我的数据集图像尺寸是4096*3000的，调整输入之后，训练保存

[njflove]

问题确认 Search before asking

• 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer.

请提出你的问题 Please ask your question

报错信息：
[11/19 12:43:36] ppdet.data.source.coco INFO: Load [2733 samples valid, 0 samples invalid] in file /configs/coco_nic/train.json.
W1119 12:43:36.647198 20760 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.6, Driver API Version: 12.1, Runtime API Version: 11.2
W1119 12:43:36.653468 20760 gpu_resources.cc:164] device: 0, cuDNN Version: 8.1.
W1119 12:43:39.662191 20760 gpu_resources.cc:299] WARNING: device: . The installed Paddle is compiled with CUDNN 8.2, but CUDNN version in your machine is 8.1, which may cause serious incompatible bug. Please recompile or reinstall Paddle with compatible CUDNN version.
Traceback (most recent call last):
File "/home/PaddleYOLO/tools/train.py", line 202, in
main()
File "/home/PaddleYOLO/tools/train.py", line 198, in main
run(FLAGS, cfg)
File "/home/PaddleYOLO/tools/train.py", line 151, in run
trainer.train(FLAGS.eval)
File "/home/PaddleYOLO/ppdet/engine/trainer.py", line 548, in train
outputs = model(data)
File "/usr/local/lib/python3.10/dist-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddle/distributed/parallel.py", line 528, in forward
outputs = self._layers(*inputs, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/home/PaddleYOLO/ppdet/modeling/architectures/meta_arch.py", line 60, in forward
out = self.get_loss()
File "/home/PaddleYOLO/ppdet/modeling/architectures/yolov8.py", line 106, in get_loss
return self._forward()
File "/home/PaddleYOLO/ppdet/modeling/architectures/yolov8.py", line 77, in _forward
neck_feats = self.neck(body_feats, self.for_mot)
File "/usr/local/lib/python3.10/dist-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/home/PaddleYOLO/ppdet/modeling/necks/yolov8_pafpn.py", line 199, in forward
f_concat0 = paddle.concat([up_feat0, c5], 1)
File "/usr/local/lib/python3.10/dist-packages/paddle/tensor/manipulation.py", line 1263, in concat
return _C_ops.concat(input, axis)
ValueError: (InvalidArgument) The 2-th dimension of input[0] and input[1] is expected to be equal.But received input[0]'s shape = [2, 640, 48, 64], input[1]'s shape = [2, 640, 47, 64].
[Hint: Expected inputs_dims[0][j] == inputs_dims[i][j], but received inputs_dims[0][j]:48 != inputs_dims[i][j]:47.] (at ../paddle/phi/kernels/funcs/concat_funcs.h:83)

reader配置：
input_height: &input_height 3000
input_width: &input_width 4096
input_size: &input_size [*input_height, *input_width]
mosaic_epoch: &mosaic_epoch 490 # last 10 epochs close mosaic, totally 500 epochs as default

在不修改原始图像尺寸的情况下，如何修改才能使模型能支持我的图像输入尺寸？