block().append_op(*args, **kwargs)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/framework.py", line 3225, in append_op
  attrs=kwargs.get("attrs", None))
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/framework.py", line 2305, in __init__
  for frame in traceback.extract_stack():

InvalidArgumentError: The 1st dimension of each Variables in Input(Anchors) must be equal to the 2nd dimension of corresponding Variables in Input(BBoxes), which represents the number of the predicted boxes, but received Anchors 1st dimension is:36864, BBoxes 2nd dimension is:147456.
  [Hint: Expected anchor_dims[0] == bbox_dims[1], but received anchor_dims[0]:36864 != bbox_dims[1]:147456.] (at /paddle/paddle/fluid/operators/detection/retinanet_detection_output_op.cc:134)
  [operator < retinanet_detection_output > error]

terminate called without an active exception

C++ Traceback (most recent call last):

0 paddle::framework::SignalHandle(char const*, int) 1 paddle::platform::GetCurrentTraceBackString[abi:cxx11]()

Error Message Summary:

FatalError: Process abort signal is detected by the operating system. [TimeInfo: Aborted at 1619760295 (unix time) try "date -d @1619760295" if you are using GNU date ] [SignalInfo: SIGABRT (@0x3e800006dbe) received by PID 28094 (TID 0x7fc33dcb3700) from PID 28094 ]

Segmentation fault (core dumped)

wangxinxin08 commented 3 years ago

请问使用PaddleDetection版本是什么？

bbbxixixixi commented 3 years ago

请问使用PaddleDetection版本是什么？

paddle master， aistudio环境，换了devolop版本的paddlepaddle_gpu-2.1.0.dev0.post101-cp37-cp37m-linux_x86_64.whl