Bump to 0.24.0 (#5)

predictionprophet_deployment/agents/prophet_agent/deploy.py

@@ -6,7 +6,7 @@
 from prediction_market_agent_tooling.deploy.agent import (
     Answer,
     BetAmount,
-    DeployableAgent,
+    DeployableTraderAgent,
 )
 from prediction_market_agent_tooling.gtypes import Probability
 from prediction_market_agent_tooling.loggers import logger
@@ -38,7 +38,7 @@
 )
 
 
-class DeployableAgentER(DeployableAgent):
+class DeployableTraderAgentER(DeployableTraderAgent):
     agent: AbstractBenchmarkedAgent
     max_markets_per_run = 5
 
@@ -134,21 +134,21 @@ def answer_binary_market(self, market: AgentMarket) -> Answer | None:
         return prediciton.outcome_prediction
 
 
-class DeployablePredictionProphetGPT3Agent(DeployableAgentER):
+class DeployablePredictionProphetGPT3Agent(DeployableTraderAgentER):
     agent = PredictionProphetAgent(model="gpt-3.5-turbo-0125")
 
 
-class DeployablePredictionProphetGPT4TurboPreviewAgent(DeployableAgentER):
+class DeployablePredictionProphetGPT4TurboPreviewAgent(DeployableTraderAgentER):
     agent = PredictionProphetAgent(model="gpt-4-0125-preview")
     # Limit to just 1, because so far it seems that 20x higher costs aren't justified by the prediction performance.
     max_markets_per_run = 1
 
 
-class DeployablePredictionProphetGPT4TurboFinalAgent(DeployableAgentER):
+class DeployablePredictionProphetGPT4TurboFinalAgent(DeployableTraderAgentER):
     agent = PredictionProphetAgent(model="gpt-4-turbo-2024-04-09")
     # Limit to just 1, because so far it seems that 20x higher costs aren't justified by the prediction performance.
     max_markets_per_run = 1
 
 
-class DeployableOlasEmbeddingOAAgent(DeployableAgentER):
+class DeployableOlasEmbeddingOAAgent(DeployableTraderAgentER):
     agent = OlasAgent(model="gpt-3.5-turbo-0125", embedding_model=EmbeddingModel.openai)

predictionprophet_deployment/run_agent.py

@@ -1,6 +1,6 @@
 """
 Entrypoint for running the agent in GKE.
-If the agent adheres to PMAT standard (subclasses DeployableAgent), 
+If the agent adheres to PMAT standard (subclasses deployable agent), 
 simply add the agent to the `RunnableAgent` enum and then `RUNNABLE_AGENTS` dict.
 
 Can also be executed locally, simply by running `python predictionprophet_deployment/run_agent.py <agent> <market_type>`.

pyproject.toml

@@ -13,7 +13,7 @@ types-requests = "^2.31.0.20240106"
 black = "^23.12.1"
 autoflake = "^2.2.1"
 isort = "^5.13.2"
-prediction-prophet = {git = "https://github.com/polywrap/predictionprophet", rev = "d74a96456704dfc21cb049393123192facce91ac"}
+prediction-prophet = {git = "https://github.com/polywrap/predictionprophet", rev = "2eb5367848f1a7c17f73d0851748eaf03c008137"}
 
 [build-system]
 requires = ["poetry-core"]