tf.keras.layers.LSTM | TensorFlow Core v2.7.0
www.tensorflow.org › python › tfBased on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation.