Spaces:
Sleeping
Sleeping
update
Browse files- agent.py +19 -21
- app.py +8 -23
- app_legacy.py +218 -0
- app_playground.ipynb +113 -6
- requirements.txt +3 -2
- test.py +0 -1
agent.py
CHANGED
|
@@ -1,14 +1,17 @@
|
|
|
|
|
| 1 |
import os
|
| 2 |
from langchain.chat_models import init_chat_model
|
| 3 |
from langchain_core.messages import HumanMessage, SystemMessage, AIMessage, AnyMessage
|
| 4 |
-
from
|
|
|
|
| 5 |
|
| 6 |
from typing_extensions import TypedDict, Annotated
|
| 7 |
|
| 8 |
|
| 9 |
class State(TypedDict):
|
| 10 |
messages: Annotated[list, add_messages]
|
| 11 |
-
|
|
|
|
| 12 |
|
| 13 |
|
| 14 |
def get_llm():
|
|
@@ -20,33 +23,28 @@ def get_graph(llm):
|
|
| 20 |
with open('prompts/system_prompt.md', 'r', encoding='utf-8') as markdown_file:
|
| 21 |
system_prompt = markdown_file.read()
|
| 22 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
|
| 24 |
-
def
|
| 25 |
-
print("
|
|
|
|
|
|
|
| 26 |
|
| 27 |
-
messages
|
| 28 |
-
|
| 29 |
-
prompt: AnyMessage = [SystemMessage(content=system_prompt), messages]
|
| 30 |
-
|
| 31 |
-
response = llm.invoke([("system", prompt), messages])
|
| 32 |
-
|
| 33 |
-
return {"message": response}
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
from IPython.display import Image, display
|
| 39 |
-
|
| 40 |
-
from langgraph.graph import MessagesState, START, END, StateGraph
|
| 41 |
|
| 42 |
# Build graph
|
| 43 |
builder = StateGraph(State)
|
| 44 |
-
builder.add_node("
|
| 45 |
|
| 46 |
|
| 47 |
# Logic
|
| 48 |
-
builder.add_edge(START, "
|
| 49 |
-
builder.add_edge("
|
| 50 |
|
| 51 |
return builder.compile()
|
| 52 |
|
|
|
|
| 1 |
+
import operator
|
| 2 |
import os
|
| 3 |
from langchain.chat_models import init_chat_model
|
| 4 |
from langchain_core.messages import HumanMessage, SystemMessage, AIMessage, AnyMessage
|
| 5 |
+
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
|
| 6 |
+
from langgraph.graph import add_messages, START, END, StateGraph
|
| 7 |
|
| 8 |
from typing_extensions import TypedDict, Annotated
|
| 9 |
|
| 10 |
|
| 11 |
class State(TypedDict):
|
| 12 |
messages: Annotated[list, add_messages]
|
| 13 |
+
aggregate: Annotated[list, operator.add]
|
| 14 |
+
# graph_state: str
|
| 15 |
|
| 16 |
|
| 17 |
def get_llm():
|
|
|
|
| 23 |
with open('prompts/system_prompt.md', 'r', encoding='utf-8') as markdown_file:
|
| 24 |
system_prompt = markdown_file.read()
|
| 25 |
|
| 26 |
+
prompt_template = ChatPromptTemplate.from_messages(
|
| 27 |
+
[
|
| 28 |
+
("system", system_prompt),
|
| 29 |
+
MessagesPlaceholder(variable_name="messages"),
|
| 30 |
+
]
|
| 31 |
+
)
|
| 32 |
|
| 33 |
+
def call_model(state: State):
|
| 34 |
+
print("\n-------------------- Agent has been called -----------------------------------\n")
|
| 35 |
+
prompt = prompt_template.invoke(state["messages"])
|
| 36 |
+
response = llm.invoke(prompt)
|
| 37 |
|
| 38 |
+
return {"messages": [response], "aggregate": ["Agent"]}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
|
| 40 |
# Build graph
|
| 41 |
builder = StateGraph(State)
|
| 42 |
+
builder.add_node("Agent", call_model)
|
| 43 |
|
| 44 |
|
| 45 |
# Logic
|
| 46 |
+
builder.add_edge(START, "Agent")
|
| 47 |
+
builder.add_edge("Agent", END)
|
| 48 |
|
| 49 |
return builder.compile()
|
| 50 |
|
app.py
CHANGED
|
@@ -3,6 +3,9 @@ import gradio as gr
|
|
| 3 |
import requests
|
| 4 |
import inspect
|
| 5 |
import pandas as pd
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
# (Keep Constants as is)
|
| 8 |
# --- Constants ---
|
|
@@ -12,34 +15,16 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
|
|
| 12 |
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
|
| 13 |
class BasicAgent:
|
| 14 |
def __init__(self):
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
self.model = init_chat_model("llama-3.3-70b-versatile", model_provider="groq")
|
| 19 |
|
| 20 |
print("BasicAgent initialized.")
