NAS Benchmarks (experimental)¶

Introduction¶

To imporve the reproducibility of NAS algorithms as well as reducing computing resource requirements, researchers proposed a series of NAS benchmarks such as NAS-Bench-101, NAS-Bench-201, NDS, etc. NNI provides a query interface for users to acquire these benchmarks. Within just a few lines of code, researcher are able to evaluate their NAS algorithms easily and fairly by utilizing these benchmarks.

Prerequisites¶

Please prepare a folder to household all the benchmark databases. By default, it can be found at ${HOME}/.nni/nasbenchmark. You can place it anywhere you like, and specify it in NASBENCHMARK_DIR before importing NNI.
Please install peewee via pip install peewee, which NNI uses to connect to database.

Data Preparation¶

To avoid storage and legality issues, we do not provide any prepared databases. Please follow the following steps.

Clone NNI to your machine and enter examples/nas/benchmarks directory.

git clone -b ${NNI_VERSION} https://github.com/microsoft/nni
cd nni/examples/nas/benchmarks

Replace ${NNI_VERSION} with a released version name or branch name, e.g., v1.7.

Install dependencies via pip3 install -r xxx.requirements.txt. xxx can be nasbench101, nasbench201 or nds.
Generate the database via ./xxx.sh. The directory that stores the benchmark file can be configured with NASBENCHMARK_DIR environment variable, which defaults to ~/.nni/nasbenchmark. Note that the NAS-Bench-201 dataset will be downloaded from a google drive.

Please make sure there is at least 10GB free disk space and note that the conversion process can take up to hours to complete.

Example Usages¶

Please refer to examples usages of Benchmarks API.

NAS-Bench-101¶

Paper link Open-source

NAS-Bench-101 contains 423,624 unique neural networks, combined with 4 variations in number of epochs (4, 12, 36, 108), each of which is trained 3 times. It is a cell-wise search space, which constructs and stacks a cell by enumerating DAGs with at most 7 operators, and no more than 9 connections. All operators can be chosen from CONV3X3_BN_RELU, CONV1X1_BN_RELU and MAXPOOL3X3, except the first operator (always INPUT) and last operator (always OUTPUT).

Notably, NAS-Bench-101 eliminates invalid cells (e.g., there is no path from input to output, or there is redundant computation). Furthermore, isomorphic cells are de-duplicated, i.e., all the remaining cells are computationally unique.

API Documentation¶

nni.nas.benchmarks.nasbench101.query_nb101_trial_stats(arch, num_epochs, isomorphism=True, reduction=None)[source]¶

Query trial stats of NAS-Bench-101 given conditions.

Parameters:	arch (dict or None) – If a dict, it is in the format that is described in `nni.nas.benchmark.nasbench101.Nb101TrialConfig`. Only trial stats matched will be returned. If none, architecture will be a wildcard. num_epochs (int or None) – If int, matching results will be returned. Otherwise a wildcard. isomorphism (boolean) – Whether to match essentially-same architecture, i.e., architecture with the same graph-invariant hash value. reduction (str or None) – If ‘none’ or None, all trial stats will be returned directly. If ‘mean’, fields in trial stats will be averaged given the same trial config.
Returns:	A generator of `nni.nas.benchmark.nasbench101.Nb101TrialStats` objects, where each of them has been converted into a dict.
Return type:	generator of dict

nasbench101.INPUT = 'input'¶

nasbench101.OUTPUT = 'output'¶

nasbench101.CONV3X3_BN_RELU = 'conv3x3-bn-relu'¶

nasbench101.CONV1X1_BN_RELU = 'conv1x1-bn-relu'¶

nasbench101.MAXPOOL3X3 = 'maxpool3x3'¶

class nni.nas.benchmarks.nasbench101.Nb101TrialConfig(*args, **kwargs)[source]¶

Trial config for NAS-Bench-101.

arch¶

A dict with keys op1, op2, … and input1, input2, … Vertices are enumerate from 0. Since node 0 is input node, it is skipped in this dict. Each op is one of nni.nas.benchmark.nasbench101.CONV3X3_BN_RELU, nni.nas.benchmark.nasbench101.CONV1X1_BN_RELU, and nni.nas.benchmark.nasbench101.MAXPOOL3X3. Each input is a list of previous nodes. For example input5 can be [0, 1, 3].

