Spaces:
Sleeping
Sleeping
File size: 2,375 Bytes
f762b8d | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 | # OpenEnv Specification for SQL Data Analyst Environment
# Hackathon: Meta x Scaler - OpenEnv Framework
name: sql_analyst
version: "1.0.0"
description: >
A Reinforcement Learning environment simulating a Data Analyst workspace
where an AI agent queries a SQLite database to answer business questions.
tags:
- openenv
- sql
- data-analyst
- reinforcement-learning
infrastructure:
vcpu: 2
memory: 8gb
timeout: 1200 # 20 minutes max runtime
entry_point: environment.env:SQLAnalystEnv
models:
action: environment.models:Action
observation: environment.models:Observation
reward: environment.models:Reward
schemas:
action:
type: object
properties:
sql_query:
type: string
description: SQL query to execute against the database
nullable: true
submit_answer:
type: string
description: Final answer to submit for grading
nullable: true
required: []
additionalProperties: false
observation:
type: object
properties:
schema_info:
type: string
description: Database schema information
current_question:
type: string
description: The current task question to answer
last_query_result:
type: string
description: Result from the last SQL query execution
error_message:
type: string
description: Error message from last action, if any
required:
- schema_info
- current_question
- last_query_result
- error_message
reward:
type: object
properties:
value:
type: number
description: Reward value for the action taken
required:
- value
endpoints:
reset:
method: POST
path: /reset
description: Reset the environment and get initial observation
response: observation
step:
method: POST
path: /step
description: Execute an action and receive observation, reward, done, info
request: action
response:
type: object
properties:
observation: observation
reward: reward
done:
type: boolean
info:
type: object
state:
method: GET
path: /state
description: Get the current internal state of the environment
|