|
| 21 |
def __call__(self, question: str) -> str:
|
| 22 |
print(f"Agent received question (first 50 chars): {question[:50]}...")
|
| 23 |
-
|
| 24 |
-
# fixed_answer = r'{"task_id": "task_id_1", "model_answer": "Between 2000 and 2009 (inclusive), Mercedes Sosa published three studio albums: Corazón Libre (2005), Cantora 1 (2009), and Cantora 2 (2009).", "reasoning_trace": "The different steps by which your model reached answer 1"}{"task_id": "task_id_2", "model_answer": "Answer 2 from your model", "reasoning_trace": "The different steps by which your model reached answer 2"}'
|
| 25 |
-
#fixed_answer = "I need to find how many studio albums Mercedes Sosa published between 2000 and 2009, inclusive. From the provided list: 2005: Corazón Libre, 2009: Cantora 1 and 2009: Cantora 2. There are three albums within the specified range. FINAL ANSWER: 3"
|
| 26 |
-
#print(f"Agent returning fixed answer: {fixed_answer}")
|
| 27 |
-
fixed_answer = self.model.invoke([("system", """You are tasked with answering questions from the GAIA benchmark for AI agents.
|
| 28 |
-
|
| 29 |
-
Provide ONLY the precise answer to the question. Do not include explanations, reasoning, or any additional text. Be direct, specific, and concise to meet the strict exact-matching requirements of the GAIA benchmark.
|
| 30 |
-
|
| 31 |
-
# Output Format
|
| 32 |
-
|
| 33 |
-
- **Single-word or short-phrase answers:** If the question necessitates a brief answer, provide just that word or phrase.
|
| 34 |
-
- **Numerical values:** Provide only the number when applicable, with no additional formatting or units unless specifically requested.
|
| 35 |
-
- **Full sentences:** If the question expects a sentence, provide the exact sentence required with no extra characters, punctuation, or formatting.
|
| 36 |
-
|
| 37 |
-
# Notes
|
| 38 |
-
|
| 39 |
-
- Be aware of strict exact-matching requirements; even minor deviations can result in an incorrect response.
|
| 40 |
-
- If any ambiguity exists in the phrasing of the input, respond with an answer that aligns with the GAIA benchmark's intended interpretation."""), ("user", question)])
|
| 41 |
|
| 42 |
-
return
|
| 43 |
|
| 44 |
def run_and_submit_all( profile: gr.OAuthProfile | None):
|
| 45 |
"""
|
|
|
|
| 3 |
import requests
|
| 4 |
import inspect
|
| 5 |
import pandas as pd
|
| 6 |
+
from dotenv import load_dotenv
|
| 7 |
+
|
| 8 |
+
from agent import *
|
| 9 |
|
| 10 |
# (Keep Constants as is)
|
| 11 |
# --- Constants ---
|
|
|
|
| 15 |
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
|
| 16 |
class BasicAgent:
|
| 17 |
def __init__(self):
|
| 18 |
+
load_dotenv()
|
| 19 |
+
self.llm = get_llm()
|
| 20 |
+
self.graph = get_graph(self.llm)
|
|
|
|
| 21 |
|
| 22 |
print("BasicAgent initialized.")
|
| 23 |
def __call__(self, question: str) -> str:
|
| 24 |
print(f"Agent received question (first 50 chars): {question[:50]}...")
