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 random search that leverages smoothness in the response surface. The annealing rate is not adaptive.
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 the `search space spec <../Tutorial/SearchSpaceSpec.rst>`__.
Suggested scenario: If the configurations you want to try have been decided, you can list them in the SearchSpace file (using ``choice``\ ) and run them using the batch tuner.
Suggested scenario: If the configurations you want to try have been decided, you can list them in the SearchSpace file (using ``choice``) and run them using the batch tuner.
Usage
-----
...
...
@@ -20,8 +17,6 @@ Example Configuration
tuner:
name: BatchTuner
:raw-html:`<br>`
Note that the search space for BatchTuner should look like:
BOHB is a robust and efficient hyperparameter tuning algorithm mentioned in `this reference paper <https://arxiv.org/abs/1807.01774>`__. BO is an abbreviation for "Bayesian Optimization" and HB is an abbreviation for "Hyperband".
...
...
@@ -81,16 +78,16 @@ BOHB advisor requires the `ConfigSpace <https://github.com/automl/ConfigSpace>`_
classArgs Requirements
^^^^^^^^^^^^^^^^^^^^^^
* **optimize_mode** (*maximize or minimize, optional, default = maximize*\ ) - If 'maximize', tuners will try to maximize metrics. If 'minimize', tuner will try to minimize metrics.
* **min_budget** (*int, optional, default = 1*\ ) - The smallest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be positive.
* **max_budget** (*int, optional, default = 3*\ ) - The largest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be larger than min_budget.
* **eta** (*int, optional, default = 3*\ ) - In each iteration, a complete run of sequential halving is executed. In it, after evaluating each configuration on the same subset size, only a fraction of 1/eta of them 'advances' to the next round. Must be greater or equal to 2.
* **min_points_in_model**\ (*int, optional, default = None*\ ): number of observations to start building a KDE. Default 'None' means dim+1; when the number of completed trials in this budget is equal to or larger than ``max{dim+1, min_points_in_model}``\ , BOHB will start to build a KDE model of this budget then use said KDE model to guide configuration selection. Needs to be positive. (dim means the number of hyperparameters in search space)
* **top_n_percent**\ (*int, optional, default = 15*\ ): percentage (between 1 and 99) of the observations which are considered good. Good points and bad points are used for building KDE models. For example, if you have 100 observed trials and top_n_percent is 15, then the top 15% of points will be used for building the good points models "l(x)". The remaining 85% of points will be used for building the bad point models "g(x)".
* **num_samples**\ (*int, optional, default = 64*\ ): number of samples to optimize EI (default 64). In this case, we will sample "num_samples" points and compare the result of l(x)/g(x). Then we will return the one with the maximum l(x)/g(x) value as the next configuration if the optimize_mode is ``maximize``. Otherwise, we return the smallest one.
* **random_fraction**\ (*float, optional, default = 0.33*\ ): fraction of purely random configurations that are sampled from the prior without the model.
* **bandwidth_factor**\ (*float, optional, default = 3.0*\ ): to encourage diversity, the points proposed to optimize EI are sampled from a 'widened' KDE where the bandwidth is multiplied by this factor. We suggest using the default value if you are not familiar with KDE.
* **min_bandwidth**\ (*float, optional, default = 0.001*\ ): to keep diversity, even when all (good) samples have the same value for one of the parameters, a minimum bandwidth (default: 1e-3) is used instead of zero. We suggest using the default value if you are not familiar with KDE.
* **optimize_mode** (*maximize or minimize, optional, default = maximize*) - If 'maximize', tuners will try to maximize metrics. If 'minimize', tuner will try to minimize metrics.
* **min_budget** (*int, optional, default = 1*) - The smallest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be positive.
* **max_budget** (*int, optional, default = 3*) - The largest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be larger than min_budget.
* **eta** (*int, optional, default = 3*) - In each iteration, a complete run of sequential halving is executed. In it, after evaluating each configuration on the same subset size, only a fraction of 1/eta of them 'advances' to the next round. Must be greater or equal to 2.
* **min_points_in_model** (*int, optional, default = None*): number of observations to start building a KDE. Default 'None' means dim+1; when the number of completed trials in this budget is equal to or larger than ``max{dim+1, min_points_in_model}``, BOHB will start to build a KDE model of this budget then use said KDE model to guide configuration selection. Needs to be positive. (dim means the number of hyperparameters in search space)
* **top_n_percent** (*int, optional, default = 15*): percentage (between 1 and 99) of the observations which are considered good. Good points and bad points are used for building KDE models. For example, if you have 100 observed trials and top_n_percent is 15, then the top 15% of points will be used for building the good points models "l(x)". The remaining 85% of points will be used for building the bad point models "g(x)".
* **num_samples** (*int, optional, default = 64*): number of samples to optimize EI (default 64). In this case, we will sample "num_samples" points and compare the result of l(x)/g(x). Then we will return the one with the maximum l(x)/g(x) value as the next configuration if the optimize_mode is ``maximize``. Otherwise, we return the smallest one.
* **random_fraction** (*float, optional, default = 0.33*): fraction of purely random configurations that are sampled from the prior without the model.
* **bandwidth_factor** (*float, optional, default = 3.0*): to encourage diversity, the points proposed to optimize EI are sampled from a 'widened' KDE where the bandwidth is multiplied by this factor. We suggest using the default value if you are not familiar with KDE.
* **min_bandwidth** (*float, optional, default = 0.001*): to keep diversity, even when all (good) samples have the same value for one of the parameters, a minimum bandwidth (default: 1e-3) is used instead of zero. We suggest using the default value if you are not familiar with KDE.
*Please note that the float type currently only supports decimal representations. You have to use 0.333 instead of 1/3 and 0.001 instead of 1e-3.*
...
...
@@ -119,16 +116,16 @@ To use BOHB, you should add the following spec in your experiment's YAML config
**classArgs Requirements:**
* **optimize_mode** (*maximize or minimize, optional, default = maximize*\ ) - If 'maximize', tuners will try to maximize metrics. If 'minimize', tuner will try to minimize metrics.
