Skip to content
Snippets Groups Projects
Select Git revision
  • 150_render_graph_d3js
  • 170_save_episode
  • 170_save_episode_b
  • 178-bug-in-step-function-reported-vetrov_andrew
  • 223-update-editor
  • 223_UpdateEditor_55_notebooks
  • 242-distance-map-performance-speed-up
  • 245-bug-distance-map-cacluation
  • 249-evaluator-bug-service-py-client-py
  • 249-evaluator-bug-service-py-client-py-2
  • 250-tests-for-agent-static
  • 263_malfunction_last_step
  • 264_fixing_examples
  • 264_revisited
  • 271-add-agent-index-to-global-agent-position-matrix
  • 273_malfunction_loading_and_saving
  • 308_fix
  • 323_unordered_cell_locking
  • 323_unordered_cell_locking_no_graphviz
  • 325_eval_step_timeout
  • generator_persistence_tag
  • v2.2.1
  • v2.2.0
  • v2.1.10
  • v2.1.9
  • v2.1.8
  • v2.1.7
  • v2.1.6
  • v2.1.5
  • v2.1.4
  • v2.1.3
  • v2.1.2
  • v2.1.1
  • v2.1.0
  • v2.0.0
  • v0.3.10
  • v0.3.9
  • v0.3.8
  • v0.3.7
  • v0.3.6
40 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.026Sep241697654328Aug27211817161514131029Jul282420191312987830Jun1213121110875432323231May24201918121110730Apr2524121013Mar42124Jan18131110763230Dec2928212017141364230Nov29282726252221201914131211876542131Oct302928272524232221181716141312111098tweaked the deadlock example in the agent-close-following notebookcount deadlocks; store deadlock status per agent in saved episodeBump version: 2.2.1 → 2.2.2Handling timeouts correctly for Round 2improved error handling in render-episode. assertion for mismatch.adding Test_9_Level_1.pkl as test env in render-episode notebook - this one has malfunctionsFixed typoClearer termination messagesAdded termination cause meta infoMerged from 323_unordered_cell_lockingBetter loggingCorrectly handling case when mean percentage done during a test gets too lowRound 2 modifications (infinite wave)Merge branch '323_unordered_cell_locking' of gitlab.aicrowd.com:flatland/flatland into 323_unordered_cell_lockingremoving notebook outputremoving explicit graphviz calls from UCF Agent Close Following notebookmerging from master into 323- branchflatland-animate change kernel to match Python 3.adding flatland-animate notebook. Simple animation.increase cell timeout from default 30s to 120s in run_all_notebooks.pychanged tox.ini to run notebooks from the tox envdir (.tox/notebooks) so they don't overwrite those in your local copy. Updated the notebooks to load env files from the package mechanism importlib_resources. This now gets them from the copy of env_data down in the tox python virtualenv.adding notebook-list, a list of notebooks to *include* in run-all-notebooks (already modified in last commit). This is so you can run tox without it choking on your local broken notebooks in development :)made the env loading more forgiving (in persistence) to allow the loading of old envs, and msgpack files with .pkl extension. Also give default 100 max_episode_steps where this is missing. Add test-saved-envs notebook to test loading all old envs.Merged from 323_unordered_cell_lockingRemoved pygraphviz dependency from requirements file323_unordered_c…323_unordered_cell_locking_no_graphvizInfinite wave: overall timeout, min percentage per Test, evaluation in order, doubled pre-planning/step time limitsignore absent pygraphviz - wrap render call (to_agraph) in try/except ImportError in agent_chainsAdditional service command line argumentsevaluator_loggingevaluator_loggingadding test episode file and a bit of a clean-up of render-episode.ipynbadding pygraphviz back into reqs_dev... sorryfixed typo imageio reqs_devadding render-episode notebookchanges prior to adding render_episode notebookLogging total number of malfunctions per episodePrint more details during evaluationremoving pygraphviz from reqs_dev - it's not used!merged from 323 UCF (again)fix typo in rail_env - remove left over m.patch from msgpackmanually merging Adrian's changes (made by Erik) from master - forgot onemanually merging Adrian's changes (made by Erik) from master