Update NATS-Bench (sss version 1.2)
This commit is contained in:
D-X-Y
@@ -11,7 +11,6 @@
|
||||
# python exps/NATS-Bench/sss-collect.py #
|
||||
##############################################################################
|
||||
import os, re, sys, time, shutil, argparse, collections
|
||||
import numpy as np
|
||||
import torch
|
||||
from tqdm import tqdm
|
||||
from pathlib import Path
|
||||
@@ -22,7 +21,7 @@ if str(lib_dir) not in sys.path: sys.path.insert(0, str(lib_dir))
|
||||
from log_utils import AverageMeter, time_string, convert_secs2time
|
||||
from config_utils import dict2config
|
||||
from models import CellStructure, get_cell_based_tiny_net
|
||||
from nas_201_api import ArchResults, ResultsCount
|
||||
from nats_bench import pickle_save, pickle_load, ArchResults, ResultsCount
|
||||
from procedures import bench_pure_evaluate as pure_evaluate, get_nas_bench_loaders
|
||||
from utils import get_md5_file
|
||||
|
||||
@@ -193,8 +192,8 @@ def simplify(save_dir, save_name, nets, total):
|
||||
arch_str = nets[index]
|
||||
hp2info = OrderedDict()
|
||||
|
||||
full_save_path = full_save_dir / '{:06d}.npy'.format(index)
|
||||
simple_save_path = simple_save_dir / '{:06d}.npy'.format(index)
|
||||
full_save_path = full_save_dir / '{:06d}.pickle'.format(index)
|
||||
simple_save_path = simple_save_dir / '{:06d}.pickle'.format(index)
|
||||
|
||||
for hp in hps:
|
||||
sub_save_dir = save_dir / 'raw-data-{:}'.format(hp)
|
||||
@@ -213,13 +212,13 @@ def simplify(save_dir, save_name, nets, total):
|
||||
to_save_data = OrderedDict({'01': hp2info['01'].state_dict(),
|
||||
'12': hp2info['12'].state_dict(),
|
||||
'90': hp2info['90'].state_dict()})
|
||||
np.save(str(full_save_path), to_save_data)
|
||||
pickle_save(to_save_data, str(full_save_path))
|
||||
|
||||
for hp in hps: hp2info[hp].clear_params()
|
||||
to_save_data = OrderedDict({'01': hp2info['01'].state_dict(),
|
||||
'12': hp2info['12'].state_dict(),
|
||||
'90': hp2info['90'].state_dict()})
|
||||
np.save(str(simple_save_path), to_save_data)
|
||||
pickle_save(to_save_data, str(simple_save_path))
|
||||
arch2infos[index] = to_save_data
|
||||
# measure elapsed time
|
||||
arch_time.update(time.time() - end_time)
|
||||
@@ -231,18 +230,23 @@ def simplify(save_dir, save_name, nets, total):
|
||||
'total_archs': total,
|
||||
'arch2infos' : arch2infos,
|
||||
'evaluated_indexes': evaluated_indexes}
|
||||
save_file_name = save_dir / '{:}.npy'.format(save_name)
|
||||
np.save(str(save_file_name), final_infos)
|
||||
save_file_name = save_dir / '{:}.pickle'.format(save_name)
|
||||
pickle_save(final_infos, str(save_file_name))
|
||||
# move the benchmark file to a new path
|
||||
hd5sum = get_md5_file(save_file_name)
|
||||
hd5_file_name = save_dir / '{:}-{:}.npy'.format(NATS_TSS_BASE_NAME, hd5sum)
|
||||
shutil.move(save_file_name, hd5_file_name)
|
||||
hd5sum = get_md5_file(str(save_file_name) + '.pbz2')
|
||||
hd5_file_name = save_dir / '{:}-{:}.pickle.pbz2'.format(NATS_TSS_BASE_NAME, hd5sum)
|
||||
shutil.move(str(save_file_name) + '.pbz2', hd5_file_name)
|
||||
print('Save {:} / {:} architecture results into {:} -> {:}.'.format(len(evaluated_indexes), total, save_file_name, hd5_file_name))
|
||||
# move the directory to a new path
|
||||
hd5_full_save_dir = save_dir / '{:}-{:}-full'.format(NATS_TSS_BASE_NAME, hd5sum)
|
||||
hd5_simple_save_dir = save_dir / '{:}-{:}-simple'.format(NATS_TSS_BASE_NAME, hd5sum)
|
||||
shutil.move(full_save_dir, hd5_full_save_dir)
|
||||
shutil.move(simple_save_dir, hd5_simple_save_dir)
|
||||
# save the meta information for simple and full
|
||||
final_infos['arch2infos'] = None
|
||||
final_infos['evaluated_indexes'] = set()
|
||||
pickle_save(final_infos, str(hd5_full_save_dir / 'meta.pickle'))
|
||||
pickle_save(final_infos, str(hd5_simple_save_dir / 'meta.pickle'))
|
||||
|
||||
|
||||
def traverse_net(candidates: List[int], N: int):
|
||||