* **min_budget** (*int, optional, default = 1*\ ) - The smallest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be positive.
* **max_budget** (*int, optional, default = 3*\ ) - The largest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be larger than min_budget.
* **eta** (*int, optional, default = 3*\ ) - In each iteration, a complete run of sequential halving is executed. In it, after evaluating each configuration on the same subset size, only a fraction of 1/eta of them 'advances' to the next round. Must be greater or equal to 2.
* **min_points_in_model**\ (*int, optional, default = None*\ ): number of observations to start building a KDE. Default 'None' means dim+1; when the number of completed trials in this budget is equal to or larger than ``max{dim+1, min_points_in_model}``\ , BOHB will start to build a KDE model of this budget then use said KDE model to guide configuration selection. Needs to be positive. (dim means the number of hyperparameters in search space)
* **top_n_percent**\ (*int, optional, default = 15*\ ): percentage (between 1 and 99) of the observations which are considered good. Good points and bad points are used for building KDE models. For example, if you have 100 observed trials and top_n_percent is 15, then the top 15% of points will be used for building the good points models "l(x)". The remaining 85% of points will be used for building the bad point models "g(x)".
* **num_samples**\ (*int, optional, default = 64*\ ): number of samples to optimize EI (default 64). In this case, we will sample "num_samples" points and compare the result of l(x)/g(x). Then we will return the one with the maximum l(x)/g(x) value as the next configuration if the optimize_mode is ``maximize``. Otherwise, we return the smallest one.
* **random_fraction**\ (*float, optional, default = 0.33*\ ): fraction of purely random configurations that are sampled from the prior without the model.
* **bandwidth_factor**\ (*float, optional, default = 3.0*\ ): to encourage diversity, the points proposed to optimize EI are sampled from a 'widened' KDE where the bandwidth is multiplied by this factor. We suggest using the default value if you are not familiar with KDE.
* **min_bandwidth**\ (*float, optional, default = 0.001*\ ): to keep diversity, even when all (good) samples have the same value for one of the parameters, a minimum bandwidth (default: 1e-3) is used instead of zero. We suggest using the default value if you are not familiar with KDE.
* **optimize_mode** (*maximize or minimize, optional, default = maximize*) - If 'maximize', tuners will try to maximize metrics. If 'minimize', tuner will try to minimize metrics.
* **min_budget** (*int, optional, default = 1*) - The smallest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be positive.
* **max_budget** (*int, optional, default = 3*) - The largest budget to assign to a trial job, (budget can be the number of mini-batches or epochs). Needs to be larger than min_budget.
* **eta** (*int, optional, default = 3*) - In each iteration, a complete run of sequential halving is executed. In it, after evaluating each configuration on the same subset size, only a fraction of 1/eta of them 'advances' to the next round. Must be greater or equal to 2.
* **min_points_in_model** (*int, optional, default = None*): number of observations to start building a KDE. Default 'None' means dim+1; when the number of completed trials in this budget is equal to or larger than ``max{dim+1, min_points_in_model}``, BOHB will start to build a KDE model of this budget then use said KDE model to guide configuration selection. Needs to be positive. (dim means the number of hyperparameters in search space)
* **top_n_percent** (*int, optional, default = 15*): percentage (between 1 and 99) of the observations which are considered good. Good points and bad points are used for building KDE models. For example, if you have 100 observed trials and top_n_percent is 15, then the top 15% of points will be used for building the good points models "l(x)". The remaining 85% of points will be used for building the bad point models "g(x)".
* **num_samples** (*int, optional, default = 64*): number of samples to optimize EI (default 64). In this case, we will sample "num_samples" points and compare the result of l(x)/g(x). Then we will return the one with the maximum l(x)/g(x) value as the next configuration if the optimize_mode is ``maximize``. Otherwise, we return the smallest one.
* **random_fraction** (*float, optional, default = 0.33*): fraction of purely random configurations that are sampled from the prior without the model.
* **bandwidth_factor** (*float, optional, default = 3.0*): to encourage diversity, the points proposed to optimize EI are sampled from a 'widened' KDE where the bandwidth is multiplied by this factor. We suggest using the default value if you are not familiar with KDE.
* **min_bandwidth** (*float, optional, default = 0.001*): to keep diversity, even when all (good) samples have the same value for one of the parameters, a minimum bandwidth (default: 1e-3) is used instead of zero. We suggest using the default value if you are not familiar with KDE.
* **config_space** (*str, optional*): directly use a .pcs file serialized by `ConfigSpace <https://automl.github.io/ConfigSpace/>` in "pcs new" format. In this case, search space file (if provided in config) will be ignored. Note that this path needs to be an absolute path. Relative path is currently not supported.
*Please note that the float type currently only supports decimal representations. You have to use 0.333 instead of 1/3 and 0.001 instead of 1e-3.*
@@ -17,235 +13,88 @@ Currently, we support the following algorithms:
* - Tuner
- Brief Introduction of Algorithm
* - `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>`__
* - `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.
* - `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 work, 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 needs to be installed by ``pip install nni[SMAC]`` 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.
* - `Grid Search <#GridSearch>`__
- Grid Search performs an exhaustive searching through the search space.
