Port from TFA the loss and split out strategies from loss

Implement a data sampler that given X, Y generate batch

Add a "layer" / loss that allows to project cluster in 2 or 3d using a different formula based of https://github.com/beringresearch/ivis

Current the model don't know what distance is used in the index when its reloaded as the variable is not set.

The threshold selected in the labels don't match the one displayed:

Label would say ['matching'] == 0.333
but the table would say -> 0.3111

Port the gradcam++ code

Port some of the SKlearn metrics such as

• adjusted rand
• v-measure

When having a multihead the indexer crash because we pass all the predict to it. Fix is to use output[0] by default and allows to specify one.

Readme content is outdated

Implement silouhette metrics and expose intra-cluster-distance and mean nearest-cluster distance

Figure out if we can have a speedup to reuse composed metric to avoid computing it twice (might not be needed)

We can do our own using a gpu accelerate lib

• https://medium.com/rapids-ai/scaling-knn-to-new-heights-using-rapids-cuml-and-dask-63410983acfe
• https://github.com/facebookresearch/faiss
• https://github.com/yahoojapan/NGT (vlad algorithm)

serving

• https://oss.redislabs.com/redisai/intro/

or use a fully fledge server

• https://github.com/vdaas/vald (NGT)
• https://github.com/milvus-io/milvus (FAISS)

Missing return prototype

Implement a mechanism to allows to schedule batch if needed be

Currently experiments only include image datasets. Adding a text dataset highlights the performance of TF.similarity on text data.

Fix documentation/docs/api/metrics.md

The Explain module is not being loaded correctly because it's path is incorrect. Loading saved models results in an error because of a missing custom_objects argument.

Benchmark various models/loss on various use-cases -- see https://github.com/KevinMusgrave/powerful-benchmarker/

Code allows to use them but we don't deal with their serialization, reloading

in documentation/docs/api/eval_metrics.md

deal with the if windows in the setup.py to use faiss-cpu on windows architecture

Signal-to-Noise Ratio: A Robust Distance Metric for Deep Metric Learning https://openaccess.thecvf.com/content_CVPR_2019/papers/Yuan_Signal-To-Noise_Ratio_A_Robust_Distance_Metric_for_Deep_Metric_Learning_CVPR_2019_paper.pdf

Add a README to make it easier for users to get started with TF.similarity serving. The README will include setup instructions, documentation and a demonstration.

Investigate why the metrics in history.history don't seems to match the values during training.

Rewrite the readme with new instructions and information.

Allows at model time to specify how many hard triplet are stored in the pool

Make TF similarity a custom model

Its unclear if we need to provide a set of layers for the embedding -- not exposing it publicly until this is thought out more.

https://deepai.org/publication/circle-loss-a-unified-perspective-of-pair-similarity-optimization

in triplet_loss serialization, adding reduction in the from_config, triggers an error unsure why, Should be investigated.

Initially discovered by Ian, TensorFlow Similarity is not compatible with TF 2.2.x and therefore the current Hello World notebook is set to TF 2.1.x

https://colab.sandbox.google.com/drive/1HRK4zLSAzGHwoM6dz2A1ygHSeVQ3VHdI#scrollTo=ST8JbEUrldut

From collecting data to transfert to train to serve

.gitignore is needed to let Git know that it should ignore certain files and not track them

Front end needs significant refactoring, including utilizing Vue.js best practices and enhancing UI/UX.

Call mkdocs in the action that merge

Make sure to fix the setup.py when all the PR are done and dusted.

implement tf.dataset SingleShot and Multishot sampler

Fix documentation/docs/api/losses.md

Ensure callback can access the distances computed during the loss computation. Potentially store them in the loss object by subclassing the loss wrapper?

https://github.com/bnu-wangxun/Deep_Metric

Implement quad loss