|
||||
@@ -1,12 +1,10 @@
|
||||
###############################################################
|
||||
# NAS-Bench-201, ICLR 2020 (https://arxiv.org/abs/2001.00326) #
|
||||
###############################################################
|
||||
# NATS-Bench: Benchmarking NAS algorithms for Architecture Topology and Size
|
||||
###############################################################
|
||||
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2020.06 #
|
||||
###############################################################
|
||||
# Usage: python exps/NAS-Bench-201/test-nas-api.py #
|
||||
###############################################################
|
||||
##############################################################################
|
||||
# NATS-Bench: Benchmarking NAS algorithms for Architecture Topology and Size #
|
||||
##############################################################################
|
||||
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2020.08 #
|
||||
##############################################################################
|
||||
# Usage: python exps/NATS-Bench/test-nats-api.py #
|
||||
##############################################################################
|
||||
import os, sys, time, torch, argparse
|
||||
import numpy as np
|
||||
from typing import List, Text, Dict, Any
|
||||
@@ -61,10 +59,12 @@ def test_api(api, is_301=True):
|
||||
print('{:}\n'.format(info))
|
||||
info = api.get_latency(12, 'cifar10')
|
||||
print('{:}\n'.format(info))
|
||||
for index in [13, 15, 19, 200]:
|
||||
info = api.get_latency(index, 'cifar10')
|
||||
|
||||
# Count the number of architectures
|
||||
info = api.statistics('cifar100', '12')
|
||||
print('{:}\n'.format(info))
|
||||
print('{:} statistics results : {:}\n'.format(time_string(), info))
|
||||
|
||||
# Show the information of the 123-th architecture
|
||||
api.show(123)
|
||||
@@ -80,33 +80,18 @@ def test_api(api, is_301=True):
|
||||
print('Compute the adjacency matrix of {:}'.format(arch_str))
|
||||
print(matrix)
|
||||
info = api.simulate_train_eval(123, 'cifar10')
|
||||
print('simulate_train_eval : {:}'.format(info))
|
||||
|
||||
|
||||
def test_issue_81_82(api):
|
||||
results = api.query_by_index(0, 'cifar10-valid', hp='12')
|
||||
results = api.query_by_index(0, 'cifar10-valid', hp='200')
|
||||
print(list(results.keys()))
|
||||
print(results[888].get_eval('valid'))
|
||||
print(results[888].get_eval('x-valid'))
|
||||
result_dict = api.get_more_info(index=0, dataset='cifar10-valid', iepoch=11, hp='200', is_random=False)
|
||||
info = api.query_by_arch('|nor_conv_3x3~0|+|skip_connect~0|nor_conv_3x3~1|+|skip_connect~0|none~1|nor_conv_3x3~2|', '200')
|
||||
print(info)
|
||||
structure = CellStructure.str2structure('|nor_conv_3x3~0|+|skip_connect~0|nor_conv_3x3~1|+|skip_connect~0|none~1|nor_conv_3x3~2|')
|
||||
info = api.query_by_arch(structure, '200')
|
||||
print(info)
|
||||
print('simulate_train_eval : {:}\n\n'.format(info))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
|
||||
api201 = create(os.path.join(os.environ['TORCH_HOME'], 'NAS-Bench-201-v1_0-e61699.pth'), 'topology', True)
|
||||
test_issue_81_82(api201)
|
||||
print ('Test {:} done'.format(api201))
|
||||
for fast_mode in [True, False]:
|
||||
for verbose in [True, False]:
|
||||
print('{:} create with fast_mode={:} and verbose={:}'.format(time_string(), fast_mode, verbose))
|
||||
api301 = create(None, 'size', fast_mode=fast_mode, verbose=True)
|
||||
print('{:} --->>> {:}'.format(time_string(), api301))
|
||||
test_api(api301, True)
|
||||
|
||||
api201 = create(None, 'topology', True) # use the default file path
|
||||
test_issue_81_82(api201)
|
||||
test_api(api201, False)
|
||||
print ('Test {:} done'.format(api201))
|
||||
|
||||
api301 = create(None, 'size', True)
|
||||
test_api(api301, True)
|
||||
# api201 = create(None, 'topology', True) # use the default file path
|