* - `Hyperband <#Hyperband>`__
- Hyperband tries to use limited resources to explore as many configurations as possible and returns the most promising ones as a final result. The basic idea is to generate many configurations and run them for a small number of trials. The half least-promising configurations are thrown out, the remaining are further trained along with a selection of new configurations. The size of these populations is sensitive to resource constraints (e.g. allotted search time). `Reference Paper <https://arxiv.org/pdf/1603.06560.pdf>`__
* - `Network Morphism <#NetworkMorphism>`__
- Network Morphism provides functions to automatically search for deep learning architectures. It generates child networks that inherit the knowledge from their parent network which it is a morph from. This includes changes in depth, width, and skip-connections. Next, it estimates the value of a child network using 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/>`__
* - `BOHB <#BOHB>`__
- BOHB is a follow-up work to 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>`__
* - `PBT Tuner <#PBTTuner>`__
- PBT Tuner is a simple asynchronous optimization algorithm which effectively utilizes a fixed computational budget to jointly optimize a population of models and their hyperparameters to maximize performance. `Reference Paper <https://arxiv.org/abs/1711.09846v1>`__
* - `DNGO Tuner <#DNGOTuner>`__
- Use of neural networks as an alternative to GPs to model distributions over functions in bayesian optimization.
Usage of Built-in Tuners
------------------------
Using a built-in tuner provided by the NNI SDK requires one to declare the **name** and **classArgs** in the ``config.yml`` file. In this part, we will introduce each tuner along with information about usage and suggested scenarios, classArg requirements, and an example configuration.
Note: Please follow the format when you write your ``config.yml`` file. Some built-in tuners have dependencies that need to be installed using ``pip install nni[<tuner>]``, like SMAC's dependencies can be installed using ``pip install nni[SMAC]``.
:raw-html:`<a name="TPE"></a>`
TPE
^^^
..
Built-in Tuner Name: **TPE**
TPE, as a black-box optimization, can be used in various scenarios and shows good performance in general. Especially when you have limited computation resources and can only try a small number of trials. From a large amount of experiments, we found that TPE is far better than Random Search. `Detailed Description <./TpeTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="Random"></a>`
Random Search
^^^^^^^^^^^^^
..
Built-in Tuner Name: **Random**
Random search is suggested when each trial does not take very long (e.g., each trial can be completed very quickly, or early stopped by the assessor), and you have enough computational resources. It's also useful if you want to uniformly explore the search space. Random Search can be considered a baseline search algorithm. `Detailed Description <./RandomTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="Anneal"></a>`
Anneal
^^^^^^
..
Built-in Tuner Name: **Anneal**
Anneal is suggested when each trial does not take very long and you have enough computation resources (very similar to Random Search). It's also useful when the variables in the search space can be sample from some prior distribution. `Detailed Description <./HyperoptTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="Evolution"></a>`
Naïve Evolution
^^^^^^^^^^^^^^^
..
Built-in Tuner Name: **Evolution**
Its computational resource requirements are relatively high. Specifically, it requires a large initial population to avoid falling into a local optimum. If your trial is short or leverages assessor, this tuner is a good choice. It is also suggested when your trial code supports weight transfer; that is, the trial could inherit the converged weights from its parent(s). This can greatly speed up the training process. `Detailed Description <./EvolutionTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="SMAC"></a>`
SMAC
^^^^
..
Built-in Tuner Name: **SMAC**
**Please note that SMAC doesn't support running on Windows currently**. For the specific reason, please refer to this `GitHub issue <https://github.com/automl/SMAC3/issues/483>`__.
Similar to TPE, SMAC is also a black-box tuner that can be tried in various scenarios and is suggested when computational resources are limited. It is optimized for discrete hyperparameters, thus, it's suggested when most of your hyperparameters are discrete. `Detailed Description <./SmacTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="Batch"></a>`
Batch Tuner
^^^^^^^^^^^
..
Built-in Tuner Name: BatchTuner
**Suggested scenario**
If the configurations you want to try have been decided beforehand, you can list them in search space file (using ``choice``\ ) and run them using batch tuner. `Detailed Description <./BatchTuner.rst>`__
:raw-html:`<a name="GridSearch"></a>`
Grid Search
^^^^^^^^^^^
..
Built-in Tuner Name: **Grid Search**
This is suggested when the search space is small. It's suggested when it is feasible to exhaustively sweep the whole search space. `Detailed Description <./GridsearchTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="Hyperband"></a>`
Hyperband
^^^^^^^^^
..
Built-in Advisor Name: **Hyperband**
This is suggested when you have limited computational resources but have a relatively large search space. It performs well in scenarios where intermediate results can indicate good or bad final results to some extent. For example, when models that are more accurate early on in training are also more accurate later on. `Detailed Description <./HyperbandAdvisor.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="NetworkMorphism"></a>`
Network Morphism
^^^^^^^^^^^^^^^^
..
Built-in Tuner Name: **NetworkMorphism**
This is suggested when you want to apply deep learning methods to your task but you have no idea how to choose or design a network. You may modify this :githublink:`example <examples/trials/network_morphism/cifar10/cifar10_keras.py>` to fit your own dataset and your own data augmentation method. Also you can change the batch size, learning rate, or optimizer. Currently, this tuner only supports the computer vision domain. `Detailed Description <./NetworkmorphismTuner.rst>`__
:raw-html:`<br>`
:raw-html:`<a name="MetisTuner"></a>`
Metis Tuner
^^^^^^^^^^^
* - `TPE <./TpeTuner.rst>`__
- 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, as a black-box optimization, can be used in various scenarios and shows good performance in general. Especially when you have limited computation resources and can only try a small number of trials. From a large amount of experiments, we found that TPE is far better than Random Search.
Built-in Tuner Name: **MetisTuner**
* - `Random Search <./RandomTuner.rst>`__
- 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>`__
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 subsequent trials. Here is an :githublink:`example <examples/trials/auto-gbdt/search_space_metis.json>` on the use of Metis. Users only need to send the final result, such as ``accuracy``\ , to the tuner by calling the NNI SDK. `Detailed Description <./MetisTuner.rst>`__
Random search is suggested when each trial does not take very long (e.g., each trial can be completed very quickly, or early stopped by the assessor), and you have enough computational resources. It's also useful if you want to uniformly explore the searchspace. Random Search can be considered a baseline search algorithm.
Note that the only acceptable types of search space types are ``quniform``\ , ``uniform``\ , ``randint``\ , and numerical ``choice``. Only numerical values are supported since the values will be used to evaluate the 'distance' between different points.
