somedaywilldo / bm-nas Goto Github PK
View Code? Open in Web Editor NEWBM-NAS: Bilevel Multimodal Neural Architecture Search (AAAI 2022 Oral)
BM-NAS: Bilevel Multimodal Neural Architecture Search (AAAI 2022 Oral)
A
May I ask when will the code and datasets be released?
Dear BM-NAS authors,
In the code the seed is fixed with lines:
np.random.seed(args.seed)
cudnn.benchmark = True
torch.manual_seed(args.seed)
cudnn.enabled=True
torch.cuda.manual_seed(args.seed)
However, every time I run the code, I get different results. Particularly with MM-IMDB, with the same hyper-parameters and the same seed, I got 62.43%, 62.55% and 62.36% Weighted F-1 scores. I believe, that fixing the score must make sure that we will get the exact same results on each run, can you please let me know where else I shall fix the seed? Furthermore, in the the weighted F1 score on MM-IMDB is reported as 62.92% +- 0.03, can you please explain whether this range is because of the different results after fixing the seed?
Thanks in advance,
Best regards,
Grigor
Thanks for providing code and insightful research in this area. However, when I try to execute mmimdb, I meet with several problems.
I meet the problems when I'm trying to execute: " python main_darts_searchable_mmimdb.py " as shown in the experiment section.
Firstly, on line 26 in main_darts_searchable_mmimdb.py, the argument should be "--parallel"? Or it will report a mistake :
And after changing this, I find an error of matrix shape:
I used the provided "python datasets/prepare_mmimdb.py" command to process the multimodal_imdb.hdf5 file.
This problem is solved by changing matrix size to 768.
However, when I'm trying to execute main_darts_found_ntu.py, I got the error mentioned in closed issue on ego and I'm not able to solve it:
Can you help fix this?
Thanks a lot for your kind help!
Hi,
I am interested in using your code for depth completion using two modalities as the ntu dataset. In the readme it mentions it can be done by using main_darts_searchable_ntu.py but I did not see it anywhere in the repository. Will it be published soon? Thank you
Hi:)
First of all, I thank you so much for the codes and the insightful research. Im doing a study on time series using Neural Architecture Search and your repository is helping greatly .
I have some questions regarding phases and status in your code though:(
Firstly,
Where it says, (in BM-NAS/models/search/train_searchable/ego.py)
phases = []
if status == 'search':
phases = ['train', 'dev']
else:
phases = ['train', 'test']
# model.load_state_dict(best_test_model_sd)
-- shouldnt the phases be just ['test'] for 'else'? (=status being 'Eval')
I couldnt understand why 'train' phase was included.:(
Another similar point that confused me was that in the scripts training the found architectures,
(For example, in the arguments of https://github.com/Somedaywilldo/BM-NAS/blob/master/main_darts_found_mmimd.py)
# loading searching experiment, if not None, perform evalution
parser.add_argument('--search_exp_dir', type=str, help='evaluate which search exp', default=None)
# loading evaluation experiment, if not None, perform test
parser.add_argument('--eval_exp_dir', type=str, help='test which eval exp', default=None)
What do 'perform evaluation' and 'perform test' mean here, respectively?
'Evaluation' seemed to actually mean 'train' (thay is, training the found architecture) if I look at the codes, so i was confused again with the use of the words.:(
Thank you again for the great model, and Id be very grateful for your answer.:)
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.