Decision forests in TensorFlow
summary
TensorFlow Decision Forests appears as a concrete library that trains random forests and gradient‑boosted trees inside Keras, with code showing a CSV→pandas→tf.data flow, one line to create a model, and defaults that often work without tuning. It natively handles numeric and categorical features so no preprocessing required, can print tree structures and report variable importances, and exports as a SavedModel usable with TensorFlow Serving.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:43Decision forests use Keras
- 2:03When should you use trees?
- 2:15It depends on your data
- 3:43When are neural networks best?
- 5:48Easier preprocessing
- 6:30Deploy with TensorFlow Serving
related talks
46mIntroduction to Machine Learning
53mA Taste of Random Decision Forests on Apache Spark
40mCART: Not only Classification and Regression Trees
57mFrom Bonsais to Random Forests
40mTensorFlow Wide & Deep: Advanced Classification the easy way
06mPërdorimi i Decision Trees dhe të dhënat për të marrë vendime
19mWhat's new in TensorFlow for R
26mMachine Learning with TensorFlow