recommenders

recommenders copied to clipboard

Excluding previously seen items from test recommendations

10

In other libraries (e.g. Lightfm) it's common to have a facility to exclude previously seen items from test\eval reccomendations by passing in a set of train_interactions into our evaluation method...

nialloh23

Optimization for big values of num_list_per_user

1

I noticed that the current algorithm works slowly for big values of num_list_per_user, as a proposal, substitute a regular python list with a NumPy array for faster indexing. You can...

zufchan

I have a large retrieval model that uses the ScaNN method for indexing the recommendations. I'm saving the model out for the retrieval speed increase. I'm starting to run models...

kylemcmearty

The docs for `tfrs.tasks.Retrieval.num_hard_negatives` note If positive, the num_hard_negatives negative examples with largest logits are kept when computing cross-entropy loss. If larger than batch size **or non-positive**, all the negative...

dwyatte

TFRS Retrieval Model with softmax for converting score of each items into probabilities

1

We are using TFRS as the recommendation solution for our usecase. Instead of returning the score associated with each item we passed the scores into the softmax and returning probability...

drtinumohan

I was following the Movielens basic retrieval model tutorial: https://www.tensorflow.org/recommenders/examples/basic_retrieval and In our use case it's crucial that the user doesn't get recommended the same items that he interacted with...

Abdelrahman-7gab

Unable to use query_with_exclusions with undefined length of exclusions

1

Hi, I am trying to save a model with `query_with_exclusions` with an indefinite length of exclusions (because every user might have seen more or fewer posts....): ``` recommendations_k_21 = tf.function(lambda...

TPME

Approach for Content Recommendation Usecase

1

Hello folks, thanks for the great package. I'm looking to build a Content Recommendation Engine. The task is to recommend content to the sales representatives based on their role, skills,...

LaxmanSinghTomar

Why does the Retrieval class use an identify matrix for labels?

13

Hi there, Thanks so much for releasing and maintaining this code base. It's really fantastic. I don't really have a bug to report, but I do have a question regarding...

jiunsiew

[Discussion] Handle user ID with continuously training

1

I want to open a discussion here as I feel the tutorial is good for running a small dataset, a one-pass model. However, in the deployment, there are more processes...

xiaoyaoyang