|
| 25 |
+
response = self.graph.invoke({"messages": [HumanMessage(content="question"),]})
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
|
| 27 |
+
return response["messages"][-1].content
|
| 28 |
|
| 29 |
def run_and_submit_all( profile: gr.OAuthProfile | None):
|
| 30 |
"""
|
app_legacy.py
ADDED
|
@@ -0,0 +1,218 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import gradio as gr
|
| 3 |
+
import requests
|
| 4 |
+
import inspect
|
| 5 |
+
import pandas as pd
|
| 6 |
+
|
| 7 |
+
# (Keep Constants as is)
|
| 8 |
+
# --- Constants ---
|
| 9 |
+
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
|
| 10 |
+
|
| 11 |
+
# --- Basic Agent Definition ---
|
| 12 |
+
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
|
| 13 |
+
class BasicAgent:
|
| 14 |
+
def __init__(self):
|
| 15 |
+
os.getenv("GROQ_API_KEY")
|
| 16 |
+
from langchain.chat_models import init_chat_model
|
| 17 |
+
|
| 18 |
+
self.model = init_chat_model("llama-3.3-70b-versatile", model_provider="groq")
|
| 19 |
+
|
| 20 |
+
print("BasicAgent initialized.")
|
| 21 |
+
def __call__(self, question: str) -> str:
|
| 22 |
+
print(f"Agent received question (first 50 chars): {question[:50]}...")
|
| 23 |
+
# fixed_answer = "This is a default answer."
|
| 24 |
+
# fixed_answer = r'{"task_id": "task_id_1", "model_answer": "Between 2000 and 2009 (inclusive), Mercedes Sosa published three studio albums: Corazón Libre (2005), Cantora 1 (2009), and Cantora 2 (2009).", "reasoning_trace": "The different steps by which your model reached answer 1"}{"task_id": "task_id_2", "model_answer": "Answer 2 from your model", "reasoning_trace": "The different steps by which your model reached answer 2"}'
|
| 25 |
+
#fixed_answer = "I need to find how many studio albums Mercedes Sosa published between 2000 and 2009, inclusive. From the provided list: 2005: Corazón Libre, 2009: Cantora 1 and 2009: Cantora 2. There are three albums within the specified range. FINAL ANSWER: 3"
|
| 26 |
+
#print(f"Agent returning fixed answer: {fixed_answer}")
|
| 27 |
+
fixed_answer = self.model.invoke([("system", """You are tasked with answering questions from the GAIA benchmark for AI agents.
|
| 28 |
+
|
| 29 |
+
Provide ONLY the precise answer to the question. Do not include explanations, reasoning, or any additional text. Be direct, specific, and concise to meet the strict exact-matching requirements of the GAIA benchmark.
|
| 30 |
+
|
| 31 |
+
# Output Format
|
| 32 |
+
|
| 33 |
+
- **Single-word or short-phrase answers:** If the question necessitates a brief answer, provide just that word or phrase.
|
| 34 |
+
- **Numerical values:** Provide only the number when applicable, with no additional formatting or units unless specifically requested.
|
| 35 |
+
- **Full sentences:** If the question expects a sentence, provide the exact sentence required with no extra characters, punctuation, or formatting.
|
| 36 |
+
|
| 37 |
+
# Notes
|
| 38 |
+
|
| 39 |
+
- Be aware of strict exact-matching requirements; even minor deviations can result in an incorrect response.
|
| 40 |
+
- If any ambiguity exists in the phrasing of the input, respond with an answer that aligns with the GAIA benchmark's intended interpretation."""), ("user", question)])
|
| 41 |
+
|
| 42 |
+
return fixed_answer.content
|
| 43 |
+
|
| 44 |
+
def run_and_submit_all( profile: gr.OAuthProfile | None):
|
| 45 |
+
"""
|
| 46 |
+
Fetches all questions, runs the BasicAgent on them, submits all answers,
|
| 47 |
+
and displays the results.
|
| 48 |
+
"""
|
| 49 |
+
# --- Determine HF Space Runtime URL and Repo URL ---
|
| 50 |
+
space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
|
| 51 |
+
|
| 52 |
+
if profile:
|
| 53 |
+
username= f"{profile.username}"
|
| 54 |
+
print(f"User logged in: {username}")
|
| 55 |
+
else:
|
| 56 |
+
print("User not logged in.")
