issues
search
alan-turing-institute
/
ARC-LoCoMoSeT
Low-Cost Model Selection for Transformers
MIT License
1
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
ImageNet validation accuracy
#104
jack89roberts
opened
8 months ago
0
Report
#103
jack89roberts
opened
8 months ago
1
Models + dataset in -> recommendation out
#102
jack89roberts
opened
8 months ago
0
Renggli implementation does not stratify by label when splitting data
#101
jack89roberts
closed
8 months ago
0
Reinstate running metrics experiments with subsets of the same training set
#100
jack89roberts
closed
8 months ago
0
Proposed models and wip configs/stats for next runs
#99
jack89roberts
closed
9 months ago
1
80 dataset drops (also 82, 83, 90)
#98
philswatton
closed
9 months ago
2
91 binary datasets
#97
eddableheath
closed
8 months ago
6
Experiments to run
#96
jack89roberts
closed
8 months ago
0
Further caching configuration
#95
jack89roberts
closed
9 months ago
1
`Intel/vit-base-patch16-224-int8-static` producing poor results
#94
jack89roberts
closed
10 months ago
0
Setting default temporary directory for `tempfile`
#93
jack89roberts
closed
10 months ago
0
[DRAFT] 60 analysis scripts
#92
eddableheath
closed
6 months ago
1
Binary datasets
#91
eddableheath
closed
8 months ago
4
`wandb_args` on lower-level configs but `wandb` on top level configs
#90
jack89roberts
closed
9 months ago
0
Training and metrics in different layers
#89
jack89roberts
opened
10 months ago
1
RVL CDIP
#88
jack89roberts
closed
8 months ago
1
Metrics analysis
#87
jack89roberts
closed
10 months ago
0
Time analysis
#86
jack89roberts
opened
10 months ago
2
Binary classification / label subsetting
#85
jack89roberts
closed
10 months ago
0
Run with multiple random seeds
#84
jack89roberts
opened
10 months ago
1
Dataset train/val/test splits
#83
jack89roberts
closed
9 months ago
0
Refactor metric experiment for config objects
#82
eddableheath
closed
9 months ago
1
Less permissive configs
#81
philswatton
opened
10 months ago
0
Dataset Manipulation
#80
philswatton
closed
9 months ago
0
Multi-GPU jobs can hang with kernel version error
#79
jack89roberts
closed
8 months ago
1
Update readme
#78
philswatton
closed
10 months ago
0
wandb training time does not include initial setup & data preprocessing
#77
jack89roberts
closed
8 months ago
1
Configurable fields and preprocessing cache
#76
jack89roberts
closed
10 months ago
1
Get relevant results files for analysis from wandb
#75
jack89roberts
opened
10 months ago
0
`facebook/dinov2-base-imagenet1k-1-layer` is training poorly
#74
jack89roberts
closed
8 months ago
2
Training fails with distilled models
#73
jack89roberts
closed
8 months ago
0
Inference with GPU
#72
eddableheath
closed
10 months ago
1
Pass `device` from config to inference
#71
eddableheath
closed
10 months ago
0
Update `README` for workpacke 3A
#70
eddableheath
closed
10 months ago
1
`transformers` bug with `google/vit-hybrid-base-bit-384`
#69
jack89roberts
opened
10 months ago
0
Further Cache Management
#68
eddableheath
closed
8 months ago
6
Track computational requirements for metrics
#67
jack89roberts
closed
8 months ago
1
Baskerville setup and config generation
#66
eddableheath
closed
10 months ago
5
[DRAFT] 63 config generating script
#65
eddableheath
closed
11 months ago
2
[DRAFT] 62 log metric results to weights and biases
#64
eddableheath
closed
11 months ago
1
Config generating script
#63
eddableheath
closed
10 months ago
3
Log metric results to weights and biases
#62
eddableheath
closed
10 months ago
2
Fine-tuning
#61
jack89roberts
closed
11 months ago
2
Results Analysis Scripts
#60
eddableheath
closed
6 months ago
3
44 common interface for all metrics
#59
eddableheath
closed
11 months ago
2
9 metric for metrics decision
#58
eddableheath
closed
1 year ago
0
49 find further metrics to consider
#57
eddableheath
closed
1 year ago
6
Investigate LEEP
#56
eddableheath
closed
8 months ago
1
H_alpha-score metric
#55
eddableheath
closed
8 months ago
1
Next