NNI provides state-of-the-art tuning algorithm as our builtin-tuners and makes them easy to use. Below is the brief summary of NNI currently built-in Tuners:
NNI provides state-of-the-art tuning algorithm as our built-intuners and makes them easy to use. Below is the brief summary of NNI currently built-in tuners:
Note: Click the **Tuner's name** to get the Tuner's installation requirements, suggested scenario and using example. The link for a detailed description of the algorithm is at the end of the suggested scenario of each tuner. Here is an [article](./CommunitySharings/HpoComparision.md) about the comparison of different Tuners on several problems.
Note: Click the **Tuner's name** to get the Tuner's installation requirements, suggested scenario and using example. The link for a detailed description of the algorithm is at the end of the suggested scenario of each tuner. Here is an [article](./CommunitySharings/HpoComparision.md) about the comparison of different Tuners on several problems.
...
@@ -11,28 +11,27 @@ Currently we support the following algorithms:
...
@@ -11,28 +11,27 @@ Currently we support the following algorithms:
|[__TPE__](#TPE)|The Tree-structured Parzen Estimator (TPE) is a sequential model-based optimization (SMBO) approach. SMBO methods sequentially construct models to approximate the performance of hyperparameters based on historical measurements, and then subsequently choose new hyperparameters to test based on this model. [Reference Paper](https://papers.nips.cc/paper/4443-algorithms-for-hyper-parameter-optimization.pdf)|
|[__TPE__](#TPE)|The Tree-structured Parzen Estimator (TPE) is a sequential model-based optimization (SMBO) approach. SMBO methods sequentially construct models to approximate the performance of hyperparameters based on historical measurements, and then subsequently choose new hyperparameters to test based on this model. [Reference Paper](https://papers.nips.cc/paper/4443-algorithms-for-hyper-parameter-optimization.pdf)|
|[__Random Search__](#Random)|In Random Search for Hyper-Parameter Optimization show that Random Search might be surprisingly simple and effective. We suggest that we could use Random Search as the baseline when we have no knowledge about the prior distribution of hyper-parameters. [Reference Paper](http://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf)|
|[__Random Search__](#Random)|In Random Search for Hyper-Parameter Optimization show that Random Search might be surprisingly simple and effective. We suggest that we could use Random Search as the baseline when we have no knowledge about the prior distribution of hyper-parameters. [Reference Paper](http://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf)|
|[__Anneal__](#Anneal)|This simple annealing algorithm begins by sampling from the prior, but tends over time to sample from points closer and closer to the best ones observed. This algorithm is a simple variation on the random search that leverages smoothness in the response surface. The annealing rate is not adaptive.|
|[__Anneal__](#Anneal)|This simple annealing algorithm begins by sampling from the prior, but tends over time to sample from points closer and closer to the best ones observed. This algorithm is a simple variation on the random search that leverages smoothness in the response surface. The annealing rate is not adaptive.|
|[__Naive Evolution__](#Evolution)|Naive Evolution comes from Large-Scale Evolution of Image Classifiers. It randomly initializes a population-based on search space. For each generation, it chooses better ones and does some mutation (e.g., change a hyperparameter, add/remove one layer) on them to get the next generation. Naive Evolution requires many trials to works, but it's very simple and easy to expand new features. [Reference paper](https://arxiv.org/pdf/1703.01041.pdf)|
|[__Naïve Evolution__](#Evolution)|Naïve Evolution comes from Large-Scale Evolution of Image Classifiers. It randomly initializes a population-based on search space. For each generation, it chooses better ones and does some mutation (e.g., change a hyperparameter, add/remove one layer) on them to get the next generation. Naïve Evolution requires many trials to works, but it's very simple and easy to expand new features. [Reference paper](https://arxiv.org/pdf/1703.01041.pdf)|
|[__SMAC__](#SMAC)|SMAC is based on Sequential Model-Based Optimization (SMBO). It adapts the most prominent previously used model class (Gaussian stochastic process models) and introduces the model class of random forests to SMBO, in order to handle categorical parameters. The SMAC supported by nni is a wrapper on the SMAC3 Github repo. Notice, SMAC need to be installed by `nnictl package` command. [Reference Paper,](https://www.cs.ubc.ca/~hutter/papers/10-TR-SMAC.pdf)[Github Repo](https://github.com/automl/SMAC3)|
