-
Notifications
You must be signed in to change notification settings - Fork 2
/
Readme.txt
35 lines (26 loc) · 1.3 KB
/
Readme.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Appendix is provided as appendix.pdf
Dataset structure
place the datasets (MNIST, CelebA and Cifar10 datasets). The datasets structure for the datasets is as follows
MNIST
- train
- test
Cifar10
- train
- test
celeba
- train
- test
celeba has just 1 folder in both train and test consisting of all the task imags
MNIST and Cifar10 should have 10 folders (numbered 0 - 9 corresponding to the 10 classes) in both train and test.
#Running evidential models
enp_run.py is used to run ENP-A and ENP-C
For eg. to run 50 shot image completion experiment with mnist for ENP-C, use
python3 enp_run.py --dataset "mnist" -use_det "true" -use_lat "false" --max_context_points 50 --model_type "CNP"
enp_l_run.py is used to run ENP-L models
For eg. to run 50 shot image completion experiment with mnist for ENP-C, use
python3 enp_l_run.py --dataset "mnist" -use_det "false" -use_lat "true" --max_context_points 50 --model_type "CNP"
#Running baseline models (NP, CNP, ANP)
np_baseline.py script can be used to run the baseline models
For eg. to run 50 shot image completion experiment with mnist for CNP, use
python3 np_baseline.py --dataset "mnist" -use_det "true" -use_lat "false" --max_context_points 50 --model_type "CNP"
Additionally, various hyperparameters and settings can be specified using utilFiles/get_args.py file