Revert to default inf max_trials and pool-size of 1

docs/src/tutorials/pytorch-mnist.rst

@@ -162,76 +162,3 @@ don't use ``--debug`` you will likely quickly fill your database with broken exp
 .. code-block:: bash
 
     $ orion --debug hunt -n orion-tutorial python main.py --lr~'loguniform(1e-5, 1.0)'
-
-Hunting Options
----------------
-
-.. code-block:: console
-
-    $ orion hunt --help
-
-    Oríon arguments (optional):
-      These arguments determine orion's behaviour
-
-      -n stringID, --name stringID
-                            experiment's unique name; (default: None - specified
-                            either here or in a config)
-      -u USER, --user USER  user associated to experiment's unique name; (default:
-                            $USER - can be overriden either here or in a config)
-      -c path-to-config, --config path-to-config
-                            user provided orion configuration file
-      --max-trials #        number of trials to be completed for the experiment.
-                            This value will be saved within the experiment
-                            configuration and reused across all workers to
-                            determine experiment's completion. (default: inf/until
-                            preempted)
-      --worker-trials #     number of trials to be completed for this worker. If
-                            the experiment is completed, the worker will die even
-                            if it did not reach its maximum number of trials
-                            (default: inf/until preempted)
-      --working-dir WORKING_DIR
-                            Set working directory for running experiment.
-      --pool-size #         number of simultaneous trials the algorithm should
-                            suggest. This is useful if many workers are executed
-                            in parallel and the algorithm has a strategy to sample
-                            non-independant trials simultaneously. Otherwise, it
-                            is better to leave `pool_size` to 1 and set a Strategy
-                            for Oríon's producer. Note that this option is not usefull useless you
-                            know the algorithm have a strategy to produce multiple trials
-                            simultaneously. If you have any doubt, leave it to 1.
-                            (default: 1)
-
-``name``
-
-The unique name of the experiment.
-
-``user``
-
-Username used to identify the experiments of a user. The default value is the system's username
-$USER.
-
-``config``
-
-Configuration file for Oríon which may define the database, the algorithm and all options of the
-command hunt, including ``name``, ``pool-size`` and ``max-trials``.
-
-``max-trials``
-
-The maximum number of trials tried during an experiment.
-
-``worker-trials``
-
-The maximum number of trials to be executed by a worker (a single call to ``orion hunt [...]``).
-
-``working-dir``
-
-The directory where configuration files are created. If not specified, Oríon will create a
-temporary directory that will be removed at end of execution of the trial.
-
-``pool-size``
-
-The number of trials which are generated by the algorithm each time it is interrogated. This is
-useful if many workers are executed in parallel and the algorithm has a strategy to sample
-non-independant trials simultaneously. Otherwise, it is better to leave ``pool_size`` to its default
-value 1. Note that this option is not usefull useless you know the algorithm have a strategy
-to produce multiple trials simultaneously. If you have any doubt, leave it to 1. :)

docs/src/user/config.rst

@@ -97,14 +97,14 @@ Full Example of Global Configuration
                 seed: None
         max_broken: 3
         max_trials: 1000000000
-        pool_size: 1
         strategy:
             MaxParallelStrategy
         worker_trials: 1000000000
         working_dir:
 
     worker:
         n_workers: 1
+        pool_size: 0
         executor: joblib
         executor_configuration: {}
         heartbeat: 120
@@ -211,7 +211,6 @@ Experiment
                 seed: None
         max_broken: 3
         max_trials: 1000000000
-        pool_size: 1
         strategy:
             MaxParallelStrategy
         worker_trials: 1000000000
@@ -322,22 +321,6 @@ working_dir
 
 
 
-.. _config_experiment_pool_size:
-
-pool_size
-~~~~~~~~~
-
-.. warning::
-
-   **DEPRECATED.** This argument will be removed in v0.3.
-
-:Type: int
-:Default: 1
-:Env var:
-:Description:
-    (DEPRECATED) This argument will be removed in v0.3.
-
-
 .. _config_experiment_algorithms:
 
 algorithms
@@ -376,6 +359,7 @@ Worker
 
     worker:
         n_workers: 1
+        pool_size: 0
         executor: joblib
         executor_configuration: {}
         heartbeat: 120
@@ -400,6 +384,20 @@ n_workers
     It is possible to run many `orion hunt` in parallel, and each will spawn
     ``n_workers``.
 
