Merge branch 'main' of github.com:davidrzs/HackOptimize

src/translate_to_math.py

@@ -1,17 +1,51 @@
-# from ai21 import AI21Client
+from ai21 import AI21Client
+from ai21.logger import set_verbose
+from ai21.models.chat import ChatMessage, ToolMessage
+from ai21.models.chat.function_tool_definition import FunctionToolDefinition
+from ai21.models.chat.tool_defintions import ToolDefinition
+from ai21.models.chat.tool_parameters import ToolParameters
+import json, requests
 
-# messages = [ChatMessage(content="Who was the first emperor of rome", role="user")]
+client = AI21Client(api_key='kXftEvJYqotOSkpTw1e2KNUhvYX0EvU1')
 
-# client = AI21Client()
+def translate_to_math(input):
 
-# response = client.chat.completions.create(
-#   messages=messages,
-#   model="jamba-1.5-mini",
-#   stream=True
-# )
+    messages = [ChatMessage(content=
+                            """
+                            You are an Operations Research scientist. You are trying to solve a scheduling problem.
+                            You will be given a problem and you will need to translate it into mathematical notation.
+                            Your task is to translate the problem into an optimization problem.
 
-# for chunk in response:
-#   print(chunk.choices[0].delta.content, end="")
+                            The problem should be formulated with one the following objectives:
+                            1. Minimize time from start to finish of all required jobs
+                            2. Minimize resource consumed
+                            3. Minimize downtime between jobs
+                            4. Maximize an output like number of items produced or profit
 
-def translate_to_math(input):
-    return input
+
+                            Decisions variables can for example be:
+                            1. Start time for each job, machine pair
+                            2. Makespan completion
+                            3. Number of items produced by each machine
+
+                            Constraints can for example be:
+                            1. Each job must be done once
+                            2. Each machine can only do one job at a time
+                            3. Each job must be done in a certain order
+                            4. Each job must be done within a certain time frame
+                            5. Maximum number of jobs that can be done in a day
+                            6. There might be downtime required between jobs
+
+                            The problem should be formulated in the following format, :
+                            Objective function: Minimize or Maximize
+                            Decision variables: x1, x2, x3, ...
+                            Constraints: c1, c2, c3, ...
+                            """, role="system")]
+    messages.append(ChatMessage(content="This is the problem: " + input, role="user"))
+    response = client.chat.completions.create(
+      messages=messages,
+      model="jamba-1.5-large",
+      stream=True
+    )
+    for chunk in response:
+      print(chunk.choices[0].delta.content, end="")