fix(ray_tune): update imports in LHC sampler#310
Merged
michael-johnston merged 6 commits intomainfrom Dec 11, 2025
Merged
Conversation
Member
|
No vulnerabilities found. |
Also decrease log-level for example tests to reduce output.
AlessandroPomponio
changed the title
github-merge-queue
bot
removed this pull request from the merge queue due to no response for status checks
danielelotito
pushed a commit
that referenced
this pull request
Dec 15, 2025
* fix(ado_ray_tune): imports * test(examples): test examples in CI * fix(test): typo Also decrease log-level for example tests to reduce output. * fix(tests): setting loglevel * fix(tests): Can't set envvars before command * fix(tests): Use strings for log level
github-merge-queue bot
pushed a commit
that referenced
this pull request
Feb 11, 2026
* refactor(core): operation cleanup Operation cleanup logic was mixed together (a single function for cleaning up everything was required/was different for different scenarios) and also mixed with signal handling (a shutdown could be successful operation or a signal). In this commit - cleanup code required by different operations/functions is separated - A single signal handler exists that is generic - operations register their cleanup requirements with the handler so they are cleaned-up - shutdown global now only indicates if a signal was received NOT if an operation finished * chore(core): fix print formatting * feat(core): Raise exception when KeyboardInterrupt caught during operation It is raised as a new InterruptedOperationError which contains the operation identifier and is subclass of KeyboardInterrupt Previously nothing was raised and the operation exited normally after interruption. However, this pattern is not easy to maintain with multiple nested operations and each operation would have to check if the inner operation exited due to KeyboardInterrupt. This way operators do not have to handle KeyboardInterrupt. Each outer operation will catch the inner interrupt and raise a new exception with its own id. The outermost handler (in operation/create.py) now catches InterruptedOperationError and prints the id of the outermost (parent) interrupted operation. * Apply suggestions from code review Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * chore(core): apply suggestions from code review * feat: initial commit Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * fix(core): Setting shutdown signal to true * fix(core): Move initialize_ray_resource_cleaner Pattern is orchestrate() creates/destroys the resource cleaner which exists in default namespace. If another function calls an explore operation that function has same responsibility. * docs: improve logging Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * feat(wip): trigger model finalization when there are no entities left in the source space Note that there is a WIP on the random_shift strategy for high D sampling fix point in time to replicate comment #281 (comment) Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor(core): explore operation owns resource cleaner This is so the actors registered with the cleaner during the operation can be cleaned BEFORE the operation clean up deletes their parents (causing them to be deleted). The other option is to have a single resource cleaner but then all actors wanting to use it must be created "detached" * refactor(core): MeasurementQueue init - Remove get_measurement_queue class method. It does nothing anymore - Add ray_namespace parameter ray_namespace parameter allows less parallel passing of same information, which potentially is source of errors, and more consistency checks. A set of actuators, discovery space manager, and queue instance should all be in same ray namespace. * refactor(core): Move actuator initialization code - Move all actuator validation and initialization code to setup_actuator. - Create actuators first so validation failures are detected early - setup_actuators obtains namespace from queue rather than having potential inconsistency. * refactor(vllm_performance): Get resource cleaner via expected namespace * fix(core): missing parameter * fix(core): shutdown operation on exception Normal shutdown and SIGTERM were handled but other exception were not leading to operation resource being cleaned up * chore(core): fix log levels * Apply suggestions from code review Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * fix: operation return Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs(trim): first commit about examples with the trim operator Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * chore(core): Additional logging for debug * chore(core): Additional logging for debug * feat(raises ValueError): add batchsize logging The batchsize passed to the trim random walk is different from what random walk uses Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor(run_experiment): replace prints with console_prints (#289) * refactor(core): operation cleanup Operation cleanup logic was mixed together (a single function for cleaning up everything was required/was different for different scenarios) and also mixed with signal handling (a shutdown could be successful operation or a signal). In this commit - cleanup code required by different operations/functions is separated - A single signal handler exists that is generic - operations register their cleanup requirements with the handler so they are cleaned-up - shutdown global now only indicates if a signal was received NOT if an operation finished * refactor(core): operation cleanup Operation cleanup logic was mixed together (a single function for cleaning up everything was required/was different for different scenarios) and also mixed with signal handling (a shutdown could be successful operation or a signal). In this commit - cleanup code required by different operations/functions is separated - A single signal handler exists that is generic - operations register their cleanup requirements with the handler so they are cleaned-up - shutdown global now only indicates if a signal was received NOT if an operation finished * refactor(core): operation cleanup Operation cleanup logic was mixed together (a single function for cleaning up everything was required/was different for different scenarios) and also mixed with signal handling (a shutdown could be successful operation or a signal). In this commit - cleanup code required by different operations/functions is separated - A single signal handler exists that is generic - operations