pytorch-metric-learning
pytorch-metric-learning copied to clipboard
KevinMusgrave
The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.
Hi Kevin, I am looking to add the **Normalized Discounted Cumulative Gain** ([NDCG](https://en.wikipedia.org/wiki/Discounted_cumulative_gain)) metric to the `AccuracyCalculator` while also using the `GlobalEmbeddingSpaceTester`. The NDCG allows the use of non binary...
First, thank you for the wonderful library. It helped me achieve amazing results in my project, and your continious support is much appreciated! I am a it struggling with adding...
The easiest way might be to make the trainers extend [Ignite](https://github.com/pytorch/ignite), [Catalyst](https://github.com/catalyst-team/catalyst), [Lightning](https://github.com/PyTorchLightning/pytorch-lightning) etc.
I am training now on a set that has a lot of outliers (not clean). As such I get a lot of class mismatches from k nearest. There are only...
Hi Kevin Would you mind add my Selectively Contrastive Triplet loss (which is published in ECCV2020)? The major idea of this paper is to overcome the local minima during the...
I see the [sampler to generate a fixed set of triplets](https://kevinmusgrave.github.io/pytorch-metric-learning/samplers/#fixedsetoftriplets) and the [miner that uses that fixed set](https://kevinmusgrave.github.io/pytorch-metric-learning/miners/#embeddingsalreadypackagedastriplets). Is there an analogous sampler/miner duo for generating and using a...
Do you have any plans to implement RLL in this DML?
See #250
I am currently training a model with semi-supervised learning and I have two separate losses and miner as well. All of them use the same distance matrix, so when I...
Metadata
Owner
Metadata
The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.