报错Dims of all Inputs(X) must be the same, but received input 2 dim is:-1, 2, -1, -1 not equal to input 0 dim:-1, 3, -1, -1.
Created by: LongLivecn
我用自己制作的coco格式的数据集在给的yolov3-darknet.yml配置文件上跑通了，但是精度太低，后来我又放到dcn/yolov3_r50vd_dcn_obj365_pretrained_coco.yml这个配置文件上训练，结果报错。 报错信息：
Error Message Summary:
InvalidArgumentError: Dims of all Inputs(X) must be the same, but received input 2 dim is:-1, 2, -1, -1 not equal to input 0 dim:-1, 3, -1, -1. [Hint: Expected input_dims[i] == input_dims[0], but received input_dims[i]:-1, 2, -1, -1 != input_dims[0]:-1, 3, -1, -1.] at (/paddle/paddle/fluid/operators/stack_op.cc:46) [operator < stack > error] 请大佬指教
完整报错信息：
aistudio@jupyter-213340-537112:~/PaddleDetection$ python -u tools/train.py -c configs/dcn/yolov3_r50vd_dcn_obj365_pretrained_coco.yml --eval warning: cornerpool_lib not found, compile in ext_op at first if needed Traceback (most recent call last): File "tools/train.py", line 376, in main() File "tools/train.py", line 123, in main train_fetches = model.train(feed_vars) File "/home/aistudio/PaddleDetection/ppdet/modeling/architectures/yolov3.py", line 156, in train return self.build(feed_vars, mode='train') File "/home/aistudio/PaddleDetection/ppdet/modeling/architectures/yolov3.py", line 85, in build gt_score, targets) File "/home/aistudio/PaddleDetection/ppdet/modeling/anchor_heads/yolo_head.py", line 285, in get_loss self.prefix_name) File "/home/aistudio/PaddleDetection/ppdet/modeling/losses/yolo_loss.py", line 59, in call mask_anchors, self._ignore_thresh) File "/home/aistudio/PaddleDetection/ppdet/modeling/losses/yolo_loss.py", line 122, in _get_fine_grained_loss num_classes) File "/home/aistudio/PaddleDetection/ppdet/modeling/losses/yolo_loss.py", line 240, in _split_output clss, axis=1), perm=[0, 1, 3, 4, 2]) File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/layers/nn.py", line 9164, in stack attrs={'axis': axis}) 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 2525, 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 1880, in init self.desc.infer_shape(self.block.desc) paddle.fluid.core_avx.EnforceNotMet:
C++ Call Stacks (More useful to developers):
0 std::string paddle::platform::GetTraceBackString<std::string const&>(std::string const&, char const*, int) 1 paddle::platform::EnforceNotMet::EnforceNotMet(std::string const&, char const*, int) 2 paddle::operators::StackOp::InferShape(paddle::framework::InferShapeContext*) const 3 paddle::framework::OpDesc::InferShape(paddle::framework::BlockDesc const&) const
Python Call Stacks (More useful to users):
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/framework.py", line 2525, in append_op attrs=kwargs.get("attrs", None)) 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/layers/nn.py", line 9164, in stack attrs={'axis': axis}) File "/home/aistudio/PaddleDetection/ppdet/modeling/losses/yolo_loss.py", line 240, in _split_output clss, axis=1), perm=[0, 1, 3, 4, 2]) File "/home/aistudio/PaddleDetection/ppdet/modeling/losses/yolo_loss.py", line 122, in _get_fine_grained_loss num_classes) File "/home/aistudio/PaddleDetection/ppdet/modeling/losses/yolo_loss.py", line 59, in call mask_anchors, self._ignore_thresh) File "/home/aistudio/PaddleDetection/ppdet/modeling/anchor_heads/yolo_head.py", line 285, in get_loss self.prefix_name) File "/home/aistudio/PaddleDetection/ppdet/modeling/architectures/yolov3.py", line 85, in build gt_score, targets) File "/home/aistudio/PaddleDetection/ppdet/modeling/architectures/yolov3.py", line 156, in train return self.build(feed_vars, mode='train') File "tools/train.py", line 123, in main train_fetches = model.train(feed_vars) File "tools/train.py", line 376, in main()
Error Message Summary:
InvalidArgumentError: Dims of all Inputs(X) must be the same, but received input 2 dim is:-1, 2, -1, -1 not equal to input 0 dim:-1, 3, -1, -1. [Hint: Expected input_dims[i] == input_dims[0], but received input_dims[i]:-1, 2, -1, -1 != input_dims[0]:-1, 3, -1, -1.] at (/paddle/paddle/fluid/operators/stack_op.cc:46) [operator < stack > error]
