Mastering Computer Vision with TensorFlow 2.x

Executing export tflite graph.py from Google Colab

In Chapter 10, Object Detection Using R-CNN, SSD, and R-FCN, and Chapter 11, Deep Learning on Edge Devices with CPU/GPU Optimization, we used Google Colab to convert the checkpoint into a frozen graph. We will use the same method here, except we will import the configuration, checkpoint, and output directory as shown:

CONFIG_FILE = '/content/sample_data/train_pipeline.config'CHECKPOINT_PATH = '/content/sample_data/train_model.ckpt-2000'OUTPUT_DIR = '/content/sample_data'

Upload the files to Google Drive and then drag them into a folder in Google Colab called sample_data. You can create a different name instead of sample_data. After that, execute the following code:

Note that if you do not ...