register their cleanup requirements with the handler so they are cleaned-up - shutdown global now only indicates if a signal was received NOT if an operation finished * build: update pre-commit hooks (#298) * build: update pre-commit hooks * style(docs): update feature request template * style(docs): update creating custom experiments * style(docs): update run_experiment * style(docs): update sft-trainer * style(docs): update vllm-performance * style(docs): update data-sharing * style(docs): update vllm-performance-full * style(docs): update ado * style(docs): update optimisation-with-ray-tune * style(docs): update random-walk * style(docs): update datacontainer * style(docs): update discovery-spaces * build: support Python 3.13 (#291) * build: enable python 3.13 * refactor(test): replace test actuatorconfiguration * refactor(test): replace test discoveryspace * test: support testing python 3.13 * build(sfttrainer): require python <3.13 * docs(sfttrainer): mention supported python versions * build: replace workspace construct * build: return most packages to workspace * fix(test): test_custom_experiments strikes again * fix(cli): do not use rich's Console.print with dataframes (#297) * docs(website): simplify cli examples (#293) * docs(website): remove unneeded field * docs(website): tidying * test(samplestore): Create separate ml_multicloud_sample_store.yaml for tests This removes the issue of users having to change the path to ml_export.csv when running the ml_multi_cloud examples. This then allows a single line command to run the example. * docs(example): Remove unneeded field * docs(example): single line version for vllm and random walk examples Also involved changing random walk example to follow vllm examples pattern * chore(docs): remove stray characters * test: fixes * docs(examples): Update to use two commands Also simplify "Scenario" * docs(website): Mention 3.13 is supported * chore(docs): apply review suggestions * Apply suggestions from code review Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> --------- Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> * docs(website): update instructions to build python wheels for ado and plugins (#301) * docs: instructions to build python wheels for ado and plugins Signed-off-by: Vassilis Vassiliadis <vassilis.vassiliadis@ibm.com> * Apply suggestion from @AlessandroPomponio Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Vassilis Vassiliadis <43679502+VassilisVassiliadis@users.noreply.github.com> * docs: instructions to build python wheels for ado and plugins Signed-off-by: Vassilis Vassiliadis <vassilis.vassiliadis@ibm.com> --------- Signed-off-by: Vassilis Vassiliadis <vassilis.vassiliadis@ibm.com> Signed-off-by: Vassilis Vassiliadis <43679502+VassilisVassiliadis@users.noreply.github.com> Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> * refactor(cli): improve sizing of live results table during operations (#299) * build(deps): update dependencies (#300) * build(deps): update dependencies * build(deps): update dependencies * build(autoconf): pin the required autogluon version (#304) * build(autoconf): fixing the autogluon version for autoconf Signed-off-by: SRIKUMAR VENUGOPAL <srikumarv@ie.ibm.com> * build(autoconf): fixing the autogluon version for autoconf Signed-off-by: SRIKUMAR VENUGOPAL <srikumarv@ie.ibm.com> --------- Signed-off-by: SRIKUMAR VENUGOPAL <srikumarv@ie.ibm.com> * fix(ray_tune): update imports in LHC sampler (#310) * fix(ado_ray_tune): imports * test(examples): test examples in CI * fix(test): typo Also decrease log-level for example tests to reduce output. * fix(tests): setting loglevel * fix(tests): Can't set envvars before command * fix(tests): Use strings for log level * feat(vllm_performance): add support for benchmarking geospatial models (#187) * feat: support geospatial benchmark - Add backend arg to execute benchmark - Add execute_geospatial_benchmark function This uses the geospatial_valencia.jsonl dataset * feat: add geospatial experiments endpoint and full * various fixes to the vllm_performance actuator Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix: add max_batch_tokens * Updated vllm performance actuator to support geospatial Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Termorarily avoiding cpu14 Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Termorarily avoiding cpu14 Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Added india dataset Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Fixed BaseSamplerConfig Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Some changes to the vllmperformance experiments Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Some changes to changes to the experiment and reverted the deployment template Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Removed some clutter from deployment template Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Few more fixes Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Fixed bug in validate_entitiy Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * One more fix to to a log message Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * One more fix to to a log message Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * One more fix to to a log message Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * Fixes to vllm_performance actuator Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix(experiment): Fixed bug in validate_entitiy Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix: Not using reference which may be parameterized * fix: validate_entity validate_entity was incorrectly identifying Entities with optional properties as invalid This was because validate_point_against_properties(allow_partial_matches=True) does not work as code expected. Code expected that given {point props} and {required props}, if {required props}.issubset(point_props) it would return True. However, it was checking the opposite i.e.