+.. _config_worker_pool_size:
+
+pool_size
+~~~~~~~~~
+
+:Type: int
+:Default: 0
+:Env var:
+:Description:
+    Number of trials to sample at a time. If 0, default to number of workers.
+    Increase it to improve the sampling speed if workers spend too much time
+    waiting for algorithms to sample points. An algorithm will try sampling `pool_size`
+    trials but may return less.
+
 
 .. _config_worker_executor:
 

src/orion/client/experiment.py

@@ -32,7 +32,7 @@
 log = logging.getLogger(__name__)
 
 
-def reserve_trial(experiment, producer, _depth=1):
+def reserve_trial(experiment, producer, pool_size, _depth=1):
     """Reserve a new trial, or produce and reserve a trial if none are available."""
     log.debug("Trying to reserve a new trial to evaluate.")
     trial = experiment.reserve_trial()
@@ -51,9 +51,9 @@ def reserve_trial(experiment, producer, _depth=1):
         producer.update()
 
         log.debug("#### Produce new trials.")
-        producer.produce()
+        producer.produce(pool_size)
 
-        return reserve_trial(experiment, producer, _depth=_depth + 1)
+        return reserve_trial(experiment, producer, pool_size, _depth=_depth + 1)
 
     return trial
 
@@ -500,7 +500,7 @@ def release(self, trial, status="interrupted"):
         finally:
             self._release_reservation(trial, raise_if_unreserved=raise_if_unreserved)
 
-    def suggest(self):
+    def suggest(self, pool_size=0):
         """Suggest a trial to execute.
 
         Experiment must be in executable ('x') mode.
@@ -509,6 +509,16 @@ def suggest(self):
         Otherwise, the algorithm is used to generate a new trial that is registered in storage and
         reserved.
 
+        Parameters
+        ----------
+        pool_size: int, optional
+            Number of trials to sample at a time. If 0, default to global config if defined,
+            else 1.  Increase it to improve the sampling speed if workers spend too much time
+            waiting for algorithms to sample points. An algorithm will try sampling `pool_size`
+            trials but may return less. Note: The method will still return only 1 trial even though
+            if the pool size is larger than 1. This is because atomic reservation of trials
+            can only be done one at a time.
+
         Returns
         -------
         `orior.core.worker.trial.Trial`
@@ -535,6 +545,10 @@ def suggest(self):
 
         """
         self._check_if_executable()
+        if not pool_size:
+            pool_size = orion.core.config.worker.pool_size
+        if not pool_size:
+            pool_size = 1
 
         if self.is_broken:
             raise BrokenExperiment("Trials failed too many times")
@@ -543,7 +557,7 @@ def suggest(self):
             raise CompletedExperiment("Experiment is done, cannot sample more trials.")
 
         try:
-            trial = reserve_trial(self._experiment, self._producer)
+            trial = reserve_trial(self._experiment, self._producer, pool_size)
 
         except (WaitingForTrials, SampleTimeout) as e:
             if self.is_broken:
@@ -629,10 +643,12 @@ def tmp_executor(self, executor, **config):
             yield self
         self.executor = old_executor
 
+    # pylint:disable=too-many-arguments
     def workon(
         self,
         fct,
         n_workers=None,
+        pool_size=0,
         max_trials=None,
         max_trials_per_worker=None,
         max_broken=None,
@@ -652,6 +668,11 @@ def workon(
             objective.
         n_workers: int, optional
             Number of workers to run in parallel. Defaults to value of global config.
+        pool_size: int, optional
+            Number of trials to sample at a time. If 0, defaults to `n_workers` or value of global
+            config if defined.  Increase it to improve the sampling speed if workers spend too much
+            time waiting for algorithms to sample points. An algorithm will try sampling
+            `pool_size` trials but may return less.
         max_trials: int, optional
             Maximum number of trials to execute within ``workon``. If the experiment or algorithm
             reach status is_done before, the execution of ``workon`` terminates.
@@ -712,6 +733,11 @@ def workon(
                     str(self.executor.n_workers),
                 )
 