Type:	dict

num_vertices¶

Number of vertices (nodes) in one cell. Should be less than or equal to 7 in default setup.

Type:	int

hash¶

Graph-invariant MD5 string for this architecture.

Type:	str

num_epochs¶

Number of epochs planned for this trial. Should be one of 4, 12, 36, 108 in default setup.

Type:	int

class nni.nas.benchmarks.nasbench101.Nb101TrialStats(*args, **kwargs)[source]¶

Computation statistics for NAS-Bench-101. Each corresponds to one trial. Each config has multiple trials with different random seeds, but unfortunately seed for each trial is unavailable. NAS-Bench-101 trains and evaluates on CIFAR-10 by default. The original training set is divided into 40k training images and 10k validation images, and the original validation set is used for test only.

config¶

Setup for this trial data.

Type:	Nb101TrialConfig

train_acc¶

Final accuracy on training data, ranging from 0 to 100.

Type:	float

valid_acc¶

Final accuracy on validation data, ranging from 0 to 100.

Type:	float

test_acc¶

Final accuracy on test data, ranging from 0 to 100.

Type:	float

parameters¶

Number of trainable parameters in million.

Type:	float

training_time¶

Duration of training in seconds.

Type:	float

class nni.nas.benchmarks.nasbench101.Nb101IntermediateStats(*args, **kwargs)[source]¶

Intermediate statistics for NAS-Bench-101.

trial¶

The exact trial where the intermediate result is produced.

Type:	Nb101TrialStats

current_epoch¶

Elapsed epochs when evaluation is done.

Type:	int

train_acc¶

Intermediate accuracy on training data, ranging from 0 to 100.

Type:	float

valid_acc¶

Intermediate accuracy on validation data, ranging from 0 to 100.

Type:	float

test_acc¶

Intermediate accuracy on test data, ranging from 0 to 100.

Type:	float

training_time¶

Time elapsed in seconds.

Type:	float

nni.nas.benchmarks.nasbench101.graph_util.nasbench_format_to_architecture_repr(adjacency_matrix, labeling)[source]¶

Computes a graph-invariance MD5 hash of the matrix and label pair. Imported from NAS-Bench-101 repo.

Parameters:	adjacency_matrix (np.ndarray) – A 2D array of shape NxN, where N is the number of vertices. `matrix[u][v]` is 1 if there is a direct edge from u to v, otherwise it will be 0. labeling (list of str) – A list of str that starts with input and ends with output. The intermediate nodes are chosen from candidate operators.
Returns:	Converted number of vertices and architecture.
Return type:	tuple and int and dict

nni.nas.benchmarks.nasbench101.graph_util.infer_num_vertices(architecture)[source]¶

Infer number of vertices from an architecture dict.

Parameters:	architecture (dict) – Architecture in NNI format.
Returns:	Number of vertices.
Return type:	int

nni.nas.benchmarks.nasbench101.graph_util.hash_module(architecture, vertices)[source]¶

Computes a graph-invariance MD5 hash of the matrix and label pair. This snippet is modified from code in NAS-Bench-101 repo.

Parameters:	matrix (np.ndarray) – Square upper-triangular adjacency matrix. labeling (list of int) – Labels of length equal to both dimensions of matrix.
Returns:	MD5 hash of the matrix and labeling.
Return type:	str

NAS-Bench-201¶

Paper link Open-source API Implementations

NAS-Bench-201 is a cell-wise search space that views nodes as tensors and edges as operators. The search space contains all possible densely-connected DAGs with 4 nodes, resulting in 15,625 candidates in total. Each operator (i.e., edge) is selected from a pre-defined operator set (NONE, SKIP_CONNECT, CONV_1X1, CONV_3X3 and AVG_POOL_3X3). Training appraoches vary in the dataset used (CIFAR-10, CIFAR-100, ImageNet) and number of epochs scheduled (12 and 200). Each combination of architecture and training approach is repeated 1 - 3 times with different random seeds.

