DeepLabV3+预测单张图片代码实现和问题解决

最新推荐文章于 2026-04-29 09:39:15 发布

原创

最新推荐文章于 2026-04-29 09:39:15 发布 · 5.6k 阅读

标签

#DeepLabV3+ #export_model #infer

本文介绍了在使用DeepLabV3+模型进行图片预测时遇到的问题及解决方案。问题一是export_model.py导出模型时，需确保crop_size与预测图片尺寸一致；问题二是正向传播过程中遇到未知错误，通过尝试新方法成功解决，同时分享了使用pdb进行Python代码调试的经验。

问题一：在使用export_model.py进行导出模型并进行网络正向传播的过程中，出现错误:

2019-03-25 17:47:42.896918: W tensorflow/core/framework/allocator.cc:122] Allocation of 62720000 exceeds 10% of system memory.
2019-03-25 17:47:43.105534: W tensorflow/core/framework/allocator.cc:122] Allocation of 62720000 exceeds 10% of system memory.
2019-03-25 17:47:43.406195: W tensorflow/core/framework/allocator.cc:122] Allocation of 62720000 exceeds 10% of system memory.
2019-03-25 17:47:43.663981: W tensorflow/core/framework/allocator.cc:122] Allocation of 63438848 exceeds 10% of system memory.
2019-03-25 17:47:51.507348: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at spacetobatch_op.cc:219 : Invalid argument: padded_shape[0]=49 is not divisible by block_shape[0]=2
2019-03-25 17:47:53.206279: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at spacetobatch_op.cc:219 : Invalid argument: padded_shape[0]=49 is not divisible by block_shape[0]=2
Traceback (most recent call last):
  File "/home/pzn/anaconda3/envs/DeepLabV3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1334, in _do_call
    return fn(*args)
  File "/home/pzn/anaconda3/envs/DeepLabV3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1319, in _run_fn
    options, feed_dict, fetch_list, target_list, run_metadata)
  File "/home/pzn/anaconda3/envs/DeepLabV3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1407, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: padded_shape[0]=49 is not divisible by block_shape[0]=2
	 [[{
  
  {node import/xception_65/exit_flow/block2/unit_1/xception_module/separable_conv1_depthwise/depthwise/SpaceToBatchND}} = SpaceToBatchND[T=DT_FLOAT, Tblock_shape=DT_INT32, Tpaddings=DT_INT32, _device="/job:localhost/replica:0/task:0/device:GPU:0"](import/xception_65/exit_flow/block1/unit_1/xception_module/add, import/xception_65/exit_flow/block2/unit_1/xception_module/separable_conv1_depth

最低0.47元/天解锁文章