* - `Anneal <./AnnealTuner.rst>`__
- 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.
:raw-html:`<br>`
Anneal is suggested when each trial does not take very long and you have enough computation resources (very similar to Random Search). It's also useful when the variables in the search space can be sample from some prior distribution.
:raw-html:`<a name="BOHB"></a>`
* - `Naïve Evolution <./EvolutionTuner.rst>`__
- 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 work, but it's very simple and easy to expand new features. `Reference paper <https://arxiv.org/pdf/1703.01041.pdf>`__
BOHB Advisor
^^^^^^^^^^^^
Its computational resource requirements are relatively high. Specifically, it requires a large initial population to avoid falling into a local optimum. If your trial is short or leverages assessor, this tuner is a good choice. It is also suggested when your trial code supports weight transfer; that is, the trial could inherit the converged weights from its parent(s). This can greatly speed up the training process.
..
* - `SMAC <./SmacTuner.rst>`__
- 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 needs to be installed by ``pip install nni[SMAC]`` command. `Reference Paper, <https://www.cs.ubc.ca/~hutter/papers/10-TR-SMAC.pdf>`__ `GitHub Repo <https://github.com/automl/SMAC3>`__
Built-in Tuner Name: **BOHB**
**Please note that SMAC doesn't support running on Windows currently**. For the specific reason, please refer to this `GitHub issue <https://github.com/automl/SMAC3/issues/483>`__.
Similar to Hyperband, BOHB is suggested when you have limited computational resources but have a relatively large search space. It performs well in scenarios where intermediate results can indicate good or bad final results to some extent. In this case, it may converge to a better configuration than Hyperband due to its usage of Bayesian optimization. `Detailed Description <./BohbAdvisor.rst>`__
Similar to TPE, SMAC is also a black-box tuner that can be tried in various scenarios and is suggested when computational resources are limited. It is optimized for discrete hyperparameters, thus, it's suggested when most of your hyperparameters are discrete.
:raw-html:`<a name="GPTuner"></a>`
* - `Batch tuner <./BatchTuner.rst>`__
- 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.
GP Tuner
^^^^^^^^
If the configurations you want to try have been decided beforehand, you can list them in search space file (using ``choice``) and run them using batch tuner.
`Detailed Description <./BatchTuner.rst>`__
..
* - `Grid Search <./GridsearchTuner.rst>`__
- Grid Search performs an exhaustive searching through the search space.
Built-in Tuner Name: **GPTuner**
This is suggested when the search space is small. It's suggested when it is feasible to exhaustively sweep the whole search space.
* - `Hyperband <./HyperbandAdvisor.rst>`__
- Hyperband tries to use limited resources to explore as many configurations as possible and returns the most promising ones as a final result. The basic idea is to generate many configurations and run them for a small number of trials. The half least-promising configurations are thrown out, the remaining are further trained along with a selection of new configurations. The size of these populations is sensitive to resource constraints (e.g. allotted search time). `Reference Paper <https://arxiv.org/pdf/1603.06560.pdf>`__
Note that the only acceptable types within the search space are ``randint``\ , ``uniform``\ , ``quniform``\ , ``loguniform``\ , ``qloguniform``\ , and numerical ``choice``. Only numerical values aresupported since the values will be used to evaluate the 'distance' between different points.
This is suggested when you have limited computational resources but have a relatively large search space. It performs well in scenarios where intermediate results can indicate good or bad final results to some extent. For example, when models that are more accurate early on in training are also more accurate later on.
As a strategy in a 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) to solve and common tools can be employed to solve it. Therefore, GP Tuner is most adequate for situations where the function to be optimized is very expensive to evaluate. GP can be used when computational resources are limited. However, 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.rst>`__
- Network Morphism provides functions to automatically search for deep learning architectures. It generates child networks that inherit the knowledge from their parent network which it is a morph from. This includes changes in depth, width, and skip-connections. Next, it estimates the value of a child network using historic architecture and metric pairs. Then it selects the most promising one to train. `Reference Paper <https://arxiv.org/abs/1806.10282>`__
:raw-html:`<a name="PBTTuner"></a>`
This is suggested when you want to apply deep learning methods to your task but you have no idea how to choose or design a network. You may modify this :githublink:`example <examples/trials/network_morphism/cifar10/cifar10_keras.py>` to fit your own dataset and your own data augmentation method. Also you can change the batch size, learning rate, or optimizer. Currently, this tuner only supports the computer vision domain.
PBT Tuner
^^^^^^^^^
* - `Metis Tuner <./MetisTuner.rst>`__
- 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/>`__
..
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 subsequent trials. Here is an :githublink:`example <examples/trials/auto-gbdt/search_space_metis.json>` on the use of Metis. Users only need to send the final result, such as ``accuracy``, to the tuner by calling the NNI SDK.
Built-in Tuner Name: **PBTTuner**
Note that the only acceptable types of search space types are ``quniform``, ``uniform``, ``randint``, and numerical ``choice``. Only numerical values are supported since the values will be used to evaluate the 'distance' between different points.
* - `BOHB <./BohbAdvisor.rst>`__
- BOHB is a follow-up work to 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>`__
**Suggested scenario**
Similar to Hyperband, BOHB is suggested when you have limited computational resources but have a relatively large search space. It performs well in scenarios where intermediate results can indicate good or bad final results to some extent. In this case, it may converge to a better configuration than Hyperband due to its usage of Bayesian optimization.
Population Based Training (PBT) bridges and extends parallel search methods and sequential optimization methods. It requires relatively small computation resource, by inheriting weights from currently good-performing ones to explore better ones periodically. With PBTTuner, users finally get a trained model, rather than a configuration that could reproduce the trained model by training the model from scratch. This is because model weights are inherited periodically through the whole search process. PBT can also be seen as a training approach. If you don't need to get a specific configuration, but just expect a good model, PBTTuner is a good choice. `See details <./PBTTuner.rst>`__
* - `GP Tuner <./GPTuner.rst>`__
- 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>`__
:raw-html:`<a name="DNGOTuner"></a>`
Note that the only acceptable types within the search space are ``randint``, ``uniform``, ``quniform``, ``loguniform``, ``qloguniform``, and numerical ``choice``. Only numerical values are supported since the values will be used to evaluate the 'distance' between different points.
