Most of the available material addresses 2D Convolutional Neural Networks for classifying images or 1D for text. I am looking for content for consultation on the use of CNN for classifying regular sequences or time series, not only string or text sequences.
Yes, indeed there is lot of work for 1-D cases. See a good work for example : https://keras.io/examples/imdb_cnn/
https://machinelearningmastery.com/cnn-models-for-human-activity-recognition-time-series-classification/
I recommend read 1D CNN with LSTM for sequence learning. Recently we applied it for Energy consumption prediction in the following research.
https://ieeexplore.ieee.org/abstract/document/8945363
Dear Manuel, you have here a good explanation and application of 1D CNN for time series data. The implementation using keras-tensorflow is also available in this blog post:
https://blog.goodaudience.com/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences-3a7ff801a2cf
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series