C.6: DeepLIFT: Difference between revisions

From Circular Twain - Reference Implementations Wiki
Jump to navigationJump to search
No edit summary
No edit summary
 
Line 5: Line 5:
• '''Reference, URL''': https://github.com/kundajelab/deeplift
• '''Reference, URL''': https://github.com/kundajelab/deeplift


• '''Relevant Domain/Industry''': Research
• '''Applicable Business Category''': Research


• '''Application in relevant Projects/Initiatives''': N.A.
• '''Application in relevant Projects/Initiatives''': N.A.


• '''Type''': Library
• '''Asset Type''': Library


• '''AI Breadth''': ML
• '''AI Breadth''': Machine Learning


• '''Learning Ability''':
• '''Learning Ability''': N.A.


• '''Related technologies''': Keras, Tensorflow
• '''Related technologies''': Keras, Tensorflow


• '''Applicable Research Area''':
• '''Applicable Research Area''': Explainable Ai


• '''Applicable Technical Category''':
• '''Applicable Technical Category''': Machine learning
 
• '''Applicable Business Category''':
 
• '''Asset Type''':


• '''License Information''': MIT license (MIT)
• '''License Information''': MIT license (MIT)

Latest revision as of 10:31, 27 December 2024

Short Description: A method for decomposing the output prediction of a neural network on a specific input and assigns contribution scores to neurons.

Reference, URL: https://github.com/kundajelab/deeplift

Applicable Business Category: Research

Application in relevant Projects/Initiatives: N.A.

Asset Type: Library

AI Breadth: Machine Learning

Learning Ability: N.A.

Related technologies: Keras, Tensorflow

Applicable Research Area: Explainable Ai

Applicable Technical Category: Machine learning

License Information: MIT license (MIT)

Related to circularity and sustainability: No

Audience: Developers



Go back to the AI portfolio of reference implementations