open-metric-learning

open-metric-learning copied to clipboard

Some changes for OML 3.0

1

**CHANGELOG** * Moved category wise metrics calculation logic from `EmbeddingMetrics` to functional metrics. * Removed `fnmr@fmr` metric from `EmbeddingMetrics` because we cannot guarantee correctness of its behaviour when postprocessor is...

AlekseySh

Differentiable version of FNMR@FMR metric to use it as loss

6

A paper for inspiration: [link](https://paperswithcode.com/paper/recall-k-surrogate-loss-with-large-batches).

AlekseySh

AlekseySh

An error will be reported during the metric calculation process of self-created data set

2

 We copied the format of the cars196 data set and made our own data set, with a total of 172 classes and a total of 450,000 pieces of data....

snow-wind-001

Is it possible to give an example of directly outputting the classification results?

4

https://open-metric-learning.readthedocs.io/en/latest/feature_extraction/python_examples.html We used the example of the above URL, but there is no example of directly outputting two photos and outputting the classification results. In order to allow more people...

snow-wind-001

AlekseySh

Example: export a model to onnx

15

Hi! Are you planning to provide a tool for exporting to various formats, for example, onnx?

Fissium

Make imports shortcuts

6

**Amendments Overview:**
 This pull request introduces the following improvements: - Simplified import statements in the 'losses' module. - Simplified imports from the 'miners' module. - Simplified import paths for the...

mimi030

Here are the steps to achieve goal: * Do all modality specific logic **only** in Dataset and Model. Rest of the classes have to be modality agnostic! * Stop using...

AlekseySh

Update _read_bytes function in images.py support for URL

1

Add support for fetching images from URLs in _read_bytes_image Previously, the function only supported reading image files from local paths. This commit extends its functionality to also fetch images from...

stoensin