Mastering Computer Vision with TensorFlow 2.x

The model config file

The model config file is a text file that defines the following important characteristics of the model:

• The model name
• The number of classes
• The image minimum/maximum dimensions
• The model parameters
• The location of the checkpoint, TFRecord, and map.pbtxt

During training, the model uses the config file to input and set parameters. You can find the list of config files in the TensorFlow directory under the following path:

models-master/research/object-detection/samples/configs

Note that at the time of writing, the preceding directory is available only in TensorFlow 1.x and not in 2.x. So, if you have TensorFlow 2.0 installed on your PC, use the alternative step discussed below to get the config file.

Alternatively, you ...