API Documentation¶

nni.nas.benchmarks.nasbench201.query_nb201_trial_stats(arch, num_epochs, dataset, reduction=None)[source]¶

Query trial stats of NAS-Bench-201 given conditions.

Parameters:	arch (dict or None) – If a dict, it is in the format that is described in `nni.nas.benchmark.nasbench201.Nb201TrialConfig`. Only trial stats matched will be returned. If none, architecture will be a wildcard. num_epochs (int or None) – If int, matching results will be returned. Otherwise a wildcard. dataset (str or None) – If specified, can be one of the dataset available in `nni.nas.benchmark.nasbench201.Nb201TrialConfig`. Otherwise a wildcard. reduction (str or None) – If ‘none’ or None, all trial stats will be returned directly. If ‘mean’, fields in trial stats will be averaged given the same trial config.
Returns:	A generator of `nni.nas.benchmark.nasbench201.Nb201TrialStats` objects, where each of them has been converted into a dict.
Return type:	generator of dict

nasbench201.NONE = 'none'¶

nasbench201.SKIP_CONNECT = 'skip_connect'¶

nasbench201.CONV_1X1 = 'conv_1x1'¶

nasbench201.CONV_3X3 = 'conv_3x3'¶

nasbench201.AVG_POOL_3X3 = 'avg_pool_3x3'¶

class nni.nas.benchmarks.nasbench201.Nb201TrialConfig(*args, **kwargs)[source]¶

Trial config for NAS-Bench-201.

arch¶

A dict with keys 0_1, 0_2, 0_3, 1_2, 1_3, 2_3, each of which is an operator chosen from nni.nas.benchmark.nasbench201.NONE, nni.nas.benchmark.nasbench201.SKIP_CONNECT, nni.nas.benchmark.nasbench201.CONV_1X1, nni.nas.benchmark.nasbench201.CONV_3X3 and nni.nas.benchmark.nasbench201.AVG_POOL_3X3.

Type:	dict

num_epochs¶

Number of epochs planned for this trial. Should be one of 12 and 200.

Type:	int

num_channels¶

Number of channels for initial convolution. 16 by default.

Type:	int

num_cells¶

Number of cells per stage. 5 by default.

Type:	int

dataset¶

Dataset used for training and evaluation. NAS-Bench-201 provides the following 4 options: cifar10-valid (training data is splited into 25k for training and 25k for validation, validation data is used for test), cifar10 (training data is used in training, validation data is splited into 25k for validation and 25k for testing), cifar100 (same protocol as cifar10), and imagenet16-120 (a subset of 120 classes in ImageNet, downscaled to 16x16, using training data for training, 6k images from validation set for validation and the other 6k for testing).

Type:	str

class nni.nas.benchmarks.nasbench201.Nb201TrialStats(*args, **kwargs)[source]¶

Computation statistics for NAS-Bench-201. Each corresponds to one trial.

config¶

Setup for this trial data.

Type:	Nb201TrialConfig

seed¶

Random seed selected, for reproduction.

Type:	int

train_acc¶

Final accuracy on training data, ranging from 0 to 100.

Type:	float

valid_acc¶

Final accuracy on validation data, ranging from 0 to 100.

Type:	float

test_acc¶

Final accuracy on test data, ranging from 0 to 100.

Type:	float

ori_test_acc¶

Test accuracy on original validation set (10k for CIFAR and 12k for Imagenet16-120), ranging from 0 to 100.

Type:	float

train_loss¶

Final cross entropy loss on training data. Note that loss could be NaN, in which case this attributed will be None.

Type:	float or None

valid_loss¶

Final cross entropy loss on validation data.

Type:	float or None

test_loss¶

Final cross entropy loss on test data.

Type:	float or None

ori_test_loss¶

Final cross entropy loss on original validation set.

Type:	float or None

parameters¶

Number of trainable parameters in million.

Type:	float

latency¶

Latency in seconds.

Type:	float

flops¶

FLOPs in million.

Type:	float

training_time¶

Duration of training in seconds.

Type:	float

valid_evaluation_time¶

Time elapsed to evaluate on validation set.

Type:	float

test_evaluation_time¶

Time elapsed to evaluate on test set.

Type:	float

ori_test_evaluation_time¶

Time elapsed to evaluate on original test set.