||||
# test_api(api201, False)
|
||||
# print ('Test {:} done'.format(api201))
|
||||
@@ -5,8 +5,8 @@
|
||||
# required to install hpbandster ##################################
|
||||
# pip install hpbandster ##################################
|
||||
###################################################################
|
||||
# OMP_NUM_THREADS=4 python exps/algos-v2/bohb.py --search_space tss --dataset cifar10 --num_samples 4 --random_fraction 0.0 --bandwidth_factor 3 --rand_seed 1
|
||||
# OMP_NUM_THREADS=4 python exps/algos-v2/bohb.py --search_space sss --dataset cifar10 --num_samples 4 --random_fraction 0.0 --bandwidth_factor 3 --rand_seed 1
|
||||
# OMP_NUM_THREADS=4 python exps/NATS-algos/bohb.py --search_space tss --dataset cifar10 --num_samples 4 --random_fraction 0.0 --bandwidth_factor 3 --rand_seed 1
|
||||
# OMP_NUM_THREADS=4 python exps/NATS-algos/bohb.py --search_space sss --dataset cifar10 --num_samples 4 --random_fraction 0.0 --bandwidth_factor 3 --rand_seed 1
|
||||
###################################################################
|
||||
import os, sys, time, random, argparse, collections
|
||||
from copy import deepcopy
|
||||
@@ -167,7 +167,7 @@ if __name__ == '__main__':
|
||||
parser.add_argument('--rand_seed', type=int, default=-1, help='manual seed')
|
||||
args = parser.parse_args()
|
||||
|
||||
api = create(None, args.search_space, verbose=False)
|
||||
api = create(None, args.search_space, fast_mode=True, verbose=False)
|
||||
|
||||
args.save_dir = os.path.join('{:}-{:}'.format(args.save_dir, args.search_space), args.dataset, 'BOHB')
|
||||
print('save-dir : {:}'.format(args.save_dir))
|
||||
@@ -3,9 +3,9 @@
|
||||
##############################################################################
|
||||
# Random Search for Hyper-Parameter Optimization, JMLR 2012 ##################
|
||||
##############################################################################
|
||||
# python ./exps/algos-v2/random_wo_share.py --dataset cifar10 --search_space tss
|
||||
# python ./exps/algos-v2/random_wo_share.py --dataset cifar100 --search_space tss
|
||||
# python ./exps/algos-v2/random_wo_share.py --dataset ImageNet16-120 --search_space tss
|
||||
# python ./exps/NATS-algos/random_wo_share.py --dataset cifar10 --search_space tss
|
||||
# python ./exps/NATS-algos/random_wo_share.py --dataset cifar100 --search_space tss
|
||||
# python ./exps/NATS-algos/random_wo_share.py --dataset ImageNet16-120 --search_space tss
|
||||
##############################################################################
|
||||
import os, sys, time, glob, random, argparse
|
||||
import numpy as np, collections
|
||||
@@ -71,7 +71,7 @@ if __name__ == '__main__':
|
||||
parser.add_argument('--rand_seed', type=int, default=-1, help='manual seed')
|
||||
args = parser.parse_args()
|
||||
|
||||
api = create(None, args.search_space, verbose=False)
|
||||
api = create(None, args.search_space, fast_mode=True, verbose=False)
|
||||
|
||||
args.save_dir = os.path.join('{:}-{:}'.format(args.save_dir, args.search_space), args.dataset, 'RANDOM')
|
||||
print('save-dir : {:}'.format(args.save_dir))
|
||||
@@ -3,12 +3,12 @@
|
||||
##################################################################
|
||||
# Regularized Evolution for Image Classifier Architecture Search #
|
||||
##################################################################
|
||||
