tensorflow/recommenders issues and pull requests

#752 - cleanup of deprecated test methods

Pull Request - State: closed - Opened by copybara-service[bot] 6 days ago - 1 comment

#751 - Clippy_adagrad raised an error: NotImplementedError: _resource_apply_dense must be implemented in subclasses.

Issue - State: open - Opened by ArachisTong 7 days ago

#750 - Update featurization.ipynb

Pull Request - State: open - Opened by aniruthraj 14 days ago - 1 comment

#749 - Bug Fix factorized_top_k.py

Pull Request - State: open - Opened by younssas 21 days ago - 1 comment

#748 - Title: ValueError when initializing tfrs.metrics.FactorizedTopK with a candidates dataset.

Issue - State: open - Opened by malinphy about 1 month ago - 1 comment

#747 - Remove special casing by sparse core as it did not work as intended

Pull Request - State: closed - Opened by copybara-service[bot] about 2 months ago - 1 comment

#745 - Revert requirement of tensorflow-macos package for MacOS platforms

Pull Request - State: open - Opened by marygiotandoi 2 months ago - 1 comment

#744 - BruteForce layer not working after loading a saved model

Issue - State: open - Opened by Toby1218 2 months ago

#743 - Update TFRS DCN tutorial notebook to use images from GitHub.

Pull Request - State: closed - Opened by copybara-service[bot] 2 months ago - 2 comments

#742 - Add DCN images to assets directory so that they can be loaded from git.

Pull Request - State: closed - Opened by copybara-service[bot] 2 months ago - 1 comment

#741 - Special case batch_size setting by hardware type.

Pull Request - State: closed - Opened by copybara-service[bot] 3 months ago - 1 comment

#740 - [Question] Uniform query size for Listwise Ranking?

Issue - State: open - Opened by kushagra-ch 3 months ago

#739 - Broken Links in Listwise Ranking Tutorial Page

Issue - State: open - Opened by james47 3 months ago

#738 - Add support for multipoint query retrieval to TFRS Retrieval task.

Pull Request - State: closed - Opened by copybara-service[bot] 3 months ago - 1 comment

#737 - Question: How can I utilize categories and other features for new users in the Sequential Retriever

Issue - State: open - Opened by hishambawa 3 months ago

#736 - cannot run initiate metrics in Basic Retrieval notebook

Issue - State: open - Opened by uncle-sann 4 months ago - 1 comment

#735 - feat: Extend 'sample_listwise' to support more movielens Dataset features

Pull Request - State: open - Opened by keivanipchihagh 5 months ago - 3 comments

#734 - Fix a bug in loss metrics calculation, and allow specifying `score_mask` in `tfrs.tasks.Retrieval.call`

Pull Request - State: closed - Opened by copybara-service[bot] 5 months ago - 1 comment

#733 - Significant drop in the model's performance metric (Top K Accuracy) when we go from 1 GPU to 2 or 4 GPUs.

Issue - State: open - Opened by hugoferrero 6 months ago

#732 - How can I use sample_weights for retrieval task?

Issue - State: open - Opened by modic23 6 months ago

#731 - Please Help ValueError: Cannot convert '('c', 'o', 'u', 'n', 't', 'e', 'r')' to a shape. Found invalid entry 'c' of type '<class 'str'>'.

Issue - State: open - Opened by Levine-Sam 6 months ago - 1 comment

#728 - TypeError: Cannot convert 1 to EagerTensor of dtype string

Issue - State: open - Opened by flynhigher 7 months ago

#727 - Fix for ValueError: The shape of the target variable and the shape of the target value in variable.assign(value) must match

Pull Request - State: open - Opened by flynhigher 7 months ago - 1 comment

#726 - ValueError: The shape of the target variable and the shape of the target value in `variable.assign(value)` must match in layers/factorized_top_k.py

Issue - State: open - Opened by flynhigher 7 months ago

#725 - Fix for 'update_ops' being None in metrics/factorized_top_k.py

Pull Request - State: open - Opened by flynhigher 7 months ago - 1 comment

#724 - Error: inputs' should be zero or more (nested) Tensors. Received 'None' with type '<class 'NoneType'>'

Issue - State: open - Opened by flynhigher 7 months ago

#723 - Fix for 'Variable' object has no attribute 'read_value' in layers/factorized_top_k.py

Pull Request - State: open - Opened by flynhigher 7 months ago - 1 comment

#722 - Error: 'Variable' object has no attribute 'read_value'

Issue - State: open - Opened by flynhigher 7 months ago

#721 - Add sparse_core_embedding_config to TPUEmbeddingLayer and add device assignment to VF in TPU distribute strategy.