Type:	float

class nni.nas.benchmarks.nasbench201.Nb201IntermediateStats(*args, **kwargs)[source]¶

Intermediate statistics for NAS-Bench-201.

trial¶

Corresponding trial.

Type:	Nb201TrialStats

current_epoch¶

Elapsed epochs.

Type:	int

train_acc¶

Current accuracy on training data, ranging from 0 to 100.

Type:	float

valid_acc¶

Current accuracy on validation data, ranging from 0 to 100.

Type:	float

test_acc¶

Current accuracy on test data, ranging from 0 to 100.

Type:	float

ori_test_acc¶

Test accuracy on original validation set (10k for CIFAR and 12k for Imagenet16-120), ranging from 0 to 100.

Type:	float

train_loss¶

Current cross entropy loss on training data.

Type:	float or None

valid_loss¶

Current cross entropy loss on validation data.

Type:	float or None

test_loss¶

Current cross entropy loss on test data.

Type:	float or None

ori_test_loss¶

Current cross entropy loss on original validation set.

Type:	float or None

NDS¶

Paper link Open-source

On Network Design Spaces for Visual Recognition released trial statistics of over 100,000 configurations (models + hyper-parameters) sampled from multiple model families, including vanilla (feedforward network loosely inspired by VGG), ResNet and ResNeXt (residual basic block and residual bottleneck block) and NAS cells (following popular design from NASNet, Ameoba, PNAS, ENAS and DARTS). Most configurations are trained only once with a fixed seed, except a few that are trained twice or three times.

Instead of storing results obtained with different configurations in separate files, we dump them into one single database to enable comparison in multiple dimensions. Specifically, we use model_family to distinguish model types, model_spec for all hyper-parameters needed to build this model, cell_spec for detailed information on operators and connections if it is a NAS cell, generator to denote the sampling policy through which this configuration is generated. Refer to API documentation for details.

Available Operators¶

Here is a list of available operators used in NDS.

constants.NONE = 'none'¶

constants.SKIP_CONNECT = 'skip_connect'¶

constants.AVG_POOL_3X3 = 'avg_pool_3x3'¶

constants.MAX_POOL_3X3 = 'max_pool_3x3'¶

constants.MAX_POOL_5X5 = 'max_pool_5x5'¶

constants.MAX_POOL_7X7 = 'max_pool_7x7'¶

constants.CONV_1X1 = 'conv_1x1'¶

constants.CONV_3X3 = 'conv_3x3'¶

constants.CONV_3X1_1X3 = 'conv_3x1_1x3'¶

constants.CONV_7X1_1X7 = 'conv_7x1_1x7'¶

constants.DIL_CONV_3X3 = 'dil_conv_3x3'¶

constants.DIL_CONV_5X5 = 'dil_conv_5x5'¶

constants.SEP_CONV_3X3 = 'sep_conv_3x3'¶

constants.SEP_CONV_5X5 = 'sep_conv_5x5'¶

constants.SEP_CONV_7X7 = 'sep_conv_7x7'¶

constants.DIL_SEP_CONV_3X3 = 'dil_sep_conv_3x3'¶

API Documentation¶

nni.nas.benchmarks.nds.query_nds_trial_stats(model_family, proposer, generator, model_spec, cell_spec, dataset, num_epochs=None, reduction=None)[source]¶

Query trial stats of NDS given conditions.

Parameters:	model_family (str or None) – If str, can be one of the model families available in `nni.nas.benchmark.nds.NdsTrialConfig`. Otherwise a wildcard. proposer (str or None) – If str, can be one of the proposers available in `nni.nas.benchmark.nds.NdsTrialConfig`. Otherwise a wildcard. generator (str or None) – If str, can be one of the generators available in `nni.nas.benchmark.nds.NdsTrialConfig`. Otherwise a wildcard. model_spec (dict or None) – If specified, can be one of the model spec available in `nni.nas.benchmark.nds.NdsTrialConfig`. Otherwise a wildcard. cell_spec (dict or None) – If specified, can be one of the cell spec available in `nni.nas.benchmark.nds.NdsTrialConfig`. Otherwise a wildcard. dataset (str or None) – If str, can be one of the datasets available in `nni.nas.benchmark.nds.NdsTrialConfig`. Otherwise a wildcard. num_epochs (float or None) – If int, matching results will be returned. Otherwise a wildcard. reduction (str or None) – If ‘none’ or None, all trial stats will be returned directly. If ‘mean’, fields in trial stats will be averaged given the same trial config.
Returns:	A generator of `nni.nas.benchmark.nds.NdsTrialStats` objects, where each of them has been converted into a dict.
Return type:	generator of dict