|
| 57 |
+
return "Please Login to Hugging Face with the button.", None
|
| 58 |
+
|
| 59 |
+
api_url = DEFAULT_API_URL
|
| 60 |
+
questions_url = f"{api_url}/questions"
|
| 61 |
+
submit_url = f"{api_url}/submit"
|
| 62 |
+
|
| 63 |
+
# 1. Instantiate Agent ( modify this part to create your agent)
|
| 64 |
+
try:
|
| 65 |
+
agent = BasicAgent()
|
| 66 |
+
except Exception as e:
|
| 67 |
+
print(f"Error instantiating agent: {e}")
|
| 68 |
+
return f"Error initializing agent: {e}", None
|
| 69 |
+
# In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public)
|
| 70 |
+
agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
|
| 71 |
+
print(agent_code)
|
| 72 |
+
|
| 73 |
+
# 2. Fetch Questions
|
| 74 |
+
print(f"Fetching questions from: {questions_url}")
|
| 75 |
+
try:
|
| 76 |
+
response = requests.get(questions_url, timeout=15)
|
| 77 |
+
response.raise_for_status()
|
| 78 |
+
questions_data = response.json()
|
| 79 |
+
if not questions_data:
|
| 80 |
+
print("Fetched questions list is empty.")
|
| 81 |
+
return "Fetched questions list is empty or invalid format.", None
|
| 82 |
+
print(f"Fetched {len(questions_data)} questions.")
|
| 83 |
+
except requests.exceptions.RequestException as e:
|
| 84 |
+
print(f"Error fetching questions: {e}")
|
| 85 |
+
return f"Error fetching questions: {e}", None
|
| 86 |
+
except requests.exceptions.JSONDecodeError as e:
|
| 87 |
+
print(f"Error decoding JSON response from questions endpoint: {e}")
|
| 88 |
+
print(f"Response text: {response.text[:500]}")
|
| 89 |
+
return f"Error decoding server response for questions: {e}", None
|
| 90 |
+
except Exception as e:
|
| 91 |
+
print(f"An unexpected error occurred fetching questions: {e}")
|
| 92 |
+
return f"An unexpected error occurred fetching questions: {e}", None
|
| 93 |
+
|
| 94 |
+
# 3. Run your Agent
|
| 95 |
+
results_log = []
|
| 96 |
+
answers_payload = []
|
| 97 |
+
print(f"Running agent on {len(questions_data)} questions...")
|
| 98 |
+
for item in questions_data:
|
| 99 |
+
task_id = item.get("task_id")
|
| 100 |
+
question_text = item.get("question")
|
| 101 |
+
if not task_id or question_text is None:
|
| 102 |
+
print(f"Skipping item with missing task_id or question: {item}")
|
| 103 |
+
continue
|
| 104 |
+
try:
|
| 105 |
+
submitted_answer = agent(question_text)
|
| 106 |
+
answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
|
| 107 |
+
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
|
| 108 |
+
except Exception as e:
|
| 109 |
+
print(f"Error running agent on task {task_id}: {e}")
|
| 110 |
+
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
|
| 111 |
+
|
| 112 |
+
if not answers_payload:
|
| 113 |
+
print("Agent did not produce any answers to submit.")
|
| 114 |
+
return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
|
| 115 |
+
|
| 116 |
+
# 4. Prepare Submission
|
| 117 |
+
submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
|
| 118 |
+
status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
|
| 119 |
+
print(status_update)
|
| 120 |
+
|
| 121 |
+
# 5. Submit
|
| 122 |
+
print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
|
| 123 |
+
try:
|
| 124 |
+
response = requests.post(submit_url, json=submission_data, timeout=60)
|
| 125 |
+
response.raise_for_status()
|
| 126 |
+
result_data = response.json()
|
| 127 |
+
final_status = (
|
| 128 |
+
f"Submission Successful!\n"
|
| 129 |
+
f"User: {result_data.get('username')}\n"
|
| 130 |
+
f"Overall Score: {result_data.get('score', 'N/A')}% "
|
| 131 |
+
f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
|
| 132 |
+
f"Message: {result_data.get('message', 'No message received.')}"
|
| 133 |
+
)
|
| 134 |
+
print("Submission successful.")