|[__SMAC__](#SMAC)|SMAC is based on Sequential Model-Based Optimization (SMBO). It adapts the most prominent previously used model class (Gaussian stochastic process models) and introduces the model class of random forests to SMBO, in order to handle categorical parameters. The SMAC supported by NNI is a wrapper on the SMAC3 GitHub repo. Notice, SMAC need to be installed by `nnictl package` command. [Reference Paper,](https://www.cs.ubc.ca/~hutter/papers/10-TR-SMAC.pdf)[GitHub Repo](https://github.com/automl/SMAC3)|
|[__Batch tuner__](#Batch)|Batch tuner allows users to simply provide several configurations (i.e., choices of hyper-parameters) for their trial code. After finishing all the configurations, the experiment is done. Batch tuner only supports the type choice in search space spec.|
|[__Batch tuner__](#Batch)|Batch tuner allows users to simply provide several configurations (i.e., choices of hyper-parameters) for their trial code. After finishing all the configurations, the experiment is done. Batch tuner only supports the type choice in search space spec.|
|[__Grid Search__](#GridSearch)|Grid Search performs an exhaustive searching through a manually specified subset of the hyperparameter space defined in the searchspace file. Note that the only acceptable types of search space are choice, quniform, qloguniform. The number q in quniform and qloguniform has special meaning (different from the spec in search space spec). It means the number of values that will be sampled evenly from the range low and high.|
|[__Grid Search__](#GridSearch)|Grid Search performs an exhaustive searching through a manually specified subset of the hyperparameter space defined in the searchspace file. Note that the only acceptable types of search space are choice, quniform, qloguniform. The number q in quniform and qloguniform has special meaning (different from the spec in search space spec). It means the number of values that will be sampled evenly from the range low and high.|
|[__Hyperband__](#Hyperband)|Hyperband tries to use the limited resource to explore as many configurations as possible, and finds out the promising ones to get the final result. The basic idea is generating many configurations and to run them for the small number of trial budget to find out promising one, then further training those promising ones to select several more promising one.[Reference Paper](https://arxiv.org/pdf/1603.06560.pdf)|
|[__Hyperband__](#Hyperband)|Hyperband tries to use the limited resource to explore as many configurations as possible, and finds out the promising ones to get the final result. The basic idea is generating many configurations and to run them for the small number of trial budget to find out promising one, then further training those promising ones to select several more promising one.[Reference Paper](https://arxiv.org/pdf/1603.06560.pdf)|
|[__Network Morphism__](#NetworkMorphism)|Network Morphism provides functions to automatically search for architecture of deep learning models. Every child network inherits the knowledge from its parent network and morphs into diverse types of networks, including changes of depth, width, and skip-connection. Next, it estimates the value of a child network using the historic architecture and metric pairs. Then it selects the most promising one to train. [Reference Paper](https://arxiv.org/abs/1806.10282)|
|[__Network Morphism__](#NetworkMorphism)|Network Morphism provides functions to automatically search for architecture of deep learning models. Every child network inherits the knowledge from its parent network and morphs into diverse types of networks, including changes of depth, width, and skip-connection. Next, it estimates the value of a child network using the historic architecture and metric pairs. Then it selects the most promising one to train. [Reference Paper](https://arxiv.org/abs/1806.10282)|
|[__Metis Tuner__](#MetisTuner)|Metis offers the following benefits when it comes to tuning parameters: While most tools only predict the optimal configuration, Metis gives you two outputs: (a) current prediction of optimal configuration, and (b) suggestion for the next trial. No more guesswork. While most tools assume training datasets do not have noisy data, Metis actually tells you if you need to re-sample a particular hyper-parameter. [Reference Paper](https://www.microsoft.com/en-us/research/publication/metis-robustly-tuning-tail-latencies-cloud-systems/)|
|[__Metis Tuner__](#MetisTuner)|Metis offers the following benefits when it comes to tuning parameters: While most tools only predict the optimal configuration, Metis gives you two outputs: (a) current prediction of optimal configuration, and (b) suggestion for the next trial. No more guesswork. While most tools assume training datasets do not have noisy data, Metis actually tells you if you need to re-sample a particular hyper-parameter. [Reference Paper](https://www.microsoft.com/en-us/research/publication/metis-robustly-tuning-tail-latencies-cloud-systems/)|
