9648fe492010

博主好屯掖，我已經(jīng)成功訓(xùn)練了20萬次，把object_detection更換成你發(fā)的最新的文件换帜，然后運(yùn)行命令：python object_detection/export_inference_graph.py --input_type=image_tensor --pipeline_config_path=training/ssdlite_mobilenet_v2_coco.config --trained_checkpoint_prefix=training/model.ckpt-200000 --output_directory=ctree_inference_graph
成功生成了ctree_inference_graph文件夾阵赠，但是文件夾里面沒有任何文件涯塔，在cmd中的錯(cuò)誤提示為：
tensorflow.python.framework.errors_impl.InvalidArgumentError: Assign requires shapes of both tensors to match. lhs shape= [256] rhs shape= [1280]
[[{{node save/Assign_26}}]]

tensorflow.python.framework.errors_impl.InvalidArgumentError: Assign requires shapes of both tensors to match. lhs shape= [256] rhs shape= [1280]
[[node save/Assign_26 (defined at Y:\MobileNet-SSD\object_detection\exporter.py:241) ]]

InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [256] rhs shape= [1280]
[[node save/Assign_26 (defined at Y:\MobileNet-SSD\object_detection\exporter.py:241) ]]

tensorflow.python.framework.errors_impl.InvalidArgumentError: Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Assign requires shapes of both tensors to match. lhs shape= [256] rhs shape= [1280]
[[node save/Assign_26 (defined at Y:\MobileNet-SSD\object_detection\exporter.py:241) ]]

InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Assign requires shapes of both tensors to match. lhs shape= [256] rhs shape= [1280]
[[node save/Assign_26 (defined at Y:\MobileNet-SSD\object_detection\exporter.py:241) ]]
看上去是結(jié)構(gòu)不一樣無法保存，請求博主幫助清蚀！

目標(biāo)檢測實(shí)踐_tensorflow版SSD模型測試

致謝聲明 本文在學(xué)習(xí)《Tensorflow object detection API 搭建屬于自己的物體識(shí)別模型（2）——訓(xùn)練并使用自己的模型》的基礎(chǔ)上優(yōu)化并總結(jié)匕荸，此博客鏈接...

瀟灑坤

8653 22 13

9648fe492010

Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.735
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.971
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.896
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.590
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.747
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.142
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.768
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.798
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.677
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.809

博主你好，這是我訓(xùn)練出來的結(jié)果枷邪。請問里面的六個(gè)精確率和召回率中area=all榛搔、small、medium、large践惑，maxDets=100腹泌、10、1分別表示什么意思童本？IoU=0.50:0.95是指IoU在0.50到0.95之間吧真屯？IoU=0.50是指IoU剛好在0.50時(shí)嗎？我想應(yīng)該是指IoU大于等于0.50吧穷娱。最后的結(jié)果=-1表示什么意思绑蔫？

目標(biāo)檢測實(shí)踐_tensorflow版SSD訓(xùn)練自己的數(shù)據(jù)

致謝聲明 本文在學(xué)習(xí)《Tensorflow object detection API 搭建屬于自己的物體識(shí)別模型（2）——訓(xùn)練并使用自己的模型》的基礎(chǔ)上優(yōu)化并總結(jié)，此博客鏈接...

瀟灑坤

17426 17 21 2