# python ./exps/algos-v2/regularized_ea.py --dataset cifar10 --search_space tss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/algos-v2/regularized_ea.py --dataset cifar100 --search_space tss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/algos-v2/regularized_ea.py --dataset ImageNet16-120 --search_space tss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/algos-v2/regularized_ea.py --dataset cifar10 --search_space sss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/algos-v2/regularized_ea.py --dataset cifar100 --search_space sss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/algos-v2/regularized_ea.py --dataset ImageNet16-120 --search_space sss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/NATS-algos/regularized_ea.py --dataset cifar10 --search_space tss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/NATS-algos/regularized_ea.py --dataset cifar100 --search_space tss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/NATS-algos/regularized_ea.py --dataset ImageNet16-120 --search_space tss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/NATS-algos/regularized_ea.py --dataset cifar10 --search_space sss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/NATS-algos/regularized_ea.py --dataset cifar100 --search_space sss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
# python ./exps/NATS-algos/regularized_ea.py --dataset ImageNet16-120 --search_space sss --ea_cycles 200 --ea_population 10 --ea_sample_size 3 --rand_seed 1
|
||||
##################################################################
|
||||
import os, sys, time, glob, random, argparse
|
||||
import numpy as np, collections
|
||||
@@ -198,7 +198,7 @@ if __name__ == '__main__':
|
||||
parser.add_argument('--rand_seed', type=int, default=-1, help='manual seed')
|
||||
args = parser.parse_args()
|
||||
|
||||
api = create(None, args.search_space, verbose=False)
|
||||
api = create(None, args.search_space, fast_mode=True, verbose=False)
|
||||
|
||||
args.save_dir = os.path.join('{:}-{:}'.format(args.save_dir, args.search_space), args.dataset, 'R-EA-SS{:}'.format(args.ea_sample_size))
|
||||
print('save-dir : {:}'.format(args.save_dir))
|
||||
@@ -3,12 +3,12 @@
|
||||
#####################################################################################################
|
||||
# modified from https://github.com/pytorch/examples/blob/master/reinforcement_learning/reinforce.py #
|
||||
#####################################################################################################
|
||||
# python ./exps/algos-v2/reinforce.py --dataset cifar10 --search_space tss --learning_rate 0.01
|
||||
# python ./exps/algos-v2/reinforce.py --dataset cifar100 --search_space tss --learning_rate 0.01
|
||||
# python ./exps/algos-v2/reinforce.py --dataset ImageNet16-120 --search_space tss --learning_rate 0.01
|
||||
# python ./exps/algos-v2/reinforce.py --dataset cifar10 --search_space sss --learning_rate 0.01
|
||||
# python ./exps/algos-v2/reinforce.py --dataset cifar100 --search_space sss --learning_rate 0.01
|
||||
# python ./exps/algos-v2/reinforce.py --dataset ImageNet16-120 --search_space sss --learning_rate 0.01
|
||||
# python ./exps/NATS-algos/reinforce.py --dataset cifar10 --search_space tss --learning_rate 0.01
|
||||
# python ./exps/NATS-algos/reinforce.py --dataset cifar100 --search_space tss --learning_rate 0.01
|
||||
# python ./exps/NATS-algos/reinforce.py --dataset ImageNet16-120 --search_space tss --learning_rate 0.01
|
||||
# python ./exps/NATS-algos/reinforce.py --dataset cifar10 --search_space sss --learning_rate 0.01
|
||||
# python ./exps/NATS-algos/reinforce.py --dataset cifar100 --search_space sss --learning_rate 0.01
|
||||
# python ./exps/NATS-algos/reinforce.py --dataset ImageNet16-120 --search_space sss --learning_rate 0.01
|
||||
#####################################################################################################
|
||||
import os, sys, time, glob, random, argparse
|
||||
import numpy as np, collections
|
||||
47
exps/NATS-algos/run-all.sh
Normal file
47
exps/NATS-algos/run-all.sh
Normal file
@@ -0,0 +1,47 @@
|
||||
#!/bin/bash
|
||||
# bash ./exps/NATS-algos/run-all.sh mul
|
||||
# bash ./exps/NATS-algos/run-all.sh ws
|
||||
set -e
|
||||
echo script name: $0
|
||||
echo $# arguments
|
||||
if [ "$#" -ne 1 ] ;then
|
||||
echo "Input illegal number of parameters " $#
|
||||
echo "Need 1 parameters for type of algorithms."