配置文件： architecture: YOLOv3 use_gpu: true max_iters: 20000 log_smooth_window: 20 save_dir: output snapshot_iter: 500 metric: COCO pretrain_weights: https://paddlemodels.bj.bcebos.com/object_detection/ResNet50_vd_dcn_db_obj365_pretrained.tar weights: output/yolov3_r50vd_dcn_db_obj365_pretrained_coco/model_final num_classes: 3 use_fine_grained_loss: true
YOLOv3: backbone: ResNet yolo_head: YOLOv3Head use_fine_grained_loss: true
ResNet: norm_type: sync_bn freeze_at: 0 freeze_norm: false norm_decay: 0. depth: 50 feature_maps: [3, 4, 5] variant: d dcn_v2_stages: [5]
YOLOv3Head: anchor_masks: [[6, 7, 8], [3, 4, 5], [0, 1, 2]] anchors: [[10, 13], [16, 30], [33, 23], [30, 61], [62, 45], [59, 119], [116, 90], [156, 198], [373, 326]] norm_decay: 0. yolo_loss: YOLOv3Loss nms: background_label: -1 keep_top_k: 100 nms_threshold: 0.45 nms_top_k: 1000 normalized: false score_threshold: 0.01
YOLOv3Loss:
batch_size here is only used for fine grained loss, not used
for training batch_size setting, training batch_size setting
is in configs/yolov3_reader.yml TrainReader.batch_size, batch
size here should be set as same value as TrainReader.batch_size
batch_size: 8 ignore_thresh: 0.7 label_smooth: true use_fine_grained_loss: true
LearningRate: base_lr: 0.001 schedulers:
- !PiecewiseDecay
gamma: 0.1
milestones:
- 5000
- 10000
- 15000
- !LinearWarmup start_factor: 0. steps: 400
OptimizerBuilder: optimizer: momentum: 0.9 type: Momentum regularizer: factor: 0.0005 type: L2
READER: 'yolov3_enhance_reader.yml'
TrainReader: inputs_def: fields: ['image', 'gt_bbox', 'gt_class', 'gt_score'] num_max_boxes: 50 use_fine_grained_loss: false #true dataset: !COCODataSet image_dir: train anno_path: annotations/instances_train.json dataset_dir: dataset/coco2020 with_background: false sample_transforms: - !DecodeImage to_rgb: True - !RandomCrop {} - !RandomFlipImage is_normalized: false - !NormalizeBox {} - !PadBox num_max_boxes: 50 - !BboxXYXY2XYWH {} batch_transforms: - !RandomShape sizes: [320, 352, 384, 416, 448, 480, 512, 544, 576, 608] random_inter: True - !NormalizeImage mean: [0.485, 0.456, 0.406] std: [0.229, 0.224, 0.225] is_scale: True #False is_channel_first: false - !Permute to_bgr: false channel_first: True # Gt2YoloTarget is only used when use_fine_grained_loss set as true, # this operator will be deleted automatically if use_fine_grained_loss # is set as false - !Gt2YoloTarget anchor_masks: [[6, 7, 8], [3, 4, 5], [0, 1, 2]] anchors: [[10, 13], [16, 30], [33, 23], [30, 61], [62, 45], [59, 119], [116, 90], [156, 198], [373, 326]] downsample_ratios: [32, 16, 8] batch_size: 4 shuffle: true drop_last: true worker_num: 2 bufsize: 16 use_process: true
EvalReader: inputs_def: image_shape: [3, 608, 608] fields: ['image', 'im_size', 'im_id'] num_max_boxes: 50 dataset: !COCODataSet dataset_dir: dataset/coco2020 anno_path: annotations/instances_val.json image_dir: val with_background: false sample_transforms: - !DecodeImage to_rgb: True with_mixup: false - !ResizeImage interp: 2 target_size: 608 - !NormalizeImage mean: [0.485, 0.456, 0.406] std: [0.229, 0.224, 0.225] is_scale: True #False is_channel_first: false - !PadBox num_max_boxes: 50 - !Permute to_bgr: false channel_first: True batch_size: 4 drop_empty: false worker_num: 2 bufsize: 16
TestReader: inputs_def: image_shape: [3, 608, 608] fields: ['image', 'im_size', 'im_id'] dataset: !ImageFolder dataset_dir: dataset/coco2020 anno_path: annotations/instances_val.json with_background: false sample_transforms: - !DecodeImage to_rgb: True with_mixup: false - !ResizeImage interp: 2 target_size: 608 - !NormalizeImage mean: [0.485, 0.456, 0.406] std: [0.229, 0.224, 0.225] is_scale: True #False is_channel_first: false - !Permute to_bgr: false channel_first: True batch_size: 1