+        if not pool_size:
+            pool_size = orion.core.config.worker.pool_size
+        if not pool_size:
+            pool_size = n_workers
+
         if max_trials is None:
             max_trials = self.max_trials
 
@@ -731,6 +757,7 @@ def workon(
             self.executor.submit(
                 self._optimize,
                 fct,
+                pool_size,
                 max_trials_per_worker,
                 max_broken,
                 trial_arg,
@@ -742,14 +769,16 @@ def workon(
 
         return sum(trials)
 
-    def _optimize(self, fct, max_trials, max_broken, trial_arg, on_error, **kwargs):
+    def _optimize(
+        self, fct, pool_size, max_trials, max_broken, trial_arg, on_error, **kwargs
+    ):
         worker_broken_trials = 0
         trials = 0
         kwargs = flatten(kwargs)
         max_trials = min(max_trials, self.max_trials)
         while not self.is_done and trials - worker_broken_trials < max_trials:
             try:
-                with self.suggest() as trial:
+                with self.suggest(pool_size=pool_size) as trial:
 
                     kwargs.update(flatten(trial.params))
 

src/orion/core/__init__.py

@@ -134,7 +134,7 @@ def define_experiment_config(config):
     experiment_config.add_option(
         "max_trials",
         option_type=int,
-        default=1000,
+        default=int(10e8),
         env_var="ORION_EXP_MAX_TRIALS",
         help="number of trials to be completed for the experiment. This value "
         "will be saved within the experiment configuration and reused "
@@ -144,7 +144,7 @@ def define_experiment_config(config):
     experiment_config.add_option(
         "worker_trials",
         option_type=int,
-        default=1000,
+        default=int(10e8),
         deprecate=dict(
             version="v0.3",
             alternative="worker.max_trials",
@@ -169,14 +169,6 @@ def define_experiment_config(config):
         help="Set working directory for running experiment.",
     )
 
-    experiment_config.add_option(
-        "pool_size",
-        option_type=int,
-        default=1,
-        deprecate=dict(version="v0.3", alternative=None, name="experiment.pool_size"),
-        help="This argument will be removed in v0.3.",
-    )
-
     experiment_config.add_option(
         "algorithms",
         option_type=dict,
@@ -210,6 +202,20 @@ def define_worker_config(config):
         ),
     )
 
+    worker_config.add_option(
+        "pool_size",
+        option_type=int,
+        default=0,
+        env_var="ORION_POOL_SIZE",
+        help=(
+            "Number of trials to sample at a time. "
+            "If 0, will default to number of executor workers. "
+            "Increase it to improve the sampling speed if workers spend too much time "
+            "waiting for algorithms to sample points. An algorithm will try sampling "
+            "`pool-size` trials but may return less."
+        ),
+    )
+
     worker_config.add_option(
         "executor",
         option_type=str,

src/orion/core/cli/hunt.py

@@ -130,6 +130,7 @@ def on_error(client, trial, error, worker_broken_trials):
 def workon(
     experiment,
     n_workers=None,
+    pool_size=None,
     max_trials=None,
     max_broken=None,
     max_idle_time=None,
@@ -163,6 +164,7 @@ def workon(
             client.workon(
                 consumer,
                 n_workers=n_workers,
+                pool_size=pool_size,
                 max_trials_per_worker=max_trials,
                 max_broken=max_broken,
                 trial_arg="trial",

src/orion/core/io/experiment_builder.py

@@ -379,11 +379,6 @@ def create_experiment(name, version, mode, space, **kwargs):
     """
     experiment = Experiment(name=name, version=version, mode=mode)
     experiment._id = kwargs.get("_id", None)  # pylint:disable=protected-access
-    experiment.pool_size = kwargs.get("pool_size")
-    if experiment.pool_size is None:
-        experiment.pool_size = orion.core.config.experiment.get(
-            "pool_size", deprecated="ignore"
-        )
     experiment.max_trials = kwargs.get(
         "max_trials", orion.core.config.experiment.max_trials
     )

src/orion/core/utils/format_terminal.py

@@ -257,7 +257,6 @@ def format_commandline(experiment):
 
 CONFIG_TEMPLATE = """\
 {title}
-pool size: {experiment.pool_size}
 max trials: {experiment.max_trials}
 max broken: {experiment.max_broken}
 working dir: {experiment.working_dir}