Pull Request - State: closed - Opened by copybara-service[bot] 7 months ago

#720 - Very low topk %, real-world accuracy around 40% at k=5000

Issue - State: open - Opened by pthieu 7 months ago

#719 - load_weights can't work

Issue - State: open - Opened by mahaidongs 8 months ago - 1 comment

#718 - [Question] two-tower-model + infoNCE how to optimize

Issue - State: open - Opened by unshaven 8 months ago - 1 comment

#717 - fix keras compatibility bug in factorized_top_k

Pull Request - State: open - Opened by getumen 8 months ago - 10 comments

#716 - GDCN implementation

Issue - State: open - Opened by an-tran528 9 months ago - 2 comments

#715 - add tf recommenders difference

Pull Request - State: closed - Opened by ZhaoyueCheng 9 months ago

#714 - Upselling recommendation with NDR for implicit feedback (millions users, few items)

Issue - State: open - Opened by lpiscusc 9 months ago - 1 comment

#713 - Package 'tensorflow-macos' is listed as a dependency of itself

Issue - State: open - Opened by marygiotandoi 10 months ago - 3 comments

#712 - tfrs.metrics.FactorizedTopK

Issue - State: open - Opened by soheil-asgari 10 months ago - 14 comments

#711 - Plot TFRS model architecture

Issue - State: open - Opened by lpiscusc 10 months ago - 1 comment

#710 - Question - Anyone Trained Retrieval Distributed Successfully?

Issue - State: open - Opened by BrianMiner 10 months ago

#709 - The loss computed in the last step is significantly smaller compared to the previous steps

Issue - State: open - Opened by rayuron 11 months ago

#708 - add multi layer DCN as a building block for DCN V2 (DLRM_DCN) model and add the option to not concat dense before top stack

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#707 - [Question] Examples of Retrieval and Ranking Models in a Single Sequential Model?

Issue - State: open - Opened by rlcauvin 12 months ago - 5 comments

#706 - RemoveAccidentalHits Not Work Correctly

Issue - State: open - Opened by mustfkeskin about 1 year ago - 8 comments

#705 - Approach for Small Number of Items in Catalog

Issue - State: open - Opened by chiragdaryani about 1 year ago - 2 comments

#704 - Add support EmbeddingFeature V2.

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago - 1 comment

#703 - Incremental retrieval model training with Hashing method

Issue - State: open - Opened by nicewenhui about 1 year ago - 2 comments

#702 - combine listwise ranking and feature preprocessing

Issue - State: open - Opened by hyyan112 about 1 year ago

#701 - Cannot install on Apple M1

Issue - State: open - Opened by drljem about 1 year ago

#700 - Zero factorizedtopk accuracy

Issue - State: closed - Opened by thani-ath-nain-khurshid over 1 year ago - 2 comments

#699 - Retrieval: shared features between query and candidate tower

Issue - State: closed - Opened by JV-Nunes over 1 year ago - 13 comments

#698 - Fix to restore from SC checkpoints

Pull Request - State: open - Opened by copybara-service[bot] over 1 year ago - 1 comment

#697 - Plumb the pipeline_execution_with_tensor_core param in PartialTpuEmbedding class.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago

#696 - How can I add Mean Average Precision metric?

Issue - State: open - Opened by ftcister over 1 year ago - 1 comment

#695 - Use same embedding initializer for Keras embeddings as used in TpuEmbedding

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago

#694 - recommendation results not making sense

Issue - State: open - Opened by riyaj8888 over 1 year ago - 2 comments

#693 - Fix typo in featurization.ipynb

Pull Request - State: open - Opened by vidhant over 1 year ago - 2 comments

#692 - [Question] Serving a Retrieval Model with multiple signatures

Issue - State: closed - Opened by ericy51 over 1 year ago - 1 comment

#691 - How to know better evaluation metrics for Retrieval or Rank will translate into better actual recommendations in the case of MovieLens dataset?

