issues
search
neuroailab
/
tfutils
Utilities for working with tensorflow
MIT License
25
stars
8
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
[TPU] Python 3 compatibility + `metric_fn` eval code rewrite
#162
eshedmargalit
opened
3 years ago
0
Workaround for partial function as online_agg_func
#161
cffan
opened
4 years ago
0
Allow callable to be an object
#160
cffan
opened
4 years ago
0
Adding seed data provider.
#159
anayebi
closed
4 years ago
0
Seed imnet data provider
#158
jvrsgsty
closed
4 years ago
1
Merge master into multiopt
#157
anayebi
closed
4 years ago
0
Loss aggfunc fx
#156
chengxuz
closed
4 years ago
1
Merge pull request #150 from neuroailab/inception_preproc
#155
anayebi
closed
4 years ago
0
adding latest multiopt changes
#154
anayebi
closed
4 years ago
0
Fixes repeated updates to the global step
#153
jvrsgsty
closed
4 years ago
1
Tpu pod
#152
anayebi
closed
4 years ago
0
Adds functionality to train on TPU pods or slices
#151
jvrsgsty
closed
4 years ago
0
Added inception preproc
#150
anayebi
closed
4 years ago
0
Added inception preproc
#149
anayebi
closed
4 years ago
0
Extending TPU optimizer to handle multiple optimizers
#148
anayebi
closed
4 years ago
0
MB fix to ensure consistency
#147
anayebi
closed
4 years ago
0
Multiple optimizer functionality
#146
anayebi
closed
4 years ago
2
Added bn trainable option
#145
anayebi
closed
5 years ago
0
Load param dict
#144
anayebi
closed
5 years ago
0
Updated tests to reflect that load_param_dict keeps all variables but…
#143
anayebi
closed
5 years ago
0
Having load params dict act as a modifier rather than complete replac…
#142
anayebi
closed
5 years ago
0
Added option to restore local vars during validation
#141
anayebi
closed
5 years ago
0
Initialize local variables in validation
#140
anayebi
opened
5 years ago
0
Option to remove global step
#139
anayebi
closed
5 years ago
0
Limits numpy version requirement
#138
jvrsgsty
closed
5 years ago
2
Python3.6 sonify will have problem with github check
#137
chengxuz
opened
5 years ago
0
pulling master in
#136
anayebi
closed
5 years ago
0
TPU Batch Norm which aggregated across shards
#135
anayebi
closed
5 years ago
0
Sets a version limit on tensorflow
#134
jvrsgsty
closed
5 years ago
0
Uses a single parameter to toggle between using tpu or not
#133
jvrsgsty
closed
5 years ago
2
Merging master updates to tpu branch
#132
anayebi
closed
5 years ago
0
New tpu master
#131
anayebi
closed
5 years ago
0
Revert "Revert "New tpu master""
#130
danielbear
closed
5 years ago
0
Revert "New tpu master"
#129
danielbear
closed
5 years ago
0
Updating with master
#128
anayebi
closed
5 years ago
0
E2e enqueue many
#127
anayebi
closed
5 years ago
0
Update tpu master
#126
anayebi
closed
5 years ago
0
Added cross gpu batch normalization
#125
anayebi
closed
5 years ago
0
When restoring a model, logs variables in checkpoint not in graph
#124
jvrsgsty
closed
5 years ago
5
Enqueue many python3
#123
anayebi
closed
5 years ago
0
Compatibility fix for DBInterface on Python 3
#122
jvrsgsty
closed
5 years ago
6
better way to handle name filtering for multi-GPU case (e.g. weight decay)
#121
yamins81
opened
5 years ago
0
cross-node training
#120
yamins81
opened
5 years ago
0
Loss computation interface
#119
yamins81
opened
5 years ago
0
New tpu master
#118
eshedmargalit
closed
5 years ago
2
Time sep
#117
anayebi
closed
5 years ago
0
Integrate check = tf.add_check_numerics_ops() which automatically checks for nan values
#116
damro
closed
5 years ago
3
Allow the specification of the gradient clipping value in the ClipOptimizer, and add the option of clipping by norm instead of by value
#115
damro
closed
5 years ago
2
python 3 compatibility
#114
yamins81
closed
5 years ago
0
Remove inactive branches
#113
chengxuz
closed
5 years ago
0
Next