DNGO Tuner
^^^^^^^^^^
As a strategy in a 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) to solve and common tools can be employed to solve it. Therefore, GP Tuner is most adequate for situations where the function to be optimized is very expensive to evaluate. GP can be used when computational resources are limited. However, 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.
..
* - `PBT Tuner <./PBTTuner.rst>`__
- PBT Tuner is a simple asynchronous optimization algorithm which effectively utilizes a fixed computational budget to jointly optimize a population of models and their hyperparameters to maximize performance. `Reference Paper <https://arxiv.org/abs/1711.09846v1>`__
Built-in Tuner Name: **DNGOTuner**
Population Based Training (PBT) bridges and extends parallel search methods and sequential optimization methods. It requires relatively small computation resource, by inheriting weights from currently good-performing ones to explore better ones periodically. With PBTTuner, users finally get a trained model, rather than a configuration that could reproduce the trained model by training the model from scratch. This is because model weights are inherited periodically through the whole search process. PBT can also be seen as a training approach. If you don't need to get a specific configuration, but just expect a good model, PBTTuner is a good choice.
Applicable to large scale hyperparameter optimization. Bayesian optimization that rapidly finds competitive models on benchmark object recognition tasks using convolutional networks, and image caption generation using neural language models. `See details <./DngoTuner.rst>`__
* - `DNGO Tuner <./DngoTuner.rst>`__
- Use of neural networks as an alternative to GPs to model distributions over functions in bayesian optimization.
Applicable to large scale hyperparameter optimization. Bayesian optimization that rapidly finds competitive models on benchmark object recognition tasks using convolutional networks, and image caption generation using neural language models.
**Reference and Feedback**
------------------------------
Usage of Built-in Tuners
------------------------
Using a built-in tuner provided by the NNI SDK requires one to declare the **name** and **classArgs** in the ``config.yml`` file.
Click tuners' name in above table to see their specification.
* To `report a bug <https://github.com/microsoft/nni/issues/new?template=bug-report.rst>`__ for this feature in GitHub;
* To `file a feature or improvement request <https://github.com/microsoft/nni/issues/new?template=enhancement.rst>`__ for this feature in GitHub;
* To know more about :githublink:`Feature Engineering with NNI <docs/en_US/FeatureEngineering/Overview.rst>`\ ;
* To know more about :githublink:`NAS with NNI <docs/en_US/NAS/Overview.rst>`\ ;
* To know more about :githublink:`Model Compression with NNI <docs/en_US/Compression/Overview.rst>`\ ;
Note: Some built-in tuners have dependencies that need to be installed using ``pip install nni[<tuner>]``, like SMAC's dependencies can be installed using ``pip install nni[SMAC]``.
@@ -36,7 +36,7 @@ Similar to tuner and assessor. NNI needs to locate your customized Advisor class
classArgs:
arg1: value1
**Note that** The working directory of your advisor is ``<home>/nni-experiments/<experiment_id>/log``\ , which can be retrieved with environment variable ``NNI_LOG_DIRECTORY``.
**Note that** The working directory of your advisor is ``<home>/nni-experiments/<experiment_id>/log``, which can be retrieved with environment variable ``NNI_LOG_DIRECTORY``.
NNI provides state-of-the-art tuning algorithm in builtin-tuners. NNI supports to build a tuner by yourself for tuning demand.
If you want to implement your own tuning algorithm, you can implement a customized Tuner, there are three things to do:
...
...
@@ -81,7 +78,7 @@ If the you implement the ``generate_parameters`` like this:
# your code implements here.
return {"dropout": 0.3, "learning_rate": 0.4}
It means your Tuner will always generate parameters ``{"dropout": 0.3, "learning_rate": 0.4}``. Then Trial will receive ``{"dropout": 0.3, "learning_rate": 0.4}`` by calling API ``nni.get_next_parameter()``. Once the trial ends with a result (normally some kind of metrics), it can send the result to Tuner by calling API ``nni.report_final_result()``\ , for example ``nni.report_final_result(0.93)``. Then your Tuner's ``receive_trial_result`` function will receied the result like：
It means your Tuner will always generate parameters ``{"dropout": 0.3, "learning_rate": 0.4}``. Then Trial will receive ``{"dropout": 0.3, "learning_rate": 0.4}`` by calling API ``nni.get_next_parameter()``. Once the trial ends with a result (normally some kind of metrics), it can send the result to Tuner by calling API ``nni.report_final_result()``, for example ``nni.report_final_result(0.93)``. Then your Tuner's ``receive_trial_result`` function will receied the result like：
.. code-block:: python
...
...
@@ -89,7 +86,7 @@ It means your Tuner will always generate parameters ``{"dropout": 0.3, "learning
**Note that** The working directory of your tuner is ``<home>/nni-experiments/<experiment_id>/log``\ , which can be retrieved with environment variable ``NNI_LOG_DIRECTORY``\ , therefore, if you want to access a file (e.g., ``data.txt``\ ) in the directory of your own tuner, you cannot use ``open('data.txt', 'r')``. Instead, you should use the following:
**Note that** The working directory of your tuner is ``<home>/nni-experiments/<experiment_id>/log``, which can be retrieved with environment variable ``NNI_LOG_DIRECTORY``, therefore, if you want to access a file (e.g., ``data.txt``) in the directory of your own tuner, you cannot use ``open('data.txt', 'r')``. Instead, you should use the following:
* **optimize_mode** (*'maximize' or 'minimize'*\ ) - If 'maximize', the tuner will target to maximize metrics. If 'minimize', the tuner will target to minimize metrics.
