Difference between revisions of "Machine Learning Resources"

From Wiki2
Jump to navigation Jump to search
Line 6: Line 6:
 
* [https://github.com/tensorflow/ranking Tensorflow Rank] (TensorFlow Ranking is a library for Learning-to-Rank (LTR) techniques on the TensorFlow platform. It contains the following components:
 
* [https://github.com/tensorflow/ranking Tensorflow Rank] (TensorFlow Ranking is a library for Learning-to-Rank (LTR) techniques on the TensorFlow platform. It contains the following components:
 
Commonly used loss functions including pointwise, pairwise, and listwise losses.
 
Commonly used loss functions including pointwise, pairwise, and listwise losses.
Commonly used ranking metrics like 'Mean Reciprocal Rank (MRR)' and 'Normalized Discounted Cumulative Gain (NDCG)'.
+
Commonly used ranking metrics like ''Mean Reciprocal Rank (MRR)'' and ''Normalized Discounted Cumulative Gain (NDCG)''.
'Multi-item (also known as groupwise) scoring functions'.
+
''Multi-item (also known as groupwise) scoring functions''.
'LambdaLoss' implementation for direct ranking metric optimization.
+
''LambdaLoss'' implementation for direct ranking metric optimization.
 
Unbiased Learning-to-Rank from biased feedback data.)
 
Unbiased Learning-to-Rank from biased feedback data.)
  

Revision as of 08:19, 10 July 2019

Key Frameworks:

Commonly used loss functions including pointwise, pairwise, and listwise losses. Commonly used ranking metrics like Mean Reciprocal Rank (MRR) and Normalized Discounted Cumulative Gain (NDCG). Multi-item (also known as groupwise) scoring functions. LambdaLoss implementation for direct ranking metric optimization. Unbiased Learning-to-Rank from biased feedback data.)

Neural Network Interpretability


Python Notebook Examples

Image Quality Assessment


Transfer Learning