Safran/LION
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

add other test data

Browse source
This commit is contained in:
xzeng 2023-03-16 12:19:31 -04:00
parent 8cd91e1c20
commit 8023fdbe79
3 changed files with 15 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
README.md
View file

@ -65,16 +65,25 @@ run `python demo.py`, will load the released text2shape model on hugging face an
* run `bash ./script/train_prior.sh $NGPU` (the released checkpoint is trained with `NGPU=8` with 2 node on V100) * run `bash ./script/train_prior.sh $NGPU` (the released checkpoint is trained with `NGPU=8` with 2 node on V100)
### evaluate a trained prior ### evaluate a trained prior
* download the test data from [here](https://drive.google.com/file/d/1uEp0o6UpRqfYwvRXQGZ5ZgT1IYBQvUSV/view?usp=share_link), unzip and put it as `./datasets/test_data/` * download the test data (Table 1) from [here](https://drive.google.com/file/d/1uEp0o6UpRqfYwvRXQGZ5ZgT1IYBQvUSV/view?usp=share_link), unzip and put it as `./datasets/test_data/`
* download the released checkpoint from above * download the released checkpoint from above
``` ```
checkpoint="./lion_ckpt/unconditional/airplane/checkpoints/model.pt" checkpoint="./lion_ckpt/unconditional/airplane/checkpoints/model.pt"
bash ./script/eval.sh $checkpoint # will take 1-2 hour bash ./script/eval.sh $checkpoint # will take 1-2 hour
``` ```
#### other test data
* ShapeNet-Vol test data:
* please check [here](https://github.com/nv-tlabs/LION/issues/20#issuecomment-1436315100) before using this data
* [all category](https://drive.google.com/file/d/1QXrCbYKjTIAnH1OhZMathwdtQEXG5TjO/view?usp=sharing): 1000 shapes are sampled from the full validation set
* [chair, airplane, car](https://drive.google.com/file/d/11ZU_Bq5JwN3ggI7Ffj4NAjIxxhc2pNZ8/view?usp=share_link)
* table 21 and table 20, point-flow test data
* check [here](https://github.com/nv-tlabs/LION/issues/26#issuecomment-1466915318) before using this data
* [mug](https://drive.google.com/file/d/1lvJh2V94Nd7nZPcRqsCwW5oygsHOD3EE/view?usp=share_link) and [bottle](https://drive.google.com/file/d/1MRl4EgW6-4hOrdRq_e2iGh348a0aCH5f/view?usp=share_link)
* 55 catergory [data](https://drive.google.com/file/d/1Rbj1_33sN_S2YUbcJu6h922tKuJyQ2Dm/view?usp=share_link)
## Evaluate the samples with the 1-NNA metrics ## Evaluate the samples with the 1-NNA metrics
* download the test data from [here](https://drive.google.com/file/d/1uEp0o6UpRqfYwvRXQGZ5ZgT1IYBQvUSV/view?usp=share_link), unzip and put it as `./datasets/test_data/` * download the test data from [here](https://drive.google.com/file/d/1uEp0o6UpRqfYwvRXQGZ5ZgT1IYBQvUSV/view?usp=share_link), unzip and put it as `./datasets/test_data/`
* run `python ./script/compute_score.py` * run `python ./script/compute_score.py` (Note: for ShapeNet-Vol data and table 21, 20, need to set `norm_box=True`)
## Citation ## Citation
``` ```

6
script/compute_score.py
View file

@ -13,7 +13,7 @@ from utils.eval_helper import compute_score
samples = './lion_ckpt/unconditional/car/samples.pt' samples = './lion_ckpt/unconditional/car/samples.pt'
ref = './datasets/test_data/ref_val_car.pt' ref = './datasets/test_data/ref_val_car.pt'
compute_score(samples, ref_name=ref) compute_score(samples, ref_name=ref, norm_box=False)
""" """
will get: will get:
[Test] MinMatDis | CD 0.000913 | EMD 0.007523 [Test] MinMatDis | CD 0.000913 | EMD 0.007523
@ -24,7 +24,7 @@ will get:
samples = './lion_ckpt/unconditional/chair/samples.pt' samples = './lion_ckpt/unconditional/chair/samples.pt'
ref = './datasets/test_data/ref_val_chair.pt' ref = './datasets/test_data/ref_val_chair.pt'
compute_score(samples, ref_name=ref) compute_score(samples, ref_name=ref, norm_box=False)
""" """
[Test] MinMatDis | CD 0.002643 | EMD 0.015516 [Test] MinMatDis | CD 0.002643 | EMD 0.015516
[Test] Coverage | CD 0.489426 | EMD 0.521148 [Test] Coverage | CD 0.489426 | EMD 0.521148
@ -34,7 +34,7 @@ compute_score(samples, ref_name=ref)
samples = './lion_ckpt/unconditional/chair/samples.pt' samples = './lion_ckpt/unconditional/chair/samples.pt'
ref = './datasets/test_data/ref_val_chair.pt' ref = './datasets/test_data/ref_val_chair.pt'
compute_score(samples, ref_name=ref) compute_score(samples, ref_name=ref, norm_box=False)
""" """
[Test] MinMatDis | CD 0.000221 | EMD 0.003706 [Test] MinMatDis | CD 0.000221 | EMD 0.003706
[Test] Coverage | CD 0.471605 | EMD 0.496296 [Test] Coverage | CD 0.471605 | EMD 0.496296

2
script/train_prior.sh
View file

@ -32,7 +32,7 @@ $ENT \
sde.prior_model 'models.latent_points_ada_localprior.PVCNN2Prior' \ sde.prior_model 'models.latent_points_ada_localprior.PVCNN2Prior' \
sde.train_vae $train_vae \ sde.train_vae $train_vae \
sde.embedding_scale 1.0 \ sde.embedding_scale 1.0 \
viz.save_freq 1 \ viz.save_freq 1000 \
viz.viz_freq -200 viz.log_freq -1 viz.val_freq -10000 \ viz.viz_freq -200 viz.log_freq -1 viz.val_freq -10000 \
data.batch_size $BS \ data.batch_size $BS \
trainer.type 'trainers.train_2prior' \ trainer.type 'trainers.train_2prior' \