Merged
Conversation
* ♻️ refactor(lots): updated absolute path loading from env; added suggested action storing >>> ⏰ 2h * ⚡️ perf(agent-files): use chain of thought only to decide an action * 🐛 fix(lots): fixed the suggested action storing; improved dashboard; played with call to actions >>> ⏰ 2h * 🧑💻 devxp(media_utils): added logging * 🐛 fix(lots): duplication fixes in scene.py;saving and metadata subsetting in prompts_and_responses >>> ⏰ 3h - structuring the _state that holds the previous actions as a list that prints nicely - removing the conditional of post/boost/reply around the duplicate check block, since the duplicate check should apply to all actionsthe app description that is printed now comes from the mastodon instructions in the config -the prompts_and_responses.jsonl is named and stored with the rest of the output * ✨ feat(dashboard-withthoughts.py;scene.py;media_utils.py): added agent&episode filtered thoughts to dashboard; updated termination and duplication prompts >>> ⏰ 5h * ♻️ refactor(main.py): simplified active player selection >>> ⏰ 1h * 🐛 fix(agent-files): adding clearer signposting to facilitate action selection * 🐛 fix(agent-files): fix phase change issue * ♻️ refactor(lots): abstracted agents using inheritance; debugged previous action flagposting >>> ⏰ 10h commented out model checkpointing. will bring back soon. * 🐛 fix(media_utils.py): fix writing to logs * 🐛 fix(agent-files): forgot to track agent files in last commmit. adding them here >>> ⏰ 5m * ✨ feat(dashboard-withthoughts.py): added heatmap for analyzing control over agent actions * ♻️ refactor(dashboard_withthoughts.py;-base_agent.py): N >>> ⏰ 1h more sizing/plot aesthetics in the dashboard; made component ordering code in base_agent.py more clear * 🐛 fix(base_agent.py): fixed bug with duplicate components * ♻️ refactor(agent-model-files): refactored agent component definitions for easy option setting >>> ⏰ 3h component are listed by their options in a generic way and options are set generically * 🎨 style(agent-files): fixed formatting of prompts >>> ⏰ 1h * 🐛 fix(apps.py,-like.py,-boost.py): stronger checks for duplicate likes/boosts * 🎨 style(gen_config.py-main.py): added model, sentence encoder as command-line/config arguments * 🎨 style(agent-files): updated formatting of context for action >>> ⏰ 2h --------- Co-authored-by: Xarangi <sneheelsarangi@gmail.com> Co-authored-by: Sneheel Sarangi <ss14247@nyu.edu>
…ut details from agent_speech_utils.py >>> ⏰ 2h
…nfo from bulk of dashboard code >>> ⏰ 1h
…type to agent file, separated game_runner to online_gamemaster
…o gen_config.py to make 4 config files; news agent treated like other agents >>> ⏰ 4h
… agent class >>> ⏰ 1h streamlining how agent role determines the agent model
…sim; leaving election specific code in examples/election >>> ⏰ 5h
…a module; updated readme >>> ⏰ 2h sim_setting now imported and added to system modules. I also updated the README to show the code structure.
…processed yaml; removed abs_path env var >>> ⏰ 2h also updated paths in stored image paths; other small changes
…ation structure, logic, and names >>> ⏰ will add hydra back later
…les are used for role-specific computation >>> ⏰ 4h
… implement loading >>> ⏰ 4h
added a fix to numpy and pandas compatibility
…ng >>> ⏰ 5h fixed bug (probes variable as DictConfig format doesn't print well so converted to dict). Will add hydra later but hardcoded similar )e.g. writing stdout; adding timestamps to output run folder)
removed compatibility fix since the project lock file has been updated
also updated concordia to most recent version
…dded hydra >>> ⏰ 3h
…s in input >>> ⏰ 4h
…tion plans; beautified dashboard; updated readme >>> ⏰ 5h
* 🐛 fix(readme): fixed image filename typo >>> ⏰ 5m * 🔥 dump(triggering.py): remove the unused triggering class >>> ⏰ 10m
* adding code and files for plotting * Adding main file for plotting * Rename plot.py to Plot.py --------- Co-authored-by: mptouzel <max.puelmatouzel@gmail.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.
No description provided.