何老师,第6章使用已有模型验证时报模型版本低: tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 **X Model directory: /home/chenhao/models/facenet/20170512-110547/ Metagraph file: model-20170512-110547.meta Checkpoint file: model-20170512-110547.ckpt-250000 WARNINGThe s**ed meta_graph is possibly from an older release: ‘model_variables’ collection should be of type ‘byte_list’, but instead is of type ‘node_list’. 然后很长时间后自己就killed了。目前我环境ubuntu 16.04 ,tensorflow 1.4.0 ,python 2.7
第三章运行train_image_classifier的时候,提示说ValueError: You must supply the dataset directory with —dataset_dir,可是我明明按照书上的操作建立了satellite文件夹和其中的data文件夹,怎么会说没有提供数据集的路径呢?
何老师,第6章使用已有模型验证时报模型版本低:
The s**ed meta_graph is possibly from an older release:
tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 **X
Model directory: /home/chenhao/models/facenet/20170512-110547/
Metagraph file: model-20170512-110547.meta
Checkpoint file: model-20170512-110547.ckpt-250000
WARNING
‘model_variables’ collection should be of type ‘byte_list’, but instead is of type ‘node_list’.
然后很长时间后自己就killed了。目前我环境ubuntu 16.04 ,tensorflow 1.4.0 ,python 2.7
第三章运行train_image_classifier的时候,提示说ValueError: You must supply the dataset directory with —dataset_dir,可是我明明按照书上的操作建立了satellite文件夹和其中的data文件夹,怎么会说没有提供数据集的路径呢?
第五章中,安装了tensorflow object detection api 后,不是模型都有了吗,为啥下一节中在执行训练好的模型中,还要再下载预训练模型?
有疑问的可进群互相交流,群号:154114989
出现导入cv2错误,我用的windows server R2012 64位版本,网上查的方法都用过了,还是没解决,希望老师给点建议。Python3.6.5,错误提示为:dll load failed