* **optimize_mode** (*'maximize' or 'minimize'*) - If 'maximize', the tuner will target to maximize metrics. If 'minimize', the tuner will target to minimize metrics.
* **sample_size** (*int, default = 1000*) - Number of samples to select in each iteration. The best one will be picked from the samples as the next trial.
* **trials_per_update** (*int, default = 20*) - Number of trials to collect before updating the model.
* **num_epochs_per_training** (*int, default = 500*) - Number of epochs to train DNGO model.
Naive Evolution comes from `Large-Scale Evolution of Image Classifiers <https://arxiv.org/pdf/1703.01041.pdf>`__. It randomly initializes a population based on the search space. For each generation, it chooses better ones and does some mutation (e.g., changes a hyperparameter, adds/removes one layer, etc.) on them to get the next generation. Naive Evolution requires many trials to works but it's very simple and it's easily expanded with new features.
...
...
@@ -14,10 +10,10 @@ classArgs Requirements
^^^^^^^^^^^^^^^^^^^^^^
*
**optimize_mode** (*maximize or minimize, optional, default = maximize*\ ) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
**optimize_mode** (*maximize or minimize, optional, default = maximize*) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
*
**population_size** (*int value (should > 0), optional, default = 20*\ ) - the initial size of the population (trial num) in the evolution tuner. It's suggested that ``population_size`` be much larger than ``concurrency`` so users can get the most out of the algorithm (and at least ``concurrency``\ , or the tuner will fail on its first generation of parameters).
**population_size** (*int value (should > 0), optional, default = 20*) - the initial size of the population (trial num) in the evolution tuner. It's suggested that ``population_size`` be much larger than ``concurrency`` so users can get the most out of the algorithm (and at least ``concurrency``, or the tuner will fail on its first generation of parameters).
Bayesian optimization works by constructing a posterior distribution of functions (a Gaussian Process) that best describes the function you want to optimize. As the number of observations grows, the posterior distribution improves, and the algorithm becomes more certain of which regions in parameter space are worth exploring and which are not.
GP Tuner is designed to minimize/maximize the number of steps required to find a combination of parameters that are close to the optimal combination. To do so, this method uses a proxy optimization problem (finding the maximum of the acquisition function) that, albeit still a hard problem, is cheaper (in the computational sense) to solve, and it's amenable to common tools. Therefore, Bayesian Optimization is suggested for situations where sampling the function to be optimized is very expensive.
Note that the only acceptable types within the search space are ``randint``\ , ``uniform``\ , ``quniform``\ , ``loguniform``\ , ``qloguniform``\ , and numerical ``choice``.
Note that the only acceptable types within the search space are ``randint``, ``uniform``, ``quniform``, ``loguniform``, ``qloguniform``, and numerical ``choice``.
This optimization approach is described in Section 3 of `Algorithms for Hyper-Parameter Optimization <https://papers.nips.cc/paper/4443-algorithms-for-hyper-parameter-optimization.pdf>`__.
...
...
@@ -18,15 +15,15 @@ Usage
classArgs requirements
^^^^^^^^^^^^^^^^^^^^^^
* **optimize_mode** (*'maximize' or 'minimize', optional, default = 'maximize'*\ ) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **utility** (*'ei', 'ucb' or 'poi', optional, default = 'ei'*\ ) - The utility function (acquisition function). 'ei', 'ucb', and 'poi' correspond to 'Expected Improvement', 'Upper Confidence Bound', and 'Probability of Improvement', respectively.
* **kappa** (*float, optional, default = 5*\ ) - Used by the 'ucb' utility function. The bigger ``kappa`` is, the more exploratory the tuner will be.
* **xi** (*float, optional, default = 0*\ ) - Used by the 'ei' and 'poi' utility functions. The bigger ``xi`` is, the more exploratory the tuner will be.
* **nu** (*float, optional, default = 2.5*\ ) - Used to specify the Matern kernel. The smaller nu, the less smooth the approximated function is.
* **alpha** (*float, optional, default = 1e-6*\ ) - Used to specify the Gaussian Process Regressor. Larger values correspond to an increased noise level in the observations.
* **cold_start_num** (*int, optional, default = 10*\ ) - Number of random explorations to perform before the 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 when getting the point which maximizes the acquisition function.
* **selection_num_starting_points** (*int, optional, default = 250*\ ) - Number of times to run L-BFGS-B from a random starting point after the warmup.
* **optimize_mode** (*'maximize' or 'minimize', optional, default = 'maximize'*) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **utility** (*'ei', 'ucb' or 'poi', optional, default = 'ei'*) - The utility function (acquisition function). 'ei', 'ucb', and 'poi' correspond to 'Expected Improvement', 'Upper Confidence Bound', and 'Probability of Improvement', respectively.
* **kappa** (*float, optional, default = 5*) - Used by the 'ucb' utility function. The bigger ``kappa`` is, the more exploratory the tuner will be.
* **xi** (*float, optional, default = 0*) - Used by the 'ei' and 'poi' utility functions. The bigger ``xi`` is, the more exploratory the tuner will be.
* **nu** (*float, optional, default = 2.5*) - Used to specify the Matern kernel. The smaller nu, the less smooth the approximated function is.
* **alpha** (*float, optional, default = 1e-6*) - Used to specify the Gaussian Process Regressor. Larger values correspond to an increased noise level in the observations.
* **cold_start_num** (*int, optional, default = 10*) - Number of random explorations to perform before the 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 when getting the point which maximizes the acquisition function.
* **selection_num_starting_points** (*int, optional, default = 250*) - Number of times to run L-BFGS-B from a random starting point after the warmup.
`Hyperband <https://arxiv.org/pdf/1603.06560.pdf>`__ is a popular autoML algorithm. The basic idea of Hyperband is to create several buckets, each having ``n`` randomly generated hyperparameter configurations, each configuration using ``r`` resources (e.g., epoch number, batch number). After the ``n`` configurations are finished, it chooses the top ``n/eta`` configurations and runs them using increased ``r*eta`` resources. At last, it chooses the best configuration it has found so far.
