tensorflow/recommenders-addons

Issues

• Demo(when indicate the device in SquashedEmbedding) not working with Apple M1: Optimizer type is not supported! got <class 'keras.src.optimizers.adam.Adam'>

  #482 opened 20 hours ago by Ross-Fan
  0

• Demo not working on Apple M1, 'NoneType' object has no attribute 'tfra_hkv_hash_table_of_tensors'

  #481 opened 20 hours ago by Ross-Fan
  1

• ShadowVariable Post applying gradients gives NaN value for embedding

  #429 opened 5 months ago by alykhantejani
  1

• Slack channel / Discord for discussion

  #430 opened 5 months ago by alykhantejani
  2

• item embedding and user sequence item embedding can not use the same embedding variable

  #450 opened 4 months ago by uzhy1987
  1

• How can I remove Horovod ops from the savedModel to use with TF Serving?

  #467 opened 2 months ago by alykhantejani
  2

• fail to run demo:movielens-1m-keras-with-horovod

  #458 opened 10 days ago by W-O-W
  3

• Does HvdAllToAllEmbedding shard embeddings across ranks? Or is it copied on all ranks?

  #435 opened 5 months ago by alykhantejani
  2

• keras-horovod运行报错

  #464 opened 2 months ago by lixiang-repo
  1

• dynamic_embedding table 插入数据报错

  #463 opened 2 months ago by lwmonster
  2

• _math_ops.so: undefined symbol

  #406 opened 2 months ago by hauturier
  3

• Error building tf serving with TFRA ops

  #453 opened 3 months ago by alykhantejani
  2

• bazel version Conflict when build tensorflow/servering with recommender-addons

  #452 opened 3 months ago by uzhy1987
  12

• How to save and restore a model with dynamic embeddings?

  #451 opened 4 months ago by alykhantejani
  2

• TFRA on Dtensor

  #432 opened 5 months ago by jq
  0

• Support for Python 3.10?

  #369 opened 5 months ago by philipkd
  2

• Variable partitioning for embedding when using in tf.distribute.ParameterServerStrategy

  #417 opened 5 months ago by alykhantejani
  3

• New Version?

  #414 opened 5 months ago by alykhantejani
  5

• The sample demo have some bug.

  #401 opened 7 months ago by sosixyz
  18

• `freq_var` is not saved in `save_weights()`, which causes `restrict()` does not work on loaded models

  #385 opened 6 months ago by huangenyan
  3

• 栈溢出风险

  #405 opened 6 months ago by dropdust
  2

• inference mode issue

  #390 opened 6 months ago by beijinggao
  2

• WarmStartHook bug: AttributeError: 'Tensor' object has no attribute '_resource_handle'

  #330 opened a year ago by lixiang-repo
  3

• return_trainable currently is not implemented when using tf.function

  #336 opened a year ago by andrew-zzz
  9

• "Variable is not running on full-size initialization" warning when using AdagradOptimizer

  #335 opened 7 months ago by dzfish
  3

• ROCM toolchain

  #274 opened 7 months ago by fsx950223
  2

• support ragged tensor or input with mask for SquashedEmbedding

  #380 opened 7 months ago by lu-sitong
  2

• Performance Issue while using shadow ops embedding look up

  #325 opened 9 months ago by sunshinenum
  3

• ERROR IN EXAMPLE: movielens-1m-keras-with-horovod

  #370 opened a year ago by VikashPeddakota999
  1

• ERROR: tensorflow_recommenders_addons/dynamic_embedding/core/kernels/cuckoo_hashtable_op_gpu.cu.cc:413: CUDA error 2: out of memory

  #371 opened a year ago by VikashPeddakota999
  15

• Support `MultiWorkerMirroredStrategy` distributed training strategy for dynamic embeddings

  #365 opened a year ago by sivukhin
  10

• keras Batch Norm Issure

  #326 opened a year ago by sunshinenum
  2

• When I try to freeze the checkpoint model, an error occurs: you must feed a value for placeholder tensor

  #367 opened a year ago by yingzhic
  7

• GPU and CPU inference are inconsistent

  #324 opened a year ago by UnknownSwordsman
  1

• TFRA load embedding weights with wrong value when use CPU table

  #363 opened a year ago by fuhailin
  1

• predict has different result even though input is same

  #353 opened a year ago by xiedeacc
  3

• online infer load model failed for model trained use gpu

  #349 opened a year ago by xiedeacc
  0

• does recommenders-addons support bare tf c++ api inference?

  #344 opened a year ago by xiedeacc
  2

• so what's the usage of trainable_wrap in dynamic_embedding.embedding_lookup?

  #334 opened a year ago by andrew-zzz
  0

• the issure of SquashedEmbedding output

  #332 opened a year ago by andrew-zzz
  3

• SavedModel load for tags { serve }; Status: fail: Not found: Op type not registered 'TFRA>CuckooHashTableOfTensors' in binary running on machine....

  #295 opened 2 years ago by breaddance
  4

• Can some documentation be provided on how to build the TensorFlow C++ library from source to use TFRA custom ops in the TensorFlow C++ API?

  #296 opened 2 years ago by WangShengguang
  6

• Can't run with ParameterServerStrategy

  #272 opened 2 years ago by songyaheng
  5

• Compilation failed (environment exception) with branch master_tf1 and TF1.15.2

  #271 opened 2 years ago by shijiexu09
  1

• Does the dynamic emb support freezing of Tensorflow models？

  #321 opened 2 years ago by UnknownSwordsman
  1

• docker.io/tensorflow/build:latest-python: not found

  #323 opened 2 years ago by fuhailin
  0

• Why prefer cuckoo hash?

  #282 opened 2 years ago by wuminbin
  1

• [Bug] Did not load optimizer variable when use keras api.

  #291 opened 2 years ago by MoFHeka
  1

• `de.shadow_ops.embedding_lookup()` is non thread-safe

  #278 opened 2 years ago by alionkun
  1

• Can you show in detail how to set up the development environment?

  #273 opened 2 years ago by GH1995
  6