Refactor microagent MarketFunctions to take market_id as input #66

evangriffiths · 2024-04-10T17:00:21Z

Changes to microagent to prevent the agent from having to pass the 'market question' around to various Functions

Changed GetMarkets to return market questions and IDs
Added GetMarketProbability for getting the market p_yes from the market id
BuyTokens Functions take in market_id instead of market question
Required a bump to PMAT 0.11.0
Also added an integration test where the agent runs GetMarketProbability

coderabbitai · 2024-04-10T17:00:32Z

Walkthrough

The updates primarily focus on enhancing type safety, renaming functions for clarity, and adjusting market-related functionalities across various modules. Changes include modifying type hints, updating function names and logic, and refining market handling and testing within the prediction market agent tooling.

Changes

Files	Change Summary
`.../known_outcome_agent/benchmark.py`, `.../known_outcome_agent/deploy.py`	Updated type hints and modified parameters for better clarity and functionality.
`.../microchain_agent/functions.py`, `.../microchain_agent/utils.py`	Renamed functions, updated type hints, and introduced new functionalities for market handling.
`pyproject.toml`	Upgraded dependency version to incorporate new features or fixes.
`tests/agents/microchain/test_agent.py`, `tests/agents/microchain/test_functions.py`	Added new test cases and modified existing ones to align with the updated functionalities and ensure robustness.

Possibly related issues

Deprecate use of Market class, replace with AgentMarket prediction-market-agent-tooling#42: The updates in known_outcome_agent/deploy.py and microchain_agent/utils.py involve changes to type hints and market class handling which align with the objectives of deprecating the Market class and transitioning to AgentMarket.
Investigate polymarket failure for tests/agents/test_microchain.py::test_get_markets #62: The new test cases added in tests/agents/microchain/test_agent.py and modifications in tests/agents/microchain/test_functions.py could help in identifying and resolving test failures related to market retrieval, addressing the issue's objectives.

Recent Review Details

Configuration used: CodeRabbit UI

Commits

Files that changed from the base of the PR and between 795cb62 and e79fa25.

Files selected for processing (1)

tests/agents/microchain/test_functions.py (4 hunks)

Files skipped from review as they are similar to previous changes (1)

tests/agents/microchain/test_functions.py

Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media?

Share

Tips

Chat

There are 3 ways to chat with CodeRabbit:

Review comments: Directly reply to a review comment made by CodeRabbit. Example:
- I pushed a fix in commit <commit_id>.
- Generate unit testing code for this file.
- Open a follow-up GitHub issue for this discussion.
Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
- @coderabbitai generate unit testing code for this file.
- @coderabbitai modularize this function.
PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
- @coderabbitai generate interesting stats about this repository and render them as a table.
- @coderabbitai show all the console.log statements in this repository.
- @coderabbitai read src/utils.ts and generate unit testing code.
- @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (invoked as PR comments)

@coderabbitai pause to pause the reviews on a PR.
@coderabbitai resume to resume the paused reviews.
@coderabbitai review to trigger a review. This is useful when automatic reviews are disabled for the repository.
@coderabbitai resolve resolve all the CodeRabbit review comments.
@coderabbitai help to get help.

Additionally, you can add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.

CodeRabbit Configration File (`.coderabbit.yaml`)

You can programmatically configure CodeRabbit by adding a .coderabbit.yaml file to the root of your repository.
Please see the configuration documentation for more information.
If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

Documentation and Community

Visit our Documentation for detailed information on how to use CodeRabbit.
Join our Discord Community to get help, request features, and share feedback.
Follow us on X/Twitter for updates and announcements.

evangriffiths · 2024-04-10T17:00:46Z

prediction_market_agent/agents/known_outcome_agent/benchmark.py

@@ -39,7 +39,7 @@ def to_market(self) -> AgentMarket:
            ),
            volume=None,
            created_time=None,
-            close_time=None,


Required change after bumping PMAT version

evangriffiths · 2024-04-10T17:00:53Z

prediction_market_agent/agents/known_outcome_agent/deploy.py

@@ -33,7 +34,7 @@ class DeployableKnownOutcomeAgent(DeployableAgent):
    def load(self) -> None:
        self.markets_with_known_outcomes: dict[str, Result] = {}

-    def pick_markets(self, markets: list[AgentMarket]) -> list[AgentMarket]:
+    def pick_markets(self, markets: t.Sequence[AgentMarket]) -> list[AgentMarket]:


Required change after bumping PMAT version

evangriffiths · 2024-04-10T17:01:31Z

prediction_market_agent/agents/microchain_agent/functions.py

@@ -63,11 +62,18 @@ def __init__(self, market_type: MarketType) -> None:
        self.market_type = market_type
        super().__init__()

+    @property


Added helper function to MarketFunction class, as it's needed in a few places

evangriffiths · 2024-04-10T17:02:29Z

prediction_market_agent/agents/microchain_agent/functions.py

+        ]
+
+
+class PredictPropabilityForQuestion(MarketFunction):


Renamed otherwise it could easily be mistaken for GetMarketProbability

I'm a bit confused - shouldn't this method be named something like PlaceBet or BuyTokensForOutcome? It's not really predicting anything, like the description says it allows the agent to purchase tokens.

@gabrielfior It's okay, just that Github hides the lines:

evangriffiths · 2024-04-10T17:03:22Z

prediction_market_agent/agents/microchain_agent/utils.py



 def get_binary_markets(market_type: MarketType) -> list[AgentMarket]:
    # Get the 5 markets that are closing soonest
    cls = market_type.market_class
-    markets: list[AgentMarket] = cls.get_binary_markets(


Required change after bumping PMAT version

evangriffiths · 2024-04-10T17:14:58Z

tests/agents/microchain/test_agent.py

+    engine.register(Stop())
+    engine.register(GetMarkets(market_type=market_type))
+    engine.register(GetMarketProbability(market_type=market_type))
+    engine.register(Jsonify())


Had to give the agent this function, otherwise it would just stop after calling GetMarketProbability, so the final output was:

>> GetMarketProbability(market_id="0x04f93bad589f989313bf3386f9dfa0f9354c9c25") ['0.43316548923547493'] Error: empty reply, aborting

gabrielfior · 2024-04-10T19:20:09Z

tests/agents/microchain/test_functions.py

+def test_get_probability(market_type: MarketType) -> None:
+    market_id = "0x0020d13c89140b47e10db54cbd53852b90bc1391"
+    get_market_probability = GetMarketProbability(market_type=market_type)
+    get_market_probability(market_id)


Is the assertion missing here?

Yes good point. I noticed that the way we calculate p_yes for closed omen markets is broken atm, so I made this task gnosis/prediction-market-agent-tooling#181

I've added an assertion here that will fail once that task is resolved, but I think that's fine.

gabrielfior · 2024-04-10T19:27:51Z

tests/agents/microchain/test_agent.py

+    m_json = json.loads(agent.history[-3]["content"])
+    m = MarketIDAndProbability.model_validate(m_json)
+    market: AgentMarket = market_type.market_class.get_binary_market(m.market_id)
+    assert market.p_yes == m.probability


Like this test a lot!
Just wondering - self.history makes sense (I don't see another way to fetch microchain output - one could use on_iteration_end and assign a variable, but don't think it would be much better).

This is a lot hackier though when compared with crewai (task.output.raw_content, task.output.json_output, etc) and autogen. So probably something to keep in mind when comparing those frameworks.

Yeah that's a good point. I thought the same

prediction_market_agent/agents/known_outcome_agent/benchmark.py

Refactor microagent MarketFunctions to take market_id as input

f5c9eca

evangriffiths commented Apr 10, 2024

View reviewed changes

Fix 'BuyTokens' Function description and example_args

d4af18f

evangriffiths commented Apr 10, 2024

View reviewed changes

evangriffiths requested review from gabrielfior and kongzii April 10, 2024 17:17

gabrielfior reviewed Apr 10, 2024

View reviewed changes

kongzii reviewed Apr 11, 2024

View reviewed changes

prediction_market_agent/agents/known_outcome_agent/benchmark.py Outdated Show resolved Hide resolved

evangriffiths added 2 commits April 11, 2024 09:56

Review comments

795cb62

fix assertion

e79fa25

kongzii approved these changes Apr 11, 2024

View reviewed changes

kongzii merged commit 284fe76 into main Apr 11, 2024
6 checks passed

kongzii deleted the evan/market-functions-id-input branch April 11, 2024 09:13

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Refactor microagent MarketFunctions to take market_id as input #66

Refactor microagent MarketFunctions to take market_id as input #66

evangriffiths commented Apr 10, 2024 •

edited

Loading

coderabbitai bot commented Apr 10, 2024 •

edited

Loading

Chat

CodeRabbit Commands (invoked as PR comments)

CodeRabbit Configration File (`.coderabbit.yaml`)

Documentation and Community

evangriffiths Apr 10, 2024

evangriffiths Apr 10, 2024

evangriffiths Apr 10, 2024

evangriffiths Apr 10, 2024

gabrielfior Apr 10, 2024

kongzii Apr 11, 2024

evangriffiths Apr 10, 2024

evangriffiths Apr 10, 2024

gabrielfior Apr 10, 2024

evangriffiths Apr 11, 2024

gabrielfior Apr 10, 2024

evangriffiths Apr 11, 2024

Refactor microagent MarketFunctions to take market_id as input #66

Refactor microagent MarketFunctions to take market_id as input #66

Conversation

evangriffiths commented Apr 10, 2024 • edited Loading

coderabbitai bot commented Apr 10, 2024 • edited Loading

Walkthrough

Changes

Possibly related issues

Chat

CodeRabbit Commands (invoked as PR comments)

CodeRabbit Configration File (.coderabbit.yaml)

Documentation and Community

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

evangriffiths commented Apr 10, 2024 •

edited

Loading

coderabbitai bot commented Apr 10, 2024 •

edited

Loading

CodeRabbit Configration File (`.coderabbit.yaml`)