|
| 135 |
+
results_df = pd.DataFrame(results_log)
|
| 136 |
+
return final_status, results_df
|
| 137 |
+
except requests.exceptions.HTTPError as e:
|
| 138 |
+
error_detail = f"Server responded with status {e.response.status_code}."
|
| 139 |
+
try:
|
| 140 |
+
error_json = e.response.json()
|
| 141 |
+
error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
|
| 142 |
+
except requests.exceptions.JSONDecodeError:
|
| 143 |
+
error_detail += f" Response: {e.response.text[:500]}"
|
| 144 |
+
status_message = f"Submission Failed: {error_detail}"
|
| 145 |
+
print(status_message)
|
| 146 |
+
results_df = pd.DataFrame(results_log)
|
| 147 |
+
return status_message, results_df
|
| 148 |
+
except requests.exceptions.Timeout:
|
| 149 |
+
status_message = "Submission Failed: The request timed out."
|
| 150 |
+
print(status_message)
|
| 151 |
+
results_df = pd.DataFrame(results_log)
|
| 152 |
+
return status_message, results_df
|
| 153 |
+
except requests.exceptions.RequestException as e:
|
| 154 |
+
status_message = f"Submission Failed: Network error - {e}"
|
| 155 |
+
print(status_message)
|
| 156 |
+
results_df = pd.DataFrame(results_log)
|
| 157 |
+
return status_message, results_df
|
| 158 |
+
except Exception as e:
|
| 159 |
+
status_message = f"An unexpected error occurred during submission: {e}"
|
| 160 |
+
print(status_message)
|
| 161 |
+
results_df = pd.DataFrame(results_log)
|
| 162 |
+
return status_message, results_df
|
| 163 |
+
|
| 164 |
+
|
| 165 |
+
# --- Build Gradio Interface using Blocks ---
|
| 166 |
+
with gr.Blocks() as demo:
|
| 167 |
+
gr.Markdown("# Basic Agent Evaluation Runner")
|
| 168 |
+
gr.Markdown(
|
| 169 |
+
"""
|
| 170 |
+
**Instructions:**
|
| 171 |
+
|
| 172 |
+
1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
|
| 173 |
+
2. Log in to your Hugging Face account using the button below. This uses your HF username for submission.
|
| 174 |
+
3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
|
| 175 |
+
|
| 176 |
+
---
|
| 177 |
+
**Disclaimers:**
|
| 178 |
+
Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
|
| 179 |
+
This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
|
| 180 |
+
"""
|
| 181 |
+
)
|
| 182 |
+
|
| 183 |
+
gr.LoginButton()
|
| 184 |
+
|
| 185 |
+
run_button = gr.Button("Run Evaluation & Submit All Answers")
|
| 186 |
+
|
| 187 |
+
status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
|
| 188 |
+
# Removed max_rows=10 from DataFrame constructor
|
| 189 |
+
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
|
| 190 |
+
|
| 191 |
+
run_button.click(
|
| 192 |
+
fn=run_and_submit_all,
|
| 193 |
+
outputs=[status_output, results_table]
|
| 194 |
+
)
|
| 195 |
+
|
| 196 |
+
if __name__ == "__main__":
|
| 197 |
+
print("\n" + "-"*30 + " App Starting " + "-"*30)
|
| 198 |
+
# Check for SPACE_HOST and SPACE_ID at startup for information
|
| 199 |
+
space_host_startup = os.getenv("SPACE_HOST")
|
| 200 |
+
space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
|
| 201 |
+
|
| 202 |
+
if space_host_startup:
|
| 203 |
+
print(f"✅ SPACE_HOST found: {space_host_startup}")
|
| 204 |
+
print(f" Runtime URL should be: https://{space_host_startup}.hf.space")
|
| 205 |
+
else:
|
| 206 |
+
print("ℹ️ SPACE_HOST environment variable not found (running locally?).")
|
| 207 |
+
|
| 208 |
+
if space_id_startup: # Print repo URLs if SPACE_ID is found
|
| 209 |
+
print(f"✅ SPACE_ID found: {space_id_startup}")
|
| 210 |
+
print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}")
|
| 211 |
+
print(f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main")
|
| 212 |
+
else:
|
| 213 |
+
print("ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.")
