issues
search
IBM
/
mi-prometheus
Enabling reproducible Machine Learning research
http://mi-prometheus.rtfd.io/
Apache License 2.0
42
stars
18
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Numpy > 1.11.0 causes errors due to numpy.float64 not being automatically converted to Int
#69
sesevgen
closed
5 years ago
1
'ThalnetModule' object has not attribute 'logger'
#68
sesevgen
opened
5 years ago
1
Store termination cause in model checkpoint
#67
tkornuta-ibm
closed
5 years ago
1
Rethink operation of grid-analyzer
#66
tkornuta-ibm
closed
5 years ago
7
Change --o to --e
#65
tkornuta-ibm
closed
5 years ago
1
Investigate the content of statistics, rethink the desired behavior of online/offline trainers
#64
tkornuta-ibm
closed
5 years ago
1
Add Ctrl+C handling to grid workers
#63
tkornuta-ibm
closed
5 years ago
0
Add basic trainer selection to config of grid_trainer_*
#62
tkornuta-ibm
closed
5 years ago
0
Move generate_feature_maps_file() from CLEVR to GenerateFeatureMaps
#61
tkornuta-ibm
opened
5 years ago
3
Push MAC and S-MAC to public github
#60
tkornuta-ibm
closed
5 years ago
1
Create grid configs enabling reproducing results from VIGIL paper
#59
tkornuta-ibm
closed
5 years ago
0
Refactor VideoTextToClass problems
#58
tkornuta-ibm
closed
5 years ago
0
Trying to fix LGTM warnings
#57
vmarois
closed
5 years ago
3
[Feature] Linking the documentation to other documentations
#56
vmarois
closed
5 years ago
0
[Fix] Documentation building
#55
vmarois
closed
5 years ago
1
Fix/workers epoch size
#54
tkornuta-ibm
closed
5 years ago
2
Vision problems and models cleanup
#53
tkornuta-ibm
closed
5 years ago
1
grid_*_cpu not working on MAC/OSX
#52
tkornuta-ibm
closed
5 years ago
2
Prepare config for grid_trainer with LeNet5 and SimpleCNN model trained on MNIST
#51
tkornuta-ibm
closed
5 years ago
0
Online and offline trainers aggregate different number samples for MNIST during training
#50
tkornuta-ibm
closed
5 years ago
3
Logging of dataset size before/after sampler initialization
#49
tkornuta-ibm
closed
5 years ago
1
Refactor and simplify MNIST and CIFAR10 problems
#48
tkornuta-ibm
closed
5 years ago
1
Release 0.3
#46
tkornuta-ibm
closed
5 years ago
9
renamed lgtm.yaml to lgtm.yml
#45
tkornuta-ibm
closed
5 years ago
0
Fix: Except block directly handles BaseException
#44
tkornuta-ibm
closed
5 years ago
1
Fixing 'old-style' classes in DNC
#43
tkornuta-ibm
closed
5 years ago
1
Feat/model loading
#42
tkornuta-ibm
closed
5 years ago
0
Doc build is failing
#41
vmarois
closed
5 years ago
1
Change the way --model is handled
#40
vmarois
closed
5 years ago
1
Reorganize helpers/workers
#39
tkornuta-ibm
opened
5 years ago
2
Simple merge with Master (badges and .lgtm.yaml)
#38
tkornuta-ibm
closed
5 years ago
0
Introduce mutex-based experiment configuration to Grid Workers GPU
#37
tkornuta-ibm
opened
5 years ago
0
Index Splitter
#36
tkornuta-ibm
closed
5 years ago
9
Add description about development mode to the documentation
#35
tkornuta-ibm
closed
5 years ago
1
Enhancement of the documentation
#34
vmarois
opened
5 years ago
1
[Feature] Enhancing the documentation
#33
vmarois
closed
5 years ago
4
Sampler factory
#32
tkornuta-ibm
closed
5 years ago
2
Implement Sampler Factory reusing the existing PyTorch samplers
#31
tkornuta-ibm
closed
5 years ago
1
Implement index_splitter worker
#30
tkornuta-ibm
closed
5 years ago
2
Investigate overhead of mi-prometheus
#29
vmarois
opened
5 years ago
0
Updated conf.py & setup.py to reflect version update.
#28
tkornuta-ibm
closed
5 years ago
0
Release/0.2.1
#27
vmarois
closed
5 years ago
0
Clear up torchvision version
#26
vmarois
closed
5 years ago
1
Clean up the flags of the different workers
#25
vmarois
opened
5 years ago
0
Fix/mnist cifar paths
#24
tkornuta-ibm
closed
5 years ago
0
Grid workers hotfix
#23
tkornuta-ibm
closed
5 years ago
0
Standardize the names of mip scripts
#22
tkornuta-ibm
closed
5 years ago
1
Standardize the way all worker scripts are reacting for lack of CUDA-compatible devices
#21
tkornuta-ibm
closed
5 years ago
1
Standardize max_concurrent_runs and experiment_repetitions for all grid trainers and testers
#20
tkornuta-ibm
closed
5 years ago
0
Grid workers should rely on mip-* scripts
#19
tkornuta-ibm
closed
5 years ago
1
Previous
Next