Or if you want to set ``exec_mode`` with ``serial`` according to the original algorithm. In this mode, the next bucket will start strictly after the current bucket.
``parallelism`` mode may lead to multiple unfinished buckets, and there is at most one unfinished bucket under ``serial`` mode. The advantage of ``parallelism`` mode is to make full use of resources, which may reduce the experiment duration multiple times. The following two pictures are the results of quick verification using `nas-bench-201 <../NAS/Benchmarks.rst>`__\ , picture above is in ``parallelism`` mode, picture below is in ``serial`` mode.
``parallelism`` mode may lead to multiple unfinished buckets, and there is at most one unfinished bucket under ``serial`` mode. The advantage of ``parallelism`` mode is to make full use of resources, which may reduce the experiment duration multiple times. The following two pictures are the results of quick verification using `nas-bench-201 <../NAS/Benchmarks.rst>`__, picture above is in ``parallelism`` mode, picture below is in ``serial`` mode.
.. image:: ../../img/hyperband_parallelism.png
...
...
@@ -54,15 +51,15 @@ To use Hyperband, you should add the following spec in your experiment's YAML co
#choice: serial, parallelism
exec_mode: parallelism
Note that once you use Advisor, you are not allowed to add a Tuner and Assessor spec in the config file. If you use Hyperband, among the hyperparameters (i.e., key-value pairs) received by a trial, there will be one more key called ``TRIAL_BUDGET`` defined by user. **By using this ``TRIAL_BUDGET``\ , the trial can control how long it runs**.
Note that once you use Advisor, you are not allowed to add a Tuner and Assessor spec in the config file. If you use Hyperband, among the hyperparameters (i.e., key-value pairs) received by a trial, there will be one more key called ``TRIAL_BUDGET`` defined by user. **By using this ``TRIAL_BUDGET``, the trial can control how long it runs**.
For ``report_intermediate_result(metric)`` and ``report_final_result(metric)`` in your trial code, **\ ``metric`` should be either a number or a dict which has a key ``default`` with a number as its value**. This number is the one you want to maximize or minimize, for example, accuracy or loss.
For ``report_intermediate_result(metric)`` and ``report_final_result(metric)`` in your trial code, **``metric`` should be either a number or a dict which has a key ``default`` with a number as its value**. This number is the one you want to maximize or minimize, for example, accuracy or loss.
``R`` and ``eta`` are the parameters of Hyperband that you can change. ``R`` means the maximum trial budget that can be allocated to a configuration. Here, trial budget could mean the number of epochs or mini-batches. This ``TRIAL_BUDGET`` should be used by the trial to control how long it runs. Refer to the example under ``examples/trials/mnist-advisor/`` for details.
``eta`` means ``n/eta`` configurations from ``n`` configurations will survive and rerun using more budgets.
Here is a concrete example of ``R=81`` and ``eta=3``\ :
Here is a concrete example of ``R=81`` and ``eta=3``:
.. list-table::
:header-rows: 1
...
...
@@ -120,10 +117,10 @@ classArgs requirements
^^^^^^^^^^^^^^^^^^^^^^
* **optimize_mode** (*maximize or minimize, optional, default = maximize*\ ) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **R** (*int, optional, default = 60*\ ) - the maximum budget given to a trial (could be the number of mini-batches or epochs). Each trial should use TRIAL_BUDGET to control how long they run.
* **eta** (*int, optional, default = 3*\ ) - ``(eta-1)/eta`` is the proportion of discarded trials.
* **exec_mode** (*serial or parallelism, optional, default = parallelism*\ ) - If 'parallelism', the tuner will try to use available resources to start new bucket immediately. If 'serial', the tuner will only start new bucket after the current bucket is done.
* **optimize_mode** (*maximize or minimize, optional, default = maximize*) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **R** (*int, optional, default = 60*) - the maximum budget given to a trial (could be the number of mini-batches or epochs). Each trial should use TRIAL_BUDGET to control how long they run.
* **eta** (*int, optional, default = 3*) - ``(eta-1)/eta`` is the proportion of discarded trials.
* **exec_mode** (*serial or parallelism, optional, default = parallelism*) - If 'parallelism', the tuner will try to use available resources to start new bucket immediately. If 'serial', the tuner will only start new bucket after the current bucket is done.
`Metis <https://www.microsoft.com/en-us/research/publication/metis-robustly-tuning-tail-latencies-cloud-systems/>`__ offers several benefits over other tuning algorithms. While most tools only predict the optimal configuration, Metis gives you two outputs, a prediction for the optimal configuration and a suggestion for the next trial. No more guess work!
...
...
@@ -19,7 +16,7 @@ Metis belongs to the class of sequential model-based optimization (SMBO) algorit
*
It identifies the next hyper-parameter candidate. This is achieved by inferring the potential information gain of exploration, exploitation, and resampling.
Note that the only acceptable types within the search space are ``quniform``\ , ``uniform``\ , ``randint``\ , and numerical ``choice``.
Note that the only acceptable types within the search space are ``quniform``, ``uniform``, ``randint``, and numerical ``choice``.
More details can be found in our `paper <https://www.microsoft.com/en-us/research/publication/metis-robustly-tuning-tail-latencies-cloud-systems/>`__.
...
...
@@ -29,7 +26,7 @@ Usage
classArgs requirements
^^^^^^^^^^^^^^^^^^^^^^
* **optimize_mode** (*'maximize' or 'minimize', optional, default = 'maximize'*\ ) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **optimize_mode** (*'maximize' or 'minimize', optional, default = 'maximize'*) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
Population Based Training (PBT) comes from `Population Based Training of Neural Networks <https://arxiv.org/abs/1711.09846v1>`__. It's a simple asynchronous optimization algorithm which effectively utilizes a fixed computational budget to jointly optimize a population of models and their hyperparameters to maximize performance. Importantly, PBT discovers a schedule of hyperparameter settings rather than following the generally sub-optimal strategy of trying to find a single fixed set to use for the whole course of training.
