Skip to main content

Keras sequence-generator sweep: complete code

Train two digit classifiers using a Talos sequence generator. This page is the standalone companion to the walkthrough.

Prerequisites and execution​

Use Python 3.11–3.13 with the TensorFlow extra (talos[tensorflow]). The dataset is an offline scikit-learn fixture. From a writable experiment directory, save the following program as generator_example.py and execute python generator_example.py.

The walkthrough owns the data split, callback explanation and interpretation of metrics. The program is bounded to two trials; successful execution passes its result-row assertion.

Program​

import talos
import numpy as np
from sklearn.model_selection import train_test_split
from tensorflow.keras import Sequential, Model
from tensorflow.keras.layers import Input, Dense, Dropout, Conv2D, Flatten, concatenate
from talos.utils import SequenceGenerator

from sklearn.datasets import load_digits
x, y = load_digits(return_X_y=True)
x = x.reshape(-1, 8, 8, 1).astype('float32') / 16
x_train, x_val, y_train, y_val = train_test_split(
    x, y, train_size=144, test_size=36, stratify=y, random_state=17)

def digits_model(x_train, y_train, x_val, y_val, params):
    model = Sequential([Input(shape=(8, 8, 1)),
                        Conv2D(4, (3, 3), activation=params['activation']),
                        Flatten(), Dense(8, activation=params['activation']),
                        Dropout(params['dropout']), Dense(10, activation='softmax')])
    model.compile(optimizer=params['optimizer'], loss=params['losses'],
                  metrics=['accuracy', talos.utils.metrics.f1score])
    batches = SequenceGenerator(x=x_train, y=y_train,
                                batch_size=params['batch_size'], backend='tensorflow')
    out = model.fit(batches, epochs=params['epochs'],
                    validation_data=(x_val, y_val), verbose=0)
    return out, model

p = {'activation': ['relu', 'elu'], 'optimizer': ['adam'],
     'losses': ['sparse_categorical_crossentropy'], 'dropout': [.1],
     'batch_size': [16], 'epochs': [2]}

scan_object = talos.Scan(x=x_train, y=y_train, x_val=x_val, y_val=y_val,
                         model=digits_model, params=p, experiment_name='digits_generator',
                         round_limit=2, seed=17, backend='tensorflow')
assert len(scan_object.data) == 2

Result and failure boundaries​

scan_object.data contains two completed rows. Each callback trains from SequenceGenerator batches and validates on the separate 36-row array split. The network produces ten class probabilities per image. The program writes result and checkpoint artifacts to its experiment run directory.

The arrays must match (8, 8, 1) image inputs and integer digit labels. Use sparse categorical cross-entropy for those labels. Modern Keras does not accept workers in fit(); configure a supported sequence/PyDataset instead. See Generator for replayability limits when using external streams.

Return to the walkthrough for the procedure and failure diagnosis. Generator covers the input contract; Analyze covers the completed sweep.