-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathproblem2_voxel.sh
executable file
·46 lines (44 loc) · 2.89 KB
/
problem2_voxel.sh
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
35
36
37
38
39
40
41
42
43
44
45
46
#!/bin/bash
command="$1" # The command to run
if [ "$command" = "train_baseline" ]; then
# Baseline Model (Linear Layers only)
echo "Training Baseline Model (Linear Layers only)"
python train_model.py --type 'vox' --max_iter 2000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 4e-4 --model_type 'baseline'
wait
python train_model.py --type 'vox' --max_iter 4000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 4e-5 --load_checkpoint --model_type 'baseline'
wait
python train_model.py --type 'vox' --max_iter 6000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 1e-6 --load_checkpoint --model_type 'baseline'
wait
python train_model.py --type 'vox' --max_iter 8000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 1e-6 --load_checkpoint --model_type 'baseline'
wait
python train_model.py --type 'vox' --max_iter 10002 --num_workers 4 --save_freq 200 --batch_size 30 --lr 5e-7 --load_checkpoint --model_type 'baseline'
wait
elif [ "$command" = "train_conv" ]; then
# Baseline Model (Conv layers Conv Model) Avg F1@0.05: 82.663
echo "Training Baseline Model (Conv layers Conv Model)"
python train_model.py --type 'vox' --max_iter 2000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 4e-4 --model_type 'conv'
wait
python train_model.py --type 'vox' --max_iter 4000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 4e-5 --load_checkpoint --model_type 'conv'
wait
python train_model.py --type 'vox' --max_iter 6000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 1e-6 --load_checkpoint --model_type 'conv'
wait
python train_model.py --type 'vox' --max_iter 8000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 1e-6 --load_checkpoint --model_type 'conv'
wait
python train_model.py --type 'vox' --max_iter 10002 --num_workers 4 --save_freq 200 --batch_size 30 --lr 5e-7 --load_checkpoint --model_type 'conv'
wait
elif [ "$command" = "train_mlp" ]; then
# Baseline Model (ImplicitMLPDecoder) Avg F1@0.05: 0.0
echo "Training Model (ImplicitMLPDecoder)"
python train_model.py --type 'vox' --max_iter 2000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 4e-4 --model_type 'mlp'
wait
python train_model.py --type 'vox' --max_iter 4000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 4e-5 --load_checkpoint --model_type 'mlp'
wait
python train_model.py --type 'vox' --max_iter 6000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 1e-6 --load_checkpoint --model_type 'mlp'
wait
python train_model.py --type 'vox' --max_iter 8000 --num_workers 4 --save_freq 200 --batch_size 16 --lr 1e-6 --load_checkpoint --model_type 'mlp'
wait
python train_model.py --type 'vox' --max_iter 10002 --num_workers 4 --save_freq 200 --batch_size 24 --lr 5e-7 --load_checkpoint --model_type 'mlp'
wait
else
echo "Usage: $0 [train_baseline|train_conv|train_implicitMLP]"
fi