|[__BOHB__](#BOHB)|BOHB is a follow-up work of Hyperband. It targets the weakness of Hyperband that new configurations are generated randomly without leveraging finished trials. For the name BOHB, HB means Hyperband, BO means Byesian Optimization. BOHB leverages finished trials by building multiple TPE models, a proportion of new configurations are generated through these models. [Reference Paper](https://arxiv.org/abs/1807.01774)|
|[__BOHB__](#BOHB)|BOHB is a follow-up work of Hyperband. It targets the weakness of Hyperband that new configurations are generated randomly without leveraging finished trials. For the name BOHB, HB means Hyperband, BO means Bayesian Optimization. BOHB leverages finished trials by building multiple TPE models, a proportion of new configurations are generated through these models. [Reference Paper](https://arxiv.org/abs/1807.01774)|
|[__GP Tuner__](#GPTuner)|Gaussian Process Tuner is a sequential model-based optimization (SMBO) approach with Gaussian Process as the surrogate. [Reference Paper, ](https://papers.nips.cc/paper/4443-algorithms-for-hyper-parameter-optimization.pdf)[Github Repo](https://github.com/fmfn/BayesianOptimization)|
|[__GP Tuner__](#GPTuner)|Gaussian Process Tuner is a sequential model-based optimization (SMBO) approach with Gaussian Process as the surrogate. [Reference Paper](https://papers.nips.cc/paper/4443-algorithms-for-hyper-parameter-optimization.pdf), [Github Repo](https://github.com/fmfn/BayesianOptimization)|
<br>
## Usage of Builtin Tuners
## Usage of Built-in Tuners
Use builtin tuner provided by NNI SDK requires to declare the **builtinTunerName** and **classArgs** in `config.yml` file. In this part, we will introduce the detailed usage about the suggested scenarios, classArg requirements and example for each tuner.
Use built-in tuner provided by NNI SDK requires to declare the **builtinTunerName** and **classArgs** in `config.yml` file. In this part, we will introduce the detailed usage about the suggested scenarios, classArg requirements and example for each tuner.
Note: Please follow the format when you write your `config.yml` file. Some builtin tuner need to be installed by `nnictl package`, like SMAC.
Note: Please follow the format when you write your `config.yml` file. Some built-in tuner need to be installed by `nnictl package`, like SMAC.
**Please note that SMAC doesn't support running on windows currently. The specific reason can be referred to this [github issue](https://github.com/automl/SMAC3/issues/483).**
**Please note that SMAC doesn't support running on windows currently. The specific reason can be referred to this [GitHub issue](https://github.com/automl/SMAC3/issues/483).**
Note that the only acceptable types of search space are `choice`, `quniform`, `uniform` and `randint`.
Note that the only acceptable types of search space are `choice`, `quniform`, `uniform` and `randint`.
**Suggested scenario**
**Suggested scenario**
Similar to TPE and SMAC, Metis is a black-box tuner. If your system takes a long time to finish each trial, Metis is more favorable than other approaches such as random search. Furthermore, Metis provides guidance on the subsequent trial. Here is an [example](https://github.com/Microsoft/nni/tree/master/examples/trials/auto-gbdt/search_space_metis.json) about the use of Metis. User only need to send the final result like `accuracy` to tuner, by calling the nni SDK. [Detailed Description](./MetisTuner.md)
Similar to TPE and SMAC, Metis is a black-box tuner. If your system takes a long time to finish each trial, Metis is more favorable than other approaches such as random search. Furthermore, Metis provides guidance on the subsequent trial. Here is an [example](https://github.com/Microsoft/nni/tree/master/examples/trials/auto-gbdt/search_space_metis.json) about the use of Metis. User only need to send the final result like `accuracy` to tuner, by calling the NNI SDK. [Detailed Description](./MetisTuner.md)
Similar to Hyperband, it is suggested when you have limited computation resource but have relatively large search space. It performs well in the scenario that intermediate result (e.g., accuracy) can reflect good or bad of final result (e.g., accuracy) to some extent. In this case, it may converges to a better configuration due to bayesian optimization usage. [Detailed Description](./BohbAdvisor.md)
Similar to Hyperband, it is suggested when you have limited computation resource but have relatively large search space. It performs well in the scenario that intermediate result (e.g., accuracy) can reflect good or bad of final result (e.g., accuracy) to some extent. In this case, it may converges to a better configuration due to Bayesian optimization usage. [Detailed Description](./BohbAdvisor.md)
Note that the only acceptable types of search space are `choice`, `randint`, `uniform`, `quniform`, `loguniform`, `qloguniform`.
