bluesky / bluesky-adaptive

Reference implementation for tightly integrated adaptive scans.

Bluesky — An Experiment Specification & Orchestration Engine

Source	https://github.com/bluesky/bluesky
PyPI	pip install bluesky
Documentation	https://bluesky.github.io/bluesky
Releases	https://github.com/bluesky/bluesky/releases

Bluesky is a library for experiment control and collection of scientific data and metadata. It emphasizes the following virtues:

• Live, Streaming Data: Available for inline visualization and processing.
• Rich Metadata: Captured and organized to facilitate reproducibility and searchability.
• Experiment Generality: Seamlessly reuse a procedure on completely different hardware.
• Interruption Recovery: Experiments are "rewindable," recovering cleanly from interruptions.
• Automated Suspend/Resume: Experiments can be run unattended, automatically suspending and resuming if needed.
• Pluggable I/O: Export data (live) into any desired format or database.
• Customizability: Integrate custom experimental procedures and commands, and get the I/O and interruption features for free.
• Integration with Scientific Python: Interface naturally with numpy and Python scientific stack.

Bluesky Documentation.

The Bluesky Project enables experimental science at the lab-bench or facility scale. It is a collection of Python libraries that are co-developed but independently useful and may be adopted a la carte.

Bluesky Project Documentation.

See https://bluesky.github.io/bluesky for more detailed documentation.

bluesky-adaptive's Issues

Migrate from Travis-CI to GitHub Actions

See bluesky/bluesky-enhancement-proposals#21 for details.

Make option to disable IOC server

Expected Behavior

Current Behavior

Possible Solution

Steps to Reproduce (for bugs)

Context

Your Environment

ENH: K-d tree based adjudicator for redundant points in discretized space.

Expected Behavior

It is commonly needed to split space up by the resolution of the measurement/device. Binning is not a completely effective way to do this, as two neighboring points can be in different bins.

Current Behavior

Current redundant adjudicators are built off hashing approaches that will consider similar, but not identical, points in a space unique.

Possible Solution

Use https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.KDTree.html instead of hashing.

Assumption of sorting data for sklearn agents is too strong.

Decomposition general preforms stronger and appears more intuitive if the data is sorted. The current assumptions for DecompositionAgentBase
and [for ClusterAgentBase] will also only work for 1-d independent data.

Expected Behavior

• Work for n-d independent vars
• Potentially don't sort for clustering?

Current Behavior

Sorting higher dims causes value errors.

Possible Solution

partial solution for 2D that needs extension:

try:
    sorted_independents, sorted_observables = zip(
        *sorted(zip(self.independent_cache, self.observable_cache))
    )
except ValueError:
    # Multidimensional case
    sorted_independents, sorted_observables = zip(
        *sorted(zip(self.independent_cache, self.observable_cache), key=lambda x: (x[0][0], x[0][1]))
    )

Agent base class is out of date with Tiled

The abstract base classes for Agents utilize the deprecated Node feature of Tiled.

Expected Behavior

Expect adaptive agent base classes to be interoperable with current version of tiled and/or have tiled as an optional dependency to allow for 'closed loop' control flow experiments to utilize them.

Current Behavior

Current agent base class utilizes the deprecated Node object in tiled.

Proposed Solution

Update code to rely on the current Container structure instead.

Context

Currently building a bluesky-adatpive dependent framework to control ALS's MAESTRO beamline and cannot leverage current code without this change.

Restore intersphinx in docs config

#21 removed the interphinx extension because it was failing with miss-configured external references. The task is

1. restore using intersphinx
2. fix the URLs for any inventories we need

Tiled deprecation in type hints

tiled.client.node.Node should be changed to tiled.client.container.Container
https://github.com/bluesky/tiled/blob/ecc232b86a17f48509f89ed6fc94076bb6c55bd3/tiled/client/node.py#L3-L8

Current Behavior

    179 class Agent(ABC):
    180     """Abstract base class for a single plan agent. These agents should consume data, decide where to measure next,
    181     and execute a single type of plan (something akin to move and count).
    182     Alternatively, these agents can be used for soley reporting.
   (...)
    242         For example, this could be a beamline three letter acronym or other distinct key.
    243     """
    245     def __init__(
    246         self,
    247         *,
    248         kafka_consumer: AgentConsumer,
--> 249         tiled_data_node: tiled.client.node.Node,
    250         tiled_agent_node: tiled.client.node.Node,
    251         qserver: API_Threads_Mixin,
    252         kafka_producer: Optional[Publisher],
    253         agent_run_suffix: Optional[str] = None,
    254         metadata: Optional[dict] = None,
    255         ask_on_tell: Optional[bool] = True,
    256         direct_to_queue: Optional[bool] = True,
    257         report_on_tell: Optional[bool] = False,
    258         default_report_kwargs: Optional[dict] = None,
    259         queue_add_position: Optional[Union[int, Literal["front", "back"]]] = None,
    260         endstation_key: Optional[str] = "",
    261     ):
    262         logger.debug("Initializing agent.")
    263         self.kafka_consumer = kafka_consumer

AttributeError: module 'tiled.client' has no attribute 'node'

Possible Solution

Find and replace

Your Environment

2024-1.0-py310-tiled