...
...
@@ -13,7 +9,7 @@ Population Based Training (PBT) comes from `Population Based Training of Neural
:alt:
PBTTuner initializes a population with several trials (i.e., ``population_size``\ ). There are four steps in the above figure, each trial only runs by one step. How long is one step is controlled by trial code, e.g., one epoch. When a trial starts, it loads a checkpoint specified by PBTTuner and continues to run one step, then saves checkpoint to a directory specified by PBTTuner and exits. The trials in a population run steps synchronously, that is, after all the trials finish the ``i``\ -th step, the ``(i+1)``\ -th step can be started. Exploitation and exploration of PBT are executed between two consecutive steps.
PBTTuner initializes a population with several trials (i.e., ``population_size``). There are four steps in the above figure, each trial only runs by one step. How long is one step is controlled by trial code, e.g., one epoch. When a trial starts, it loads a checkpoint specified by PBTTuner and continues to run one step, then saves checkpoint to a directory specified by PBTTuner and exits. The trials in a population run steps synchronously, that is, after all the trials finish the ``i``-th step, the ``(i+1)``-th step can be started. Exploitation and exploration of PBT are executed between two consecutive steps.
Usage
-----
...
...
@@ -21,7 +17,7 @@ Usage
Provide checkpoint directory
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Since some trials need to load other trial's checkpoint, users should provide a directory (i.e., ``all_checkpoint_dir``\ ) which is accessible by every trial. It is easy for local mode, users could directly use the default directory or specify any directory on the local machine. For other training services, users should follow `the document of those training services <../TrainingService/Overview.rst>`__ to provide a directory in a shared storage, such as NFS, Azure storage.
Since some trials need to load other trial's checkpoint, users should provide a directory (i.e., ``all_checkpoint_dir``) which is accessible by every trial. It is easy for local mode, users could directly use the default directory or specify any directory on the local machine. For other training services, users should follow `the document of those training services <../TrainingService/Overview.rst>`__ to provide a directory in a shared storage, such as NFS, Azure storage.
Modify your trial code
^^^^^^^^^^^^^^^^^^^^^^
...
...
@@ -47,11 +43,11 @@ The complete example code can be found :githublink:`here <examples/trials/mnist-
classArgs requirements
^^^^^^^^^^^^^^^^^^^^^^
* **optimize_mode** (*'maximize' or 'minimize'*\ ) - If 'maximize', the tuner will target to maximize metrics. If 'minimize', the tuner will target to minimize metrics.
* **all_checkpoint_dir** (*str, optional, default = None*\ ) - Directory for trials to load and save checkpoint, if not specified, the directory would be "~/nni/checkpoint/\ :raw-html:`<exp-id>`\ ". Note that if the experiment is not local mode, users should provide a path in a shared storage which can be accessed by all the trials.
* **population_size** (*int, optional, default = 10*\ ) - Number of trials in a population. Each step has this number of trials. In our implementation, one step is running each trial by specific training epochs set by users.
* **factors** (*tuple, optional, default = (1.2, 0.8)*\ ) - Factors for perturbation of hyperparameters.
* **fraction** (*float, optional, default = 0.2*\ ) - Fraction for selecting bottom and top trials.
* **optimize_mode** (*'maximize' or 'minimize'*) - If 'maximize', the tuner will target to maximize metrics. If 'minimize', the tuner will target to minimize metrics.
* **all_checkpoint_dir** (*str, optional, default = None*) - Directory for trials to load and save checkpoint, if not specified, the directory would be "~/nni/checkpoint/\ :raw-html:`<exp-id>`\ ". Note that if the experiment is not local mode, users should provide a path in a shared storage which can be accessed by all the trials.
* **population_size** (*int, optional, default = 10*) - Number of trials in a population. Each step has this number of trials. In our implementation, one step is running each trial by specific training epochs set by users.
* **factors** (*tuple, optional, default = (1.2, 0.8)*) - Factors for perturbation of hyperparameters.
* **fraction** (*float, optional, default = 0.2*) - Fraction for selecting bottom and top trials.
In `Random Search for Hyper-Parameter Optimization <http://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf>`__ we show that Random Search might be surprisingly effective despite its simplicity.
We suggest using Random Search as a baseline when no knowledge about the prior distribution of hyper-parameters is available.
`SMAC <https://www.cs.ubc.ca/~hutter/papers/10-TR-SMAC.pdf>`__ 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 <https://github.com/automl/SMAC3>`__.
Note that SMAC on nni only supports a subset of the types in the `search space spec <../Tutorial/SearchSpaceSpec.rst>`__\ : ``choice``\ , ``randint``\ , ``uniform``\ , ``loguniform``\ , and ``quniform``.
Note that SMAC on nni only supports a subset of the types in the `search space spec <../Tutorial/SearchSpaceSpec.rst>`__: ``choice``, ``randint``, ``uniform``, ``loguniform``, and ``quniform``.
Usage
-----
...
...
@@ -25,8 +20,8 @@ SMAC has dependencies that need to be installed by following command before the
classArgs requirements
^^^^^^^^^^^^^^^^^^^^^^
* **optimize_mode** (*maximize or minimize, optional, default = maximize*\ ) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **config_dedup** (*True or False, optional, default = False*\ ) - If True, the tuner will not generate a configuration that has been already generated. If False, a configuration may be generated twice, but it is rare for a relatively large search space.
* **optimize_mode** (*maximize or minimize, optional, default = maximize*) - If 'maximize', the tuner will try to maximize metrics. If 'minimize', the tuner will try to minimize metrics.
* **config_dedup** (*True or False, optional, default = False*) - If True, the tuner will not generate a configuration that has been already generated. If False, a configuration may be generated twice, but it is rare for a relatively large search space.