class nni.nas.benchmarks.nds.NdsTrialConfig(*args, **kwargs)[source]¶

Trial config for NDS.

model_family¶

Could be nas_cell, residual_bottleneck, residual_basic or vanilla.

Type:	str

model_spec¶

If model_family is nas_cell, it contains num_nodes_normal, num_nodes_reduce, depth, width, aux and drop_prob. If model_family is residual_bottleneck, it contains bot_muls, ds (depths), num_gs (number of groups) and ss (strides). If model_family is residual_basic or vanilla, it contains ds, ss and ws.

Type:	dict

cell_spec¶

If model_family is not nas_cell it will be an empty dict. Otherwise, it specifies <normal/reduce>_<i>_<op/input>_<x/y>, where i ranges from 0 to num_nodes_<normal/reduce> - 1. If it is an op, the value is chosen from the constants specified previously like nni.nas.benchmark.nds.CONV_1X1. If it is i’s input, the value range from 0 to i + 1, as nas_cell uses previous two nodes as inputs, and node 0 is actually the second node. Refer to NASNet paper for details. Finally, another two key-value pairs normal_concat and reduce_concat specify which nodes are eventually concatenated into output.

Type:	dict

dataset¶

Dataset used. Could be cifar10 or imagenet.

Type:	str

generator¶

Can be one of random which generates configurations at random, while keeping learning rate and weight decay fixed, fix_w_d which further keeps width and depth fixed, only applicable for nas_cell. tune_lr_wd which further tunes learning rate and weight decay.

Type:	str

proposer¶

Paper who has proposed the distribution for random sampling. Available proposers include nasnet, darts, enas, pnas, amoeba, vanilla, resnext-a, resnext-b, resnet, resnet-b (ResNet with bottleneck). See NDS paper for details.

Type:	str

base_lr¶

Initial learning rate.

Type:	float

weight_decay¶

L2 weight decay applied on weights.

Type:	float

num_epochs¶

Number of epochs scheduled, during which learning rate will decay to 0 following cosine annealing.

Type:	int

class nni.nas.benchmarks.nds.NdsTrialStats(*args, **kwargs)[source]¶

Computation statistics for NDS. Each corresponds to one trial.

config¶

Corresponding config for trial.

Type:	NdsTrialConfig

seed¶

Random seed selected, for reproduction.

Type:	int

final_train_acc¶

Final accuracy on training data, ranging from 0 to 100.

Type:	float

final_train_loss¶

Final cross entropy loss on training data. Could be NaN (None).

Type:	float or None

final_test_acc¶

Final accuracy on test data, ranging from 0 to 100.

Type:	float

best_train_acc¶

Best accuracy on training data, ranging from 0 to 100.

Type:	float

best_train_loss¶

Best cross entropy loss on training data. Could be NaN (None).

Type:	float or None

best_test_acc¶

Best accuracy on test data, ranging from 0 to 100.

Type:	float

parameters¶

Number of trainable parameters in million.

Type:	float

flops¶

FLOPs in million.

Type:	float

iter_time¶

Seconds elapsed for each iteration.

Type:	float

class nni.nas.benchmarks.nds.NdsIntermediateStats(*args, **kwargs)[source]¶

Intermediate statistics for NDS.

trial¶

Corresponding trial.

Type:	NdsTrialStats

current_epoch¶

Elapsed epochs.

Type:	int

train_loss¶

Current cross entropy loss on training data. Can be NaN (None).

Type:	float or None

train_acc¶

Current accuracy on training data, ranging from 0 to 100.

Type:	float

test_acc¶

Current accuracy on test data, ranging from 0 to 100.

Type:	float