recommenders

recommenders copied to clipboard

Update tpu_embedding_layer.ipynb

1

Update the Download notebook link.

tiruk007

Multi-task Recommendations is a great tutorial! https://www.tensorflow.org/recommenders/examples/multitask I tried to override train_step to measure rating_loss and retrieval_loss respectively, as shown below. ```py class MovielensModel(tfrs.models.Model): ... def train_step(self, inputs): with tf.GradientTape()...

fuchami

Basic Recommender Ranking: Big Optimization issue

6

Hi everyone, I am a Data Scientist and I start following your examples on _Tensorflow recommenders_ few weeks ago. I noticed that the [Basic Recommender (ranking)](https://www.tensorflow.org/recommenders/examples/basic_ranking) is deeply flawed. Here...

steven-ja

Sample weights for both Retrieval and Ranking

3

I'd like to ask about using the `'sample_weight'` argument for both the Retrieval and the Ranking Tasks' call methods. [Retrieval docs](https://www.tensorflow.org/recommenders/api_docs/python/tfrs/tasks/Retrieval) [Ranking docs](https://www.tensorflow.org/recommenders/api_docs/python/tfrs/tasks/Ranking) 1. For both models, does it make...

hkristof03

Hi, The current value of 1e-6 is set [here](https://github.com/tensorflow/recommenders/commit/2f89ed85f0005d2368f0fae4a3028a81525bcffd). The rationale seems to be to avoid zeros but the value set could be too high for larger datasets and result...

aismlv

I am implementing Retrieval model on sequential query data. Each historical data contain several categorical variables, which converted to embeddings, concatenated and pushed to GRU to create a single query...

schistyakov

chore: fix variable miss in Multi-task recommenders

1

Bug: Multi-task recommenders embedding_dimension variable defined inner class but refer before definition. Fix: reclaim variable embedding_dimension as 32

BichengWang

Metrics computed by FactorizedTopK seems very off when running on M1 Max. Tried it on the "Quickstart" movielens example with Adam(0.005) and received the following evaluation results: ``` {'factorized_top_k/top_5_categorical_accuracy': 0.007550000213086605,...

Tractorage

user model as an object instead of keras sequential

1

I try to make it short, basically if I have a model as ``` user_model = tf.keras.Sequential([ tf.keras.layers.IntegerLookup(vocabulary=unique_user_ids, mask_token=None), tf.keras.layers.Embedding(input_dim=len(unique_user_ids) + 1, output_dim=embedding_dimension) ]) content_model = tf.keras.Sequential([ tf.keras.layers.experimental.preprocessing.IntegerLookup(vocabulary=contents_df, mask_token=None), tf.keras.layers.Embedding(input_dim=len(contents_df)...

naarkhoo

[Question] How can I detect and prevent folding ?

2

Hi, Thanks for your great library and tutorials ! In a previous blog post from Google, a phenomenon known as folding was mentioned for recommender systems that only use positive...

jeremybastin1207