Photo by Alina Grubnyak on Unsplash
Types of RNN in Deep Learning
RNN is most important in Deep Learning
Table of contents
No headings in the article.
Deep Learning is popular famous after few years and after winning some competition of data Science Before learning type of RNNs first learn about what is RNN (Recurrent Neural Network) RNN is special type of neural which specially designed for for solving sequence related task , NLP, image captioning, language translation, etc.
One to One In this RNN you give fixed size of the input and expect output in fixed size. This is also call plain Neural network. Example : Image classification.
One to many In this RNN you give fixed size of the input also your model give you output in sequences. Example : Image captioning.
Many to One In this RNN you give sequence size of the input and your model give output in fixed size. Example : Sentiment Analysis.
- Many to Many In this RNN you give sequence size of the input and your Model give output in sequence size. Example : Machine Translation.