The Algoliterator: Difference between revisions
From Algolit
(Inserted a table mentioned the authors) |
|||
| Line 1: | Line 1: | ||
by Algolit | by Algolit | ||
| + | |||
| + | {| | ||
| + | |- | ||
| + | | Type: || Algoliterary exploration | ||
| + | |- | ||
| + | | Technique: || Recursive Neural Network | ||
| + | |- | ||
| + | | Developed by: || Andrej Karphaty, Justin Johnson, Algolit | ||
| + | |} | ||
| + | |||
The Algoliterator is an interface to a neural network trained using the selection of digitized works of the Mundaneum archive. | The Algoliterator is an interface to a neural network trained using the selection of digitized works of the Mundaneum archive. | ||
Revision as of 15:25, 1 March 2019
by Algolit
| Type: | Algoliterary exploration |
| Technique: | Recursive Neural Network |
| Developed by: | Andrej Karphaty, Justin Johnson, Algolit |
The Algoliterator is an interface to a neural network trained using the selection of digitized works of the Mundaneum archive.
With the Algoliterator you can write a text in the style of the International Institutions Bureau. The Algoliterator starts by picking a sentence from the archive or corpus it was trained on. You can then continue writing yourself or, at any time, ask the Algoliterator to suggest a next sentence: the network will generate three new fragments based on the texts it has read. You can control level of training of the network and have it generate sentences based on primitive training, intermediate training or final training.
Sources: https://gitlab.constantvzw.org/algolit/algoliterator.clone
Concept, code & interface: Gijs de Heij & An Mertens
Original model: