AI Understanding: What is an Elephant? | by Gergely D. Németh | Oct, 2020
The image above reveals seven various elephant photos and their DenseNet  classifications carried out in Keras with pre-trained ImageNet weights. As we are able to see, in lots of instances, the mannequin acknowledges the format (drawing, plush) as a substitute of the animal within the picture. Only two of the predictions has an elephant within the top5 predictions.
The photos within the image above:
- Reproduction of an illustration from early printed version of Speculum Humanae Salvationis (1430s)
- André Thevet. Cosmographie de Levant par F. André Thevet d’Angoulême. Revue et augmentée de plusieurs figures, Lyon, Jean Tournes et Guil.Gazeau, 1556.
- Objecten Wajangfiguur van perkament voorstellende een olifant Bandung (stadsgemeente) voor 1976
- Chess Piece, Bishop late 17th–early 18th century
- The elephants portray on T shirts that had been offered later. by Dennis Jarvis
- Blauer Elephant ClipArt von mir mit GIMP gemacht
- Picture of an opulent elephant, personal photograph
The elephants are iconic animals of wildlife. According to Save The Elephants:
Elephants are Africa’s gardeners and panorama engineers, planting seeds and creating habitat wherever they roam.
Without pressing motion to save lots of their species, elephants might disappear from the wild inside a single era.
Approximately 100,000 elephants in Africa had been killed for his or her ivory in simply three years between the years 2010 & 2012.
: I need to donate the income of this publish to save lots of elephants. Therefore, each time the quantity for an African Elephant Virtual Adoption donation ($55) on WWF Gifts reached, I’ll donate it to the organisation.
 “elephant.” Merriam-Webster.com. 2011. https://www.merriam-webster.com (22 Oct 2020).
 Mikolov, T., Sutskever, I., Chen, Okay., Corrado, G. S., & Dean, J. (2013). Distributed representations of phrases and phrases and their compositionality. In Advances in neural info processing techniques (pp. 3111–3119).
 Mikolov, T., Chen, Okay., Corrado, G., & Dean, J. (2013). Efficient estimation of phrase representations in vector area. arXiv preprint arXiv:1301.3781.
 Devlin, J., Chang, M. W., Lee, Okay., & Toutanova, Okay. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
 Honnibal, M., & Montani, I. (2017). spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing.
 Krasin I., Duerig T., Alldrin N., Ferrari V., Abu-El-Haija S., Kuznetsova A., Rom H., Uijlings J., Popov S., Kamali S., Malloci M., Pont-Tuset J., Veit A., Belongie S., Gomes V., Gupta A., Sun C., Chechik G., Cai D., Feng Z., Narayanan D., Murphy Okay.
OpenImages: A public dataset for large-scale multi-label and multi-class picture classification, 2017.
Available from https://storage.googleapis.com/openimages/internet/index.html.
 Deng, J., Dong, W., Socher, R., Li, L. J., Li, Okay., & Fei-Fei, L. (2009, June). Imagenet: A big-scale hierarchical picture database. In 2009 IEEE convention on pc imaginative and prescient and sample recognition (pp. 248–255). Ieee.
 Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, Okay. Q. (2017). Densely linked convolutional networks. In Proceedings of the IEEE convention on pc imaginative and prescient and sample recognition (pp. 4700–4708).