[英]TensorFlow object detection API fine tune checkpoint
我正在尝试从头开始训练模型,在这种情况下,我应该在 Protobuf 文件中放入 Fine_tune_checkpoint 值。
Nothing.没有。 It should initialize your entire model.
它应该初始化您的整个模型。
If for some reason that doesn't work for you, you can give it a checkpoint of something else entirely.如果由于某种原因对您不起作用,您可以给它一个完全其他东西的检查点。 The API builds the graph of the model, and compares between its variables to the variables in the checkpoint (both names and shapes), and every variable it doesn't find - it initializes.
API 构建模型的图形,并将其变量与检查点中的变量(名称和形状)进行比较,并且它没有找到每个变量 - 它进行初始化。
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.