[英]Where can I find the label map file(pbtxt) of ssd_mobilenet_v1_coco when using tensorflow?
I learn object detection on windows 10 with tensorflow object detection .我使用tensorflow object detection在 windows 10 上学习对象检测。
I download ssd_mobilenet_v1_coco_2018_01_28.tar.gz from https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md我从https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md下载 ssd_mobilenet_v1_coco_2018_01_28.tar.gz
After I unzipped the ssd_mobilenet_v1_coco_2018_01_28.tar.gz
file, I didn't find the pbtxt file.我解压
ssd_mobilenet_v1_coco_2018_01_28.tar.gz
文件后,没有找到pbtxt文件。
Where can I find the related pbtxt file of ssd_mobilenet_v1_coco? ssd_mobilenet_v1_coco的相关pbtxt文件在哪里可以找到?
I know that there some pbtxt files in models-master\\research\\object_detection\\data
folder, but which file is related to ssd_mobilenet_v1_coco?我知道在
models-master\\research\\object_detection\\data
文件夹中有一些 pbtxt 文件,但是哪个文件与 ssd_mobilenet_v1_coco 相关?
The label map is not specific to an architecture, but rather to a dataset (which classes you support, and accordingly you should set the number of channels of the classification layer).标签图并不特定于架构,而是特定于数据集(您支持哪些类,因此您应该设置分类层的通道数)。 Therefore, you simply need the label map which corresponds to coco, which is
object_detection/data/mscoco_label_map.pbtxt
.因此,您只需要与 coco 对应的标签图,即
object_detection/data/mscoco_label_map.pbtxt
。
Dataset is implemented in the model.数据集在模型中实现。 Model is delivered by tar.gz or .zip.
模型由 tar.gz 或 .zip 提供。 If you use pretrained basic model, then label map can be found code tree githup object_detection/data/mscoco_label_map.pbtxt as netanel-sam explains.
如果您使用预训练的基本模型,则可以在代码树 githup object_detection/data/mscoco_label_map.pbtxt 中找到标签图,如 netanel-sam 解释的那样。
But if you start to train your pretrained model and add items to be detected to your dataset and start to deliver your modified model, then your must offer your label map also and there is no better way than include it to the .tar.gz or .zip.但是,如果您开始训练您的预训练模型并将要检测的项目添加到您的数据集中并开始交付您修改后的模型,那么您还必须提供您的标签地图,并且没有比将其包含到 .tar.gz 或。压缩。 Same situation is with lite-model, because conversion from trainable model to lite often loses items from dataset.
lite-model 的情况相同,因为从可训练模型到 lite 的转换通常会丢失数据集中的项目。 Lite-model uses also other format to labelmap than basic model.
Lite-model 还使用其他格式来标记地图而不是基本模型。 Confusing?
令人困惑?
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.