|
||||
exit 1
|
||||
fi
|
||||
|
||||
|
||||
datasets="cifar10 cifar100 ImageNet16-120"
|
||||
alg_type=$1
|
||||
|
||||
if [ "$alg_type" == "mul" ]; then
|
||||
search_spaces="tss sss"
|
||||
|
||||
for dataset in ${datasets}
|
||||
do
|
||||
for search_space in ${search_spaces}
|
||||
do
|
||||
python ./exps/NATS-algos/reinforce.py --dataset ${dataset} --search_space ${search_space} --learning_rate 0.01
|
||||
python ./exps/NATS-algos/regularized_ea.py --dataset ${dataset} --search_space ${search_space} --ea_cycles 200 --ea_population 10 --ea_sample_size 3
|
||||
python ./exps/NATS-algos/random_wo_share.py --dataset ${dataset} --search_space ${search_space}
|
||||
python ./exps/NATS-algos/bohb.py --dataset ${dataset} --search_space ${search_space} --num_samples 4 --random_fraction 0.0 --bandwidth_factor 3
|
||||
done
|
||||
done
|
||||
|
||||
python exps/experimental/vis-bench-algos.py --search_space tss
|
||||
python exps/experimental/vis-bench-algos.py --search_space sss
|
||||
else
|
||||
seeds="777 888 999"
|
||||
algos="darts-v1 darts-v2 gdas setn random enas"
|
||||
epoch=200
|
||||
for seed in ${seeds}
|
||||
do
|
||||
for alg in ${algos}
|
||||
do
|
||||
python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo ${alg} --rand_seed ${seed} --overwite_epochs ${epoch}
|
||||
python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo ${alg} --rand_seed ${seed} --overwite_epochs ${epoch}
|
||||
python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo ${alg} --rand_seed ${seed} --overwite_epochs ${epoch}
|
||||
done
|
||||
done
|
||||
fi
|
||||
|
||||
@@ -1,29 +1,29 @@
|
||||
##################################################
|
||||
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2020 #
|
||||
######################################################################################
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo darts-v1 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo darts-v1 --drop_path_rate 0.3
|
||||
# python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo darts-v1
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo darts-v1 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo darts-v1 --drop_path_rate 0.3
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo darts-v1
|
||||
####
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo darts-v2 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo darts-v2
|
||||
# python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo darts-v2
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo darts-v2 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo darts-v2
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo darts-v2
|
||||
####
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo gdas --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo gdas
|
||||
# python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo gdas
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo gdas --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo gdas
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo gdas
|
||||
####
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo setn --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo setn
|
||||
# python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo setn
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo setn --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo setn
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo setn
|
||||
####
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo random --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo random
|
||||
# python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo random
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo random --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo random
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo random
|
||||
####
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo enas --arch_weight_decay 0 --arch_learning_rate 0.001 --arch_eps 0.001 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo enas --arch_weight_decay 0 --arch_learning_rate 0.001 --arch_eps 0.001 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo enas --arch_weight_decay 0 --arch_learning_rate 0.001 --arch_eps 0.001 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo enas --arch_weight_decay 0 --arch_learning_rate 0.001 --arch_eps 0.001 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo enas --arch_weight_decay 0 --arch_learning_rate 0.001 --arch_eps 0.001 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo enas --arch_weight_decay 0 --arch_learning_rate 0.001 --arch_eps 0.001 --rand_seed 777
|
||||
######################################################################################
|
||||
import os, sys, time, random, argparse
|
||||
import numpy as np
|
||||
@@ -364,7 +364,7 @@ def main(xargs):
|
||||
logger.log('The parameters of the search model = {:.2f} MB'.format(params))
|
||||
logger.log('search-space : {:}'.format(search_space))
|
||||
if bool(xargs.use_api):
|
||||
api = create(None, 'topology', verbose=False)
