issues
search
DeepRec-AI
/
HybridBackend
A high-performance framework for training wide-and-deep recommender systems on heterogeneous cluster
Apache License 2.0
156
stars
30
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Support default values for filename dataset
#160
2sin18
opened
6 months ago
0
Op type not registered 'HbGetNcclId' in binary
#159
ZhuYuJin
opened
7 months ago
0
fix pyarrow error
#158
Nov11
opened
7 months ago
0
Support default value for fields not exist
#157
2sin18
closed
8 months ago
0
ParquetDataset support configuration with default value
#156
Markz2z
opened
9 months ago
0
Training is very slow
#155
dixingxing0
closed
1 year ago
4
Error in multi-card in a single machine mode
#154
dixingxing0
closed
1 year ago
0
Train got error died with <Signals.SIGSEGV: 11>
#153
dixingxing0
closed
1 year ago
3
hb.data.ParquetDataset in hb.estimator.train_and_evaluate will loss data
#152
karterotte
closed
1 year ago
1
No OpKernel was registered to support Op 'HbSparseSegmentMeanGrad1' used by node
#151
karterotte
opened
1 year ago
0
[DOC] Add user docs for data deduplication and so forth.
#150
francktcheng
closed
1 year ago
0
[DATA] Implement zero-copied string dtype and accelerate shuffle.
#149
francktcheng
closed
1 year ago
1
[DATA] Implement zero-copied string dtype and accelerate shuffle.
#148
francktcheng
closed
1 year ago
0
Exception occurs when call `batch` with ragged tensor
#147
DelightRun
closed
1 year ago
0
[DATA] Implement zero-copied string dtype and accelerate shuffle.
#146
francktcheng
closed
1 year ago
0
[DATA] Implement zero-copied string dtype and accelerate shuffle.
#145
francktcheng
closed
1 year ago
0
[DATA] Implement zero-copied string dtype and accelerate shuffle.
#144
francktcheng
closed
1 year ago
0
[DIST] Set data_sync_drop_remainder as true by default.
#143
francktcheng
closed
1 year ago
0
[DIST] Set data_sync_drop_remainder as true by default.
#142
francktcheng
closed
1 year ago
0
[CI] Upgrade DeepRec docker to 2302
#141
2sin18
closed
1 year ago
1
[DIST] Set data_sync_drop_remainder as true by default.
#140
francktcheng
closed
1 year ago
0
[CI] Add more build information
#139
2sin18
closed
1 year ago
1
Throughput is lower than TFRecords when there are many strings in Parquets file
#138
deepllz
opened
1 year ago
0
[DIST] Rewrite NanTensorHook for SyncReplicasDataset.
#137
francktcheng
closed
1 year ago
0
[CI] Refines logging and cibuild
#136
2sin18
closed
1 year ago
1
[DATA] Refine data deduplication and add example.
#135
francktcheng
closed
1 year ago
0
[DIST] Refines NCCL logging messages for debugging
#134
2sin18
closed
1 year ago
0
[DIST] Using empty data at the end of SyncReplicasDataset.
#133
francktcheng
closed
1 year ago
0
[CI] Upgrade to 1.0.0
#132
2sin18
closed
1 year ago
1
[EMB] Add experimental support for embedding service acceleration
#131
2sin18
closed
1 year ago
0
[DIST] Implement a hierarchical embedding lookup.
#130
francktcheng
closed
1 year ago
0
[DATA] Support transfer optimization
#129
2sin18
closed
1 year ago
0
[DATA] Support ORC format and data deduplication
#128
2sin18
closed
1 year ago
0
[DIST] Use variable_scope with partitioner for sharded deeprecev.
#127
francktcheng
closed
1 year ago
1
[DIST] Refine naming of deeprecev variables in unsharded cases.
#126
francktcheng
closed
1 year ago
0
Deeprec hangs in distributed mode.
#125
silingtong123
opened
1 year ago
0
[DIST] Support a standalone evaluation in `hb.keras`.
#124
francktcheng
closed
1 year ago
0
[DIST] Fix: Shared embedding shape rewriting.
#123
francktcheng
closed
1 year ago
0
Failed to train with multiple GPUs in single node
#122
ZhuYuJin
closed
1 year ago
0
hb.data.ParquetDataset will discard some data
#121
silingtong123
closed
1 year ago
0
init_from_checkpoint throw Exception when using hb.keras.Model
#120
karterotte
closed
1 year ago
1
[CI] Upgrade DeepRec
#119
2sin18
closed
1 year ago
1
hb.keras.model evaluate error
#118
karterotte
closed
1 year ago
0
EmbeddingLookupRewritingForDeepRecEV Add "part0" to op-name twice
#117
karterotte
closed
1 year ago
0
[CI] Fix GitHub workflow files
#116
2sin18
closed
1 year ago
0
[DIST] Support standalone evaluation and prediction in `hb.Estimator`.
#115
francktcheng
closed
1 year ago
0
[DIST] Support pipeline-based semi-synchronous training.
#114
francktcheng
closed
1 year ago
0
[DIST] Refactor directories of embedding sharding
#113
2sin18
closed
1 year ago
1
[DATA] Refines data transfer prefetching and synchronization
#112
2sin18
closed
1 year ago
1
[DIST] Support Non-intrusive embedding APIs
#111
2sin18
closed
1 year ago
1
Next