# LSTM Benchmarks for Deep Learning Frameworks

arXiv: https://arxiv.org/abs/1806.01818 · HTML: https://mlsanity.com/paper/1806.01818/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jun 2018.

## Built on by (most cited first, 3 of 3)

- [DA-LSTM: A Dynamic Drift-Adaptive Learning Framework for Interval Load Forecasting with LSTM Networks](https://mlsanity.com/paper/2305.08767/) (2023)
- [Predicting Li-ion Battery Cycle Life with LSTM RNN](https://mlsanity.com/paper/2207.03687/) (2022)
- [CarbonNet: How Computer Vision Plays a Role in Climate Change? Application: Learning Geomechanics from Subsurface Geometry of CCS to Mitigate Global Warming](https://mlsanity.com/paper/2403.06025/) (2024)

## Abstract (arXiv)

This study provides benchmarks for different implementations of LSTM units between the deep learning frameworks PyTorch, TensorFlow, Lasagne and Keras. The comparison includes cuDNN LSTMs, fused LSTM variants and less optimized, but more flexible LSTM implementations. The benchmarks reflect two typical scenarios for automatic speech recognition, notably continuous speech recognition and isolated digit recognition. These scenarios cover input sequences of fixed and variable length as well as the loss functions CTC and cross entropy. Additionally, a comparison between four different PyTorch versions is included. The code is available online https://github.com/stefbraun/rnn_benchmarks.