Note that the only acceptable types of search space are `choice`, `randint`, `uniform`, `quniform`, `loguniform`, `qloguniform`.
**Suggested scenario**
**Suggested scenario**
As a strategy in Sequential Model-based Global Optimization(SMBO) algorithm, GP Tuner uses a proxy optimization problem (finding the maximum of the acquisition function) that, albeit still a hard problem, is cheaper (in the computational sense) and common tools can be employed. Therefore GP Tuner is most adequate for situations where the function to be optimized is a very expensive endeavor. GP can be used when the computation resource is limited. While GP Tuner has a computationoal cost that grows at *O(N^3)* due to the requirement of inverting the Gram matrix, so it's not suitable when lots of trials are needed. [Detailed Description](./GPTuner.md)
As a strategy in Sequential Model-based Global Optimization(SMBO) algorithm, GP Tuner uses a proxy optimization problem (finding the maximum of the acquisition function) that, albeit still a hard problem, is cheaper (in the computational sense) and common tools can be employed. Therefore GP Tuner is most adequate for situations where the function to be optimized is a very expensive endeavor. GP can be used when the computation resource is limited. While GP Tuner has a computational cost that grows at *O(N^3)* due to the requirement of inverting the Gram matrix, so it's not suitable when lots of trials are needed. [Detailed Description](./GPTuner.md)
**Requirement of classArg**
**Requirement of classArg**
...
@@ -390,7 +388,7 @@ As a strategy in Sequential Model-based Global Optimization(SMBO) algorithm, GP
...
@@ -390,7 +388,7 @@ As a strategy in Sequential Model-based Global Optimization(SMBO) algorithm, GP
***alpha** (*float, optional, default = 1e-6*) - Used to specify Gaussian Process Regressor. Larger values correspond to increased noise level in the observations.
***alpha** (*float, optional, default = 1e-6*) - Used to specify Gaussian Process Regressor. Larger values correspond to increased noise level in the observations.
***cold_start_num** (*int, optional, default = 10*) - Number of random exploration to perform before Gaussian Process. Random exploration can help by diversifying the exploration space.
***cold_start_num** (*int, optional, default = 10*) - Number of random exploration to perform before Gaussian Process. Random exploration can help by diversifying the exploration space.
***selection_num_warm_up** (*int, optional, default = 1e5*) - Number of random points to evaluate for getting the point which maximizes the acquisition function.
***selection_num_warm_up** (*int, optional, default = 1e5*) - Number of random points to evaluate for getting the point which maximizes the acquisition function.
***selection_num_starting_points** (*int, optional, default = 250*) - Nnumber of times to run L-BFGS-B from a random starting point after the warmup.
***selection_num_starting_points** (*int, optional, default = 250*) - Number of times to run L-BFGS-B from a random starting point after the warmup.
3. 参考此[指南](https://docs.microsoft.com/en-us/azure/storage/common/storage-quickstart-create-account?tabs=portal)来创建 Azure 文件存储账户。 NNI 需要 Azure Storage Service 来存取代码和输出文件。
3. 参考此[指南](https://docs.microsoft.com/zh-cn/azure/storage/common/storage-quickstart-create-account?tabs=portal)来创建 Azure 文件存储账户。 NNI 需要 Azure Storage Service 来存取代码和输出文件。
3. 在 Azure Kubernetes Service 上部署 Kubeflow，参考此[指南](https://www.kubeflow.org/docs/started/getting-started/)。
3. 在 Azure Kubernetes Service 上部署 Kubeflow，参考此[指南](https://www.kubeflow.org/docs/started/getting-started/)。
4. 参考此[指南](https://docs.microsoft.com/en-us/azure/storage/common/storage-quickstart-create-account?tabs=portal)来创建 Azure 文件存储账户。 NNI 需要 Azure Storage Service 来存取代码和输出文件。
4. 参考此[指南](https://docs.microsoft.com/zh-cn/azure/storage/common/storage-quickstart-create-account?tabs=portal)来创建 Azure 文件存储账户。 NNI 需要 Azure Storage Service 来存取代码和输出文件。
