Mastering Computer Vision with TensorFlow 2.x

Preprocessing within the Keras–Python code

In this section, we will read the CSV file as X and Y data, where X is the image corresponding to each filename, and Y has 32 values for 16 key-points coordinate. We will then slice the Y data into 16 Yx and Yy coordinates for each of the key-points. The detailed steps are shown as follows:

1. Read the CSV file from the previous section using the standard Python command. Here, we use two CSVs, trainimgface.csv, and testimgface.csv, located in the faceimagestrain directory. You can use a different folder if you want:

train_path = 'faceimagestrain/trainimgface.csv'test_path = 'faceimagestrain/testimgface.csv'train_data = pd.read_csv(train_path) test_data = pd.read_csv(test_path)

2. Next, we find the ...