Distributed XGBoost with Dask

Dask is a parallel computing library built on Python. Dask allows easy management of distributed workers and excels at handling large distributed data science workflows. The implementation in XGBoost originates from dask-xgboost with some extended functionalities and a different interface. Right now it is still under construction and may change (with proper warnings) in the future. The tutorial here focuses on basic usage of dask with CPU tree algorithms. For an overview of GPU based training and internal workings, see A New, Official Dask API for XGBoost.

Contents

Requirements

Dask can be installed using either pip or conda (see the dask installation documentation for more information). For accelerating XGBoost with GPUs, dask-cuda is recommended for creating GPU clusters.

Overview

A dask cluster consists of three different components: a centralized scheduler, one or more workers, and one or more clients which act as the user-facing entry point for submitting tasks to the cluster. When using XGBoost with dask, one needs to call the XGBoost dask interface from the client side. Below is a small example which illustrates basic usage of running XGBoost on a dask cluster:

import xgboost as xgb
import dask.distributed

cluster = dask.distributed.LocalCluster(n_workers=4, threads_per_worker=1)
client = dask.distributed.Client(cluster)

dtrain = xgb.dask.DaskDMatrix(client, X, y)  # X and y are dask dataframes or arrays

output = xgb.dask.train(client,
                        {'verbosity': 2,
                         'tree_method': 'hist'},
                        dtrain,
                        num_boost_round=4, evals=[(dtrain, 'train')])

Here we first create a cluster in single-node mode with dask.distributed.LocalCluster, then connect a dask.distributed.Client to this cluster, setting up an environment for later computation.

We then create a DMatrix object and pass it to train, along with some other parameters, much like XGBoost’s normal, non-dask interface. The primary difference with XGBoost’s dask interface is we pass our dask client as an additional argument for carrying out the computation. Note that if client is set to None, XGBoost will use the default client returned by dask.

There are two sets of APIs implemented in XGBoost. The first set is functional API illustrated in above example. Given the data and a set of parameters, the train function returns a model and the computation history as a Python dictionary:

{'booster': Booster,
 'history': dict}

For prediction, pass the output returned by train into xgb.dask.predict:

prediction = xgb.dask.predict(client, output, dtrain)

Or equivalently, pass output['booster']:

prediction = xgb.dask.predict(client, output['booster'], dtrain)

Here prediction is a dask Array object containing predictions from model.

Alternatively, XGBoost also implements the Scikit-Learn interface with DaskXGBClassifier and DaskXGBRegressor. See xgboost/demo/dask for more examples.

Threads

XGBoost has built in support for parallel computation through threads by the setting nthread parameter (n_jobs for scikit-learn). If these parameters are set, they will override the configuration in Dask. For example:

with dask.distributed.LocalCluster(n_workers=7, threads_per_worker=4) as cluster:

There are 4 threads allocated for each dask worker. Then by default XGBoost will use 4 threads in each process for both training and prediction. But if nthread parameter is set:

output = xgb.dask.train(client,
                        {'verbosity': 1,
                         'nthread': 8,
                         'tree_method': 'hist'},
                        dtrain,
                        num_boost_round=4, evals=[(dtrain, 'train')])

XGBoost will use 8 threads in each training process.

Working with asyncio

New in version 1.2.0.

XGBoost’s dask interface supports the new asyncio in Python and can be integrated into asynchronous workflows. For using dask with asynchronous operations, please refer to this dask example and document in distributed. To use XGBoost’s dask interface asynchronously, the client which is passed as an argument for training and prediction must be operating in asynchronous mode by specifying asynchronous=True when the client is created (example below). All functions (including DaskDMatrix) provided by the functional interface will then return coroutines which can then be awaited to retrieve their result.

Functional interface:

async with dask.distributed.Client(scheduler_address, asynchronous=True) as client:
    X, y = generate_array()
    m = await xgb.dask.DaskDMatrix(client, X, y)
    output = await xgb.dask.train(client, {}, dtrain=m)

    with_m = await xgb.dask.predict(client, output, m)
    with_X = await xgb.dask.predict(client, output, X)
    inplace = await xgb.dask.inplace_predict(client, output, X)

    # Use `client.compute` instead of the `compute` method from dask collection
    print(await client.compute(with_m))

While for the Scikit-Learn interface, trivial methods like set_params and accessing class attributes like evals_result_ do not require await. Other methods involving actual computation will return a coroutine and hence require awaiting:

async with dask.distributed.Client(scheduler_address, asynchronous=True) as client:
    X, y = generate_array()
    regressor = await xgb.dask.DaskXGBRegressor(verbosity=1, n_estimators=2)
    regressor.set_params(tree_method='hist')  # trivial method, synchronous operation
    regressor.client = client  #  accessing attribute, synchronous operation
    regressor = await regressor.fit(X, y, eval_set=[(X, y)])
    prediction = await regressor.predict(X)

    # Use `client.compute` instead of the `compute` method from dask collection
    print(await client.compute(prediction))

Be careful that XGBoost uses all the workers supplied by the client object. If you are training on GPU cluster and have 2 GPUs, the client object passed to XGBoost should return 2 workers.

Why is the initialization of DaskDMatrix so slow and throws weird errors

The dask API in XGBoost requires construction of DaskDMatrix. With the Scikit-Learn interface, DaskDMatrix is implicitly constructed for all input data during the fit or predict steps. You might have observed that DaskDMatrix construction can take large amounts of time, and sometimes throws errors that don’t seem to be relevant to DaskDMatrix. Here is a brief explanation for why. By default most dask computations are lazily evaluated, which means that computation is not carried out until you explicitly ask for a result by, for example, calling compute(). See the previous link for details in dask, and this wiki for information on the general concept of lazy evaluation. The DaskDMatrix constructor forces lazy computations to be evaluated, which means it’s where all your earlier computation actually being carried out, including operations like dd.read_csv(). To isolate the computation in DaskDMatrix from other lazy computations, one can explicitly wait for results of input data before constructing a DaskDMatrix. Also dask’s diagnostics dashboard can be used to monitor what operations are currently being performed.

Limitations

Basic functionality including model training and generating classification and regression predictions have been implemented. However, there are still some other limitations we haven’t addressed yet:

• Label encoding for the DaskXGBClassifier classifier may not be supported. So users need to encode their training labels into discrete values first.

• Ranking is not yet supported.

• Empty worker is not well supported by classifier. If the training hangs for classifier with a warning about empty DMatrix, please consider balancing your data first. But regressor works fine with empty DMatrix.

• Callback functions are not tested.

• Only GridSearchCV from Scikit-Learn is supported. Meaning that we can distribute data among workers but have to train one model at a time. If you want to scale up grid searching with model parallelism with Dask-ML, please consider using XGBoost’s non-dask Scikit-Learn interface, for example xgboost.XGBRegressor, for now.