Reference · Glossary
TensorFlow
Last updated
Google's open-source deep learning framework, often used through its high-level **Keras** API — define layers, compile with a loss/optimizer, and call `.fit()` to train, with strong tooling for mobile and production deployment (TensorFlow Lite, TensorFlow Serving).
#When to use
Production deployment paths that specifically want TensorFlow's serving/mobile tooling, or teams already standardized on Keras for readability.
#When not to
Following a research paper or a Hugging Face model card — most of those ship PyTorch weights first, so PyTorch is usually the smoother path.
#Example
import tensorflow as tf
model = tf.keras.Sequential([tf.keras.layers.Dense(1, input_shape=(3,))])
model.compile(optimizer="adam", loss="mse")
model.fit(x_train, y_train, epochs=5)