{point_probs}.issubset{required_props} * fix: missing return * test: for validate_entity * chore: fixed formatting with black Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore: Removed dataset file as it was relocated to a different folder Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * feat: Added custom dataset geospatial experiment Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix: Reworked vllm_catalog actuator experiments catalog loading Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore: Improved experiment definition language and properties typing Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix: Improved logic for fetching a dataset in the geospatial benchmark Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore: various fixes around after review Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(performance_testing_geospatial): removing target properties that are irrelevant to geospatial tests Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(execute_benchmark): docstrings cleanup Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * feat(group_samplers): Improved GroupSampler groups building process Moved to reading the entities from storage only when are returned by the generathor rather than getching the whole space when initializing the sampler. Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix(group_samplers): Last fixes to make the group samplers tests pass Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Generic cleanup Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Applied snake casing wherever possible Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Addressed review comments Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Reverted wrong conversion to snake casing Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Simplified handling of points for creating groups Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Some leftover types and snake_cases Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(group_sampler): Tests cleanup Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(vllm_performance): removed custom entity_for_point in facvor of the new one from the discovery space Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(vllm_performance): removed unnecessary entity_identifier field from point dict Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix(vllm_performance): Fixed geospatial experiments after merge with master Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix(vllm_performance): Fixed yet another issue after merging with main Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * fix(chore): overall cleanup Signed-off-by: Christian Pinto <christian.pinto@ibm.com> * chore(vllm_performance): addressed review comments Signed-off-by: Christian Pinto <christian.pinto@ibm.com> --------- Signed-off-by: Christian Pinto <christian.pinto@ibm.com> Co-authored-by: michaelj <michaelj@ie.ibm.com> * build(deps): update dependencies (#311) * feat(ray_tune): support multi-objective optimization with optuna (#307) * feat(ado_ray_tune): support multi-objective opt with optuna Optuna can take multiple metrics/modes. However the code needed to be updated to handle when this is the case * docs(examples): multi-objective optimization * feat(ado_ray_tune): enable configuring optuna sampler * docs(website): optuna and multi-opt * Apply suggestions from code review Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * docs(website): Update multi-objective * chore(ado_ray_tune): rename function --------- Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * refactor(core): separate cleanup logic from signal handling and fix nested-operation shutdown (#281) * refactor(core): operation cleanup Operation cleanup logic was mixed together (a single function for cleaning up everything was required/was different for different scenarios) and also mixed with signal handling (a shutdown could be successful operation or a signal). In this commit - cleanup code required by different operations/functions is separated - A single signal handler exists that is generic - operations register their cleanup requirements with the handler so they are cleaned-up - shutdown global now only indicates if a signal was received NOT if an operation finished * chore(core): fix print formatting * feat(core): Raise exception when KeyboardInterrupt caught during operation It is raised as a new InterruptedOperationError which contains the operation identifier and is subclass of KeyboardInterrupt Previously nothing was raised and the operation exited normally after interruption. However, this pattern is not easy to maintain with multiple nested operations and each operation would have to check if the inner operation exited due to KeyboardInterrupt. This way operators do not have to handle KeyboardInterrupt. Each outer operation will catch the inner interrupt and raise a new exception with its own id. The outermost handler (in operation/create.py) now catches InterruptedOperationError and prints the id of the outermost (parent) interrupted operation. * Apply suggestions from code review Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * chore(core): apply suggestions from code review * fix(core): Setting shutdown signal to true * fix(core): Move initialize_ray_resource_cleaner Pattern is orchestrate() creates/destroys the resource cleaner which exists in default namespace. If another function calls an explore operation that function has same responsibility. * refactor(core): explore operation owns resource cleaner This is so the actors registered with the cleaner during the operation can be cleaned BEFORE the operation clean up deletes their parents (causing them to be deleted). The other option is to have a single resource cleaner but then all actors wanting to use it must be created "detached" * refactor(core): MeasurementQueue init - Remove get_measurement_queue class method. It does nothing anymore - Add ray_namespace parameter ray_namespace parameter allows less parallel passing of same information, which potentially is source of errors, and more consistency checks. A set of actuators, discovery space manager, and queue instance should all be in same ray namespace. * refactor(core): Move actuator initialization code - Move all actuator validation and initialization code to setup_actuator. - Create actuators first so validation failures are detected early - setup_actuators obtains namespace from queue rather than having potential inconsistency. * refactor(vllm_performance): Get resource cleaner via expected namespace * fix(core): missing parameter * Apply suggestions from code review Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * fix(core): shutdown operation on exception Normal shutdown and SIGTERM were handled but other exception were not leading to operation resource being cleaned up * chore(core): fix log levels * chore(core): Additional logging for debug * chore(core): Additional logging for debug * Apply suggestions from code review Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * chore(ado_ray_tune): black * chore(ado_ray_tune): code review suggestions --------- Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> * feat: access updated matching entities table Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor: logging module Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * feat(error): add ring structure for holdout set and declassi batch size variable in trim as the batch size of the random walk is always one Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * feat(wip): trim retrieves entity from discoveryspacemanager correctly the tests I am doing concerns now the stability of the iterative modeling process Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * test(data structure): add tests Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * fix: delete legacy code Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs: walkthrough example Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor(logs): improve logging logic Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * fix: add missing function Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * fix: func call Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor: put stopping criterion in the utils module Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor: delete unused func Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs: add comment Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs: improve example Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor: delete tests at the end of this file Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs: solve conflicts in logs and docstrings Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * feat: add sudoku and recursive aggregation sampling Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * refactor: remove comment Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * fix: ensure consistency of shifts eprmutations Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * feat: enable different sampling strategies in no priors characterization Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs: add example on how to run the operator Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs(log): handle better logging about mean column presence in the source space * fix(type): change type and default value of function used to retrieve source space points * docs(log): improve logging in get_source_and_target * refactor: improve logic * fix(delay_issue): set batchSIze to 1 Since randomwalk tries to perform continuous batching when it submits the first entity for measurement it immediately sends the second (trying to send concurrently up to five). This means it's very possible the sampler will look for the first entity results before it has actually been measured. This continues to happen whenever random walk sees the number of active experiments N=3 or less (as it tries to submit 5-N experiment concurrently to get back to 5 active again). If you want to do logic after each entity is measured then the batchSize passed to random walk (continuous batching size) must be 1. * build: update autogluon version * build: Update plugins/operators/trim/pyproject.toml This causes only trim and not the utils sub-package to be installed. Also the section is not necessary as by default it will find and install all packages if nothing is specified. follows: #329 (comment) * refactor(examples): rename examples/trim_custom_experiments -> examples/trim as it's an example of using trim, this also removes the double "trim_custom_experiments". * docs(refactor): update folder structure * refactor(example): move custom experiments config up one directory - Relocate config/ from custom experiments package to repo-level for clearer structure and reuse. - Updates import and path references accordingly. * docs(trim-examples): clarify repo root assumption and simplify commands - Add note that all commands assume execution from repository root. - Remove unnecessary `cd` steps by using full paths for config files. - Keep original context and links for custom experiment reference. * docs: move trim_operator README one level higher - Relocate README to operator’s parent directory to match repo structure * fix(logs): fix duplicate logs * fix(examples): move trim_experiments package into subfolder, install gets confused by config folder otherwise * fix(trim): circular import * chore(delay_issue): remove legacy code, made legacy by commit f8d5493 * style(ruff): make ruff happier * style(ruff): remove S101 from trim test path * refactor: change algorithm name * fix: removing 'dead' code * feat: add sobol sampling * feat(wip): add sobol sampling, plot if run as main and improve performance * style: improve readability * docs: improve comment * feat(minor): change default * style(ruff): automatic fix of type annotations * fix: signature * chore: remove not optimized version * chore: remove deprecated strategies * style: make ruff happier about type annotations, even though it still complains * style(ruff): add ann * refactor(chore): delete file to conform to repo structure * refactor: remove comments * docs: improve description * fix: set seed if none * docs: improve logging message * docs: change log level * docs: improve logging message * fix(stopping): use holdout score to evaluate stopping * docs: improve logging * docs: improve logging * docs(pydantic): improve logs and ensure coherence of sampling budget * docs: improve logs * fix(stopping): compute variance correctly * fix(finalization): add DiscoverySpaceManager to finalize function signature * docs(log): change to info level * docs(example): present trim to a broad public with an example * chore: remove unused pydantic fields * feat(stopping): connect enabled field of params.StoppingCriterion to the sampling logic * refactor: simplify code logic * docs: improve docstring * docs(wip): first draft of long docs * docs(refactor): improve formatting, add ref to autogluon * docs: simplify * style(ruff): make ruff happier about ANN * docs: improve docstrings * chore(typing): annotations * chore(logging): harmonize * Apply suggestions from code review Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> * build: change name of experiments folder * refactor: remove legacy function * docs: improve docstring * refactor: simplify entity space iteration using sequential point iterator apply mj's suggestion https://github.com/notifications#discussion_r2735895736 * test: add tests for concatenated_latin_hypercube_sampling in various scenarios * docs: update example config