New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Is NATS Extension of NAS_201 bench #48
Comments
Hi @Mars-204 , thanks for your interests. For the |
Hi @D-X-Y , thanks for the reply. It works for tss as you mentioned. But when I am trying to obtain the pre-trained weights for this index I am getting a an empty dic in return. Currently I am using the benchmark compressed pickle for tss for the same. Should I be using any other files for the pre-trained weights? d = pickle_load('/work/ws-tmp/g059997-naslib/g059997-naslib-1667607005/NASLib_mod/naslib/NATS-bench/Copy of NATS-tss-v1_0-3ffb9.pickle.pbz2') |
The network parameters are contained in full archive files (please follow the instructions at https://github.com/D-X-Y/NATS-Bench#preparation-and-download). Note that those files are pretty large. |
Thanks for the response. It is working. |
I was working with NAS_201 bench earlier and now am shifting to NATS bench. I have the following doubts:
Following is my implementation. I am using benchmark file with sss.
api = create(d, 'sss', fast_mode=False, verbose=True)
index = api.query_index_by_arch(convert_naslib_to_str(best_arch))
config = api.get_net_config(index, 'cifar10')
best_arch = get_cell_based_tiny_net(config)
logger.info("Queried results ({}): {}".format(metric, best_arch))
params = api.get_net_param(index, 'cifar10', None)
best_arch.load_state_dict(next(iter(params.values())))
The text was updated successfully, but these errors were encountered: