Step 2: Select which agent to run by modifying the following line in ingestion_program_2/ingestion.py:
For example:
from random_search_agent import Agent
All agents name can be found in sample_code_submission
As the results for different agents are stored in output/agent_name/ after Step 3, you should modify the following line in scoring_program_2/score.py:
For example:
default_output_dir = os.path.join(root_dir, "output/random_search_agent/")
NOTE: To collect the trajectory of an agent, you can check which action is selected at each step by the agent in the suggest()
function implemented in each agent.