Fase de entrenamiento del modelo

Instalar la librería de object Detection en Google Colab:

import os
%cd /content
!git clone --quiet https://github.com/tensorflow/models.git
%cd /content/models/
#!git checkout 58d19c67e1d30d905dd5c6e5092348658fed80af
!apt-get update && apt-get install -y -qq protobuf-compiler python-pil python-lxml python-tk
!pip install -q Cython contextlib2 pillow lxml matplotlib
!pip install -q pycocotools
%cd /content/models/research
!protoc object_detection/protos/*.proto --python_out=.
os.environ['PYTHONPATH'] += ':/content/models/research/:/content/models/research/slim/'
!python object_detection/builders/model_builder_test.py

Descargar los modelos pre-entrenados en este caso SSD + MobileNetV2

!wget --no-check-certificate http://download.tensorflow.org/models/object_detection/tf2/20200711/ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8.tar.gz \
-O /content/ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8.tar.gz