Epoch

Appears in 2 papers · 1 tutorial

One complete pass through all training examples.

As used in Paper 02 — The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain →

One complete pass through all training examples. Training typically requires many epochs before convergence. Each epoch gives every training example a chance to update the weights.

As used in Paper 03 — Learning Representations by Back-propagating Errors →

One complete pass through the entire training dataset. Networks typically train for many epochs — sometimes hundreds — before converging. Each epoch gives every training example the chance to update the weights.

As used in Fine-Tuning & Model Customization →

One full pass through the entire training dataset. Fine-tuning often uses just 1–3. (M03)

Paper 02 — The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain → Paper 03 — Learning Representations by Back-propagating Errors → Fine-Tuning & Model Customization →