train.py

import cv2
import csv
import random
import numpy as np
import tensorflow as tf

encoding = '123456789'
labels_file = 'labels.csv'

img_list = []
label_list = []

with open(labels_file, newline='') as csvfile:
    reader = csv.DictReader(csvfile)
    fieldnames = reader.fieldnames
    for row in reader:
        img_list.append(row[fieldnames[0]])
        label_list.append(encoding.index(row[fieldnames[1]]))

items=[]
for i in range(len(img_list)):
    items.append(i)
X = []
y = []
for i in random.sample(items,len(img_list)):
    img = cv2.imread(img_list[i], 0)
    img = tf.keras.preprocessing.image.img_to_array(img)
    X.append(img)
    y.append(label_list[i])

X = 1 - np.array(X).astype(float)/255 # invert and scale
y = np.array(y).astype(float)

model = tf.keras.Sequential()
model.add(tf.keras.layers.Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(24, 31, 1)))
model.add(tf.keras.layers.BatchNormalization())
model.add(tf.keras.layers.MaxPooling2D(pool_size=(2, 2)))
model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu'))
model.add(tf.keras.layers.BatchNormalization())
model.add(tf.keras.layers.MaxPooling2D(pool_size=(2, 2)))
model.add(tf.keras.layers.Conv2D(128, (3, 3), activation='relu'))
model.add(tf.keras.layers.BatchNormalization())
model.add(tf.keras.layers.Flatten())
model.add(tf.keras.layers.Dropout(rate=0.25))
model.add(tf.keras.layers.Dense(128, activation='relu'))
model.add(tf.keras.layers.Dropout(rate=0.5))
model.add(tf.keras.layers.Dense(9, activation='softmax'))
print(model.summary())

initial_learning_rate = 1e-3
lr_schedule = tf.keras.optimizers.schedules.ExponentialDecay(
    initial_learning_rate,
    decay_steps=100000,
    decay_rate=0.9,
)
model.compile(
    loss=tf.keras.losses.SparseCategoricalCrossentropy(),
    optimizer=tf.keras.optimizers.Adam(learning_rate=lr_schedule),
    metrics=[tf.keras.metrics.SparseCategoricalAccuracy()]
)
batch_size = 32
epochs = 80

model.fit(
    X,
    y,
    batch_size = batch_size,
    epochs = epochs,
    validation_split = 0.2,
    callbacks = [tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=5)]
)

import tensorflowjs as tfjs
tfjs.converters.save_keras_model(model, 'outputs/model_tfjs')