|
| 214 |
+
|
| 215 |
+
print("-"*(60 + len(" App Starting ")) + "\n")
|
| 216 |
+
|
| 217 |
+
print("Launching Gradio Interface for Basic Agent Evaluation...")
|
| 218 |
+
demo.launch(debug=True, share=False)
|
app_playground.ipynb
CHANGED
|
@@ -6,8 +6,8 @@
|
|
| 6 |
"metadata": {
|
| 7 |
"collapsed": true,
|
| 8 |
"ExecuteTime": {
|
| 9 |
-
"end_time": "2025-04-
|
| 10 |
-
"start_time": "2025-04-
|
| 11 |
}
|
| 12 |
},
|
| 13 |
"source": [
|
|
@@ -22,14 +22,121 @@
|
|
| 22 |
"\n",
|
| 23 |
"graph = get_graph(llm)\n"
|
| 24 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
"outputs": [],
|
| 26 |
-
"execution_count":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"metadata": {
|
| 30 |
"ExecuteTime": {
|
| 31 |
-
"end_time": "2025-04-
|
| 32 |
-
"start_time": "2025-04-
|
| 33 |
}
|
| 34 |
},
|
| 35 |
"cell_type": "code",
|
|
@@ -43,7 +150,7 @@
|
|
| 43 |
{
|
| 44 |
"data": {
|
| 45 |
"text/plain": [
|
| 46 |
-
"\"Hello
|
| 47 |
]
|
| 48 |
},
|
| 49 |
"execution_count": 3,
|
|
|
|
| 6 |
"metadata": {
|
| 7 |
"collapsed": true,
|
| 8 |
"ExecuteTime": {
|
| 9 |
+
"end_time": "2025-04-27T11:27:04.428213Z",
|
| 10 |
+
"start_time": "2025-04-27T11:26:58.244180Z"
|
| 11 |
}
|
| 12 |
},
|
| 13 |
"source": [
|
|
|
|
| 22 |
"\n",
|
| 23 |
"graph = get_graph(llm)\n"
|
| 24 |
],
|
| 25 |
+
"outputs": [
|
| 26 |
+
{
|
| 27 |
+
"name": "stderr",
|
| 28 |
+
"output_type": "stream",
|
| 29 |
+
"text": [
|
| 30 |
+
"C:\\Users\\dennis.binzen\\PycharmProjects\\Final_Assignment\\.venv\\Lib\\site-packages\\tqdm\\auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
|
| 31 |
+
" from .autonotebook import tqdm as notebook_tqdm\n"
|
| 32 |
+
]
|
| 33 |
+
}
|
| 34 |
+
],
|
| 35 |
+
"execution_count": 1
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"metadata": {
|
| 39 |
+
"ExecuteTime": {
|
| 40 |
+
"end_time": "2025-04-27T11:27:04.828108Z",
|
| 41 |
+
"start_time": "2025-04-27T11:27:04.434834Z"
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"cell_type": "code",
|
| 45 |
+
"source": [
|
| 46 |
+
"from IPython.display import Image, display\n",
|
| 47 |
+
"display(Image(graph.get_graph().draw_mermaid_png()))"
|
| 48 |
+
],
|
| 49 |
+
"id": "da8311fc7f1acf6",
|
| 50 |
+
"outputs": [
|
| 51 |
+
{
|
| 52 |
+
"data": {
|
| 53 |
+
"image/png": "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",
|
| 54 |
+
"text/plain": [
|
| 55 |
+
"<IPython.core.display.Image object>"
|
| 56 |
+
]
|
| 57 |
+
},
|
| 58 |
+
"metadata": {},
|
| 59 |
+
"output_type": "display_data"
|
| 60 |
+
}
|
| 61 |
+
],
|
| 62 |
+
"execution_count": 2
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"metadata": {
|
| 66 |
+
"ExecuteTime": {
|
| 67 |
+
"end_time": "2025-04-27T11:30:44.925757Z",
|
| 68 |
+
"start_time": "2025-04-27T11:30:44.688378Z"
|
| 69 |
+
}
|
| 70 |
+
},
|
| 71 |
+
"cell_type": "code",
|
| 72 |
+