and default to avoid a warning about lightGBM failed import * feat: save dataset with prediction and model card * style: Update examples/trim/custom_experiments/trim_custom_experiments/experiments.py Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * style: avoid Any in typing Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: simplify the first lines, change trim to TRIM Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs(refactor): markdown linter error Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs(refactor): markdown linter error Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs(style): update trim->TRIM, formatting, removing "Declarative Configuration" section, remove a bit of self-glazing from "Reusable Artifacts" Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * style: apply new pre-commit hooks (copyright + type_checking block) Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> * docs(fix): improve docs, address #329 (comment) * docs(fix): path Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: improve paragraph trim->TRIM, formatting, rewrite first line, remove reference to the value of the outputDirectory field (already drifted - it's currently trim_models), rewrite second line to clarify that the model can be loaded as a TabularPredictor (first time it's mentioned) Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: enable linters Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: disable linter Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: format better Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: leave a blank line after the title of the message to avoid formatting issues down the line Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: format better Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: avoid the case where people think "wait, the identifier I got is different" Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: ap suggestion Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: trim -> TRIM Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: include the file to avoid drift Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: trim->TRIM, formatting, verb tense change Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: avoid a possibly confusing "no point present in the discovery space" by just removing that part. The next anchor correctly contextualises it and uses sample store. Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: good suggestion from ap Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: fix path Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: format better Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: format better Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * style: improve readability Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * refactor: avoid duplicated code Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: move to the top Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs(fix): undo unintended modification * style: logging function cannot raise errors * docs: address #329 (comment) * docs: clarify where final model is saved * docs: add 'why' * refactor: use clearer name of variables * build: expand python support * docs: fix typo Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * build: version is dynamic atm * style: apply #329 (comment) * refactor: improve name of variables * style: If the seed is not provided it's better to use the default RNG rather than set the seed to a hardcoded value * docs: improve function annotation * docs: improve function annotation * docs: raise error * style: address #329 (comment) * docs: address #329 (comment) * docs: load examples of configurations from preprocessor, apply #329 (comment) * docs: address #329 (comment) * refactor: adopt 'refresh', remove refs to DiscoverySpaceManager if not needed * docs: fix preprocessor Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: preprocessor fix Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs(refactor): rename files and remove duplicated information * fix(refactor): propagate refactoring dims->dimensions to this file * docs: refine config guide * docs: remove legacy ref * docs: refine readme * refactor: restore file as the new methods are no longer needed * refactor: invert if statement * refactor(log): create logger name and use match/case syntax * refactor: remove legacy * refactor: improve checks * fix(refactor): propagate refactoring dims->dimensions to this file * refactor: remove legacy code * refactor: remove legacy code * refactor: better handling of pydantic models splat * refactor: fix async programming * style: add expected argument metadata * refactor(build): uniform python version style * build: add trim * docs(website): make website available * feat(refactor): refactor TRIM logic so that you reuse code in the non remote entity iterator NOTE: I still need to test programmatically the entity iterator * docs: update link Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * test: change fixture * test: add trim to tox tests * docs: update link Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: update link Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: update link Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> * docs: update link Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> --------- Signed-off-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> Signed-off-by: Daniele Lotito <daniele.lotito@ibm.com> Signed-off-by: Vassilis Vassiliadis <vassilis.vassiliadis@ibm.com> Signed-off-by: Vassilis Vassiliadis <43679502+VassilisVassiliadis@users.noreply.github.com> Signed-off-by: SRIKUMAR VENUGOPAL <srikumarv@ie.ibm.com> Signed-off-by: Christian Pinto <christian.pinto@ibm.com> Signed-off-by: Daniele Lotito <99284466+danielelotito@users.noreply.github.com> Co-authored-by: michaelj <michaelj@ie.ibm.com> Co-authored-by: Michael Johnston <66301584+michael-johnston@users.noreply.github.com> Co-authored-by: Alessandro Pomponio <10339005+AlessandroPomponio@users.noreply.github.com> Co-authored-by: Vassilis Vassiliadis <43679502+VassilisVassiliadis@users.noreply.github.com> Co-authored-by: Srikumar Venugopal <srikumar003@users.noreply.github.com> Co-authored-by: Christian Pinto <christian.pinto@ibm.com> Co-authored-by: Alessandro Pomponio <alessandro.pomponio1@ibm.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fix incorrect import and also added test to tox that the LHC and RandomWalk examples can be executed to protect against this issue.