|
||||
api = create(None, 'topology', fast_mode=True, verbose=False)
|
||||
else:
|
||||
api = None
|
||||
logger.log('{:} create API = {:} done'.format(time_string(), api))
|
||||
@@ -1,17 +1,17 @@
|
||||
##################################################
|
||||
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2020 #
|
||||
######################################################################################
|
||||
# python ./exps/algos-v2/search-size.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo tas --rand_seed 777
|
||||
# python ./exps/algos-v2/search-size.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo tas --rand_seed 777
|
||||
# python ./exps/algos-v2/search-size.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo tas --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo tas --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo tas --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo tas --rand_seed 777
|
||||
####
|
||||
# python ./exps/algos-v2/search-size.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo fbv2 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-size.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo fbv2 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-size.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo fbv2 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo fbv2 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo fbv2 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo fbv2 --rand_seed 777
|
||||
####
|
||||
# python ./exps/algos-v2/search-size.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo tunas --arch_weight_decay 0 --rand_seed 777 --use_api 0
|
||||
# python ./exps/algos-v2/search-size.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo tunas --arch_weight_decay 0 --rand_seed 777
|
||||
# python ./exps/algos-v2/search-size.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo tunas --arch_weight_decay 0 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo tunas --arch_weight_decay 0 --rand_seed 777 --use_api 0
|
||||
# python ./exps/NATS-algos/search-size.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo tunas --arch_weight_decay 0 --rand_seed 777
|
||||
# python ./exps/NATS-algos/search-size.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo tunas --arch_weight_decay 0 --rand_seed 777
|
||||
######################################################################################
|
||||
import os, sys, time, random, argparse
|
||||
import numpy as np
|
||||
@@ -176,7 +176,7 @@ def main(xargs):
|
||||
logger.log('The parameters of the search model = {:.2f} MB'.format(params))
|
||||
logger.log('search-space : {:}'.format(search_space))
|
||||
if bool(xargs.use_api):
|
||||
api = create(None, 'size', verbose=False)
|
||||
api = create(None, 'size', fast_mode=True, verbose=False)
|
||||
else:
|
||||
api = None
|
||||
logger.log('{:} create API = {:} done'.format(time_string(), api))
|
||||
@@ -1,47 +0,0 @@
|
||||
#!/bin/bash
|
||||
# bash ./exps/algos-v2/run-all.sh mul
|
||||
# bash ./exps/algos-v2/run-all.sh ws
|
||||
set -e
|
||||
echo script name: $0
|
||||
echo $# arguments
|
||||
if [ "$#" -ne 1 ] ;then
|
||||
echo "Input illegal number of parameters " $#
|
||||
echo "Need 1 parameters for type of algorithms."
|
||||
exit 1
|
||||
fi
|
||||
|
||||
|
||||
datasets="cifar10 cifar100 ImageNet16-120"
|
||||
alg_type=$1
|
||||
|
||||
if [ "$alg_type" == "mul" ]; then
|
||||
search_spaces="tss sss"
|
||||
|
||||
for dataset in ${datasets}
|
||||
do
|
||||
for search_space in ${search_spaces}
|
||||
do
|
||||
python ./exps/algos-v2/reinforce.py --dataset ${dataset} --search_space ${search_space} --learning_rate 0.01
|
||||
python ./exps/algos-v2/regularized_ea.py --dataset ${dataset} --search_space ${search_space} --ea_cycles 200 --ea_population 10 --ea_sample_size 3
|
||||
python ./exps/algos-v2/random_wo_share.py --dataset ${dataset} --search_space ${search_space}
|
||||
python ./exps/algos-v2/bohb.py --dataset ${dataset} --search_space ${search_space} --num_samples 4 --random_fraction 0.0 --bandwidth_factor 3
|
||||
done
|
||||
done
|
||||
|
||||
python exps/experimental/vis-bench-algos.py --search_space tss
|
||||
python exps/experimental/vis-bench-algos.py --search_space sss
|
||||
else
|
||||
seeds="777 888 999"
|
||||
algos="darts-v1 darts-v2 gdas setn random enas"
|
||||
epoch=200
|
||||
for seed in ${seeds}
|
||||
do
|
||||
for alg in ${algos}
|
||||
do
|
||||
python ./exps/algos-v2/search-cell.py --dataset cifar10 --data_path $TORCH_HOME/cifar.python --algo ${alg} --rand_seed ${seed} --overwite_epochs ${epoch}
|
||||
python ./exps/algos-v2/search-cell.py --dataset cifar100 --data_path $TORCH_HOME/cifar.python --algo ${alg} --rand_seed ${seed} --overwite_epochs ${epoch}
|
||||
python ./exps/algos-v2/search-cell.py --dataset ImageNet16-120 --data_path $TORCH_HOME/cifar.python/ImageNet16 --algo ${alg} --rand_seed ${seed} --overwite_epochs ${epoch}
|
||||
done
|
||||
done
|
||||
fi
|
||||
|
||||
Reference in New Issue
Block a user