"source": "res = graph.invoke({\"messages\": [HumanMessage(content=\"Hello, how are you?\"),]})",
|
| 73 |
+
"id": "a9fdfecc1af0975e",
|
| 74 |
+
"outputs": [
|
| 75 |
+
{
|
| 76 |
+
"name": "stdout",
|
| 77 |
+
"output_type": "stream",
|
| 78 |
+
"text": [
|
| 79 |
+
"\n",
|
| 80 |
+
"-------------------- Agent has been called -----------------------------------\n",
|
| 81 |
+
"\n"
|
| 82 |
+
]
|
| 83 |
+
}
|
| 84 |
+
],
|
| 85 |
+
"execution_count": 4
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"metadata": {
|
| 89 |
+
"ExecuteTime": {
|
| 90 |
+
"end_time": "2025-04-27T11:32:02.443950Z",
|
| 91 |
+
"start_time": "2025-04-27T11:32:02.438038Z"
|
| 92 |
+
}
|
| 93 |
+
},
|
| 94 |
+
"cell_type": "code",
|
| 95 |
+
"source": "res[\"messages\"][-1].content",
|
| 96 |
+
"id": "66f2dcaeb343c836",
|
| 97 |
+
"outputs": [
|
| 98 |
+
{
|
| 99 |
+
"data": {
|
| 100 |
+
"text/plain": [
|
| 101 |
+
"\"I'm doing well.\""
|
| 102 |
+
]
|
| 103 |
+
},
|
| 104 |
+
"execution_count": 7,
|
| 105 |
+
"metadata": {},
|
| 106 |
+
"output_type": "execute_result"
|
| 107 |
+
}
|
| 108 |
+
],
|
| 109 |
+
"execution_count": 7
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"metadata": {},
|
| 113 |
+
"cell_type": "code",
|
| 114 |
"outputs": [],
|
| 115 |
+
"execution_count": null,
|
| 116 |
+
"source": "",
|
| 117 |
+
"id": "8d3a94be44f4859a"
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"metadata": {},
|
| 121 |
+
"cell_type": "code",
|
| 122 |
+
"outputs": [],
|
| 123 |
+
"execution_count": null,
|
| 124 |
+
"source": "",
|
| 125 |
+
"id": "3bf392ddf3ce5c96"
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"metadata": {},
|
| 129 |
+
"cell_type": "code",
|
| 130 |
+
"outputs": [],
|
| 131 |
+
"execution_count": null,
|
| 132 |
+
"source": "",
|
| 133 |
+
"id": "abe6c705b7557bdc"
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"metadata": {
|
| 137 |
"ExecuteTime": {
|
| 138 |
+
"end_time": "2025-04-27T11:18:19.448376Z",
|
| 139 |
+
"start_time": "2025-04-27T11:18:19.025565Z"
|
| 140 |
}
|
| 141 |
},
|
| 142 |
"cell_type": "code",
|
|
|
|
| 150 |
{
|
| 151 |
"data": {
|
| 152 |
"text/plain": [
|
| 153 |
+
"\"Hello. I'm just a language model, so I don't have feelings or emotions like humans do, but I'm functioning properly and ready to assist you with any questions or topics you'd like to discuss. How can I help you today?\""
|
| 154 |
]
|
| 155 |
},
|
| 156 |
"execution_count": 3,
|
requirements.txt
CHANGED
|
@@ -5,5 +5,6 @@ langchain~=0.3.24
|
|
| 5 |
dotenv~=0.9.9
|
| 6 |
python-dotenv~=1.1.0
|
| 7 |
typing_extensions~=4.13.2
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
| 5 |
dotenv~=0.9.9
|
| 6 |
python-dotenv~=1.1.0
|
| 7 |
typing_extensions~=4.13.2
|
| 8 |
+
langgraph~=0.3.34
|
| 9 |
+
langchain-core~=0.3.56
|
| 10 |
+
ipython~=9.2.0
|
test.py
DELETED
|
@@ -1 +0,0 @@
|
|
| 1 |
-
print("Hello")
|
|
|
|
|
|