Issue - State: open - Opened by houghtonweihu over 1 year ago - 1 comment

#690 - How to know the Rank model improves the output of the Retrieval model in the case of MovieLens dataset?

Issue - State: open - Opened by houghtonweihu over 1 year ago - 15 comments

#689 - Enable weighted factorized topK metric calculation.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago - 1 comment

#688 - candidates argument for FactorizedTopK

Issue - State: open - Opened by datasciyj over 1 year ago - 3 comments

#687 - how to return NDCG as a metric

Issue - State: open - Opened by Lulu20220 over 1 year ago - 2 comments

#686 - Initial effort to support EmbeddingFeature V2.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago

#685 - invalid results on during evaluation

Issue - State: open - Opened by riyaj8888 over 1 year ago - 3 comments

#684 - installation bug

Issue - State: open - Opened by riyaj8888 over 1 year ago

#683 - Adding the training script for upsell model.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago - 1 comment

#682 - Cross-reference the Clippy paper and the open-sourced implementation in TFRS.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago

#681 - Fix a typo in the doc.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago - 1 comment

#680 - [Question] Brute Force Query with Inclusions?

Issue - State: closed - Opened by ericy51 over 1 year ago - 1 comment

#679 - Not possible to run the basic retrieval example

Issue - State: open - Opened by pksinghus over 1 year ago - 8 comments

#678 - Spelling and grammar corrections

Pull Request - State: closed - Opened by Nik-code over 1 year ago - 1 comment

#677 - Spell and grammar corrections

Pull Request - State: closed - Opened by Nik-code over 1 year ago - 2 comments

#676 - Training took too long time with pre-trained embedding BERT layer

Issue - State: open - Opened by beubeu13220 over 1 year ago - 1 comment

#675 - [Idea] Learn from Positives and Negatives Using Two Retrieval Models

Issue - State: open - Opened by rlcauvin over 1 year ago - 3 comments

#674 - Can you share the code of clippyAdam based on pytorch?

Issue - State: open - Opened by sugarcandy over 1 year ago

#673 - Tutorial "Recommending movies: retrieval" is broken due to protobuf version conflict

Issue - State: open - Opened by marco-ve over 1 year ago

#672 - changed vocab_size to vocabulary_size

Pull Request - State: closed - Opened by AmgadHasan over 1 year ago - 2 comments

#671 - Please release scann compiled against Tensorflow 2.12

Issue - State: closed - Opened by IzakMaraisTAL over 1 year ago - 2 comments

#670 - Include sample_weight in batch metric update

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago - 1 comment

#669 - Unable to save model checkpoint on Tensorflow Recommender's subclass model

Issue - State: open - Opened by karndeepsingh over 1 year ago - 3 comments

#668 - [Question] Is it possible to get batch predictions from tfrs.layers.factorized_top_k.BruteForce

Issue - State: closed - Opened by ericy51 over 1 year ago - 1 comment

#667 - [Question] Is it Possible to change the SignatureDef for a BruteForce Model

Issue - State: closed - Opened by ericy51 over 1 year ago - 1 comment

#666 - Creating a frozen graph from TF Recommender Model to use model on server.

Issue - State: open - Opened by nikhil-krishna over 1 year ago

#665 - DELETED

Issue - State: closed - Opened by patrickorlando over 1 year ago

#664 - Silence some pytype errors.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago

#663 - Enable TFRS models to handle non-scalar regularization losses by using `tf.reduce_sum` instead of `sum`.

Pull Request - State: closed - Opened by copybara-service[bot] over 1 year ago - 1 comment

#662 - Enable tfrs models to handle non-scalar regularization losses by applying reduce_sum instead of sum on regularization losses.

Pull Request - State: open - Opened by copybara-service[bot] almost 2 years ago - 1 comment

GitHub / tensorflow/recommenders issues and pull requests