File size: 16,348 Bytes
10e9b7d
 
eccf8e4
9469c9b
3c4371f
9086500
10e9b7d
d59f015
e80aab9
3db6293
e80aab9
31243f4
d59f015
55126e6
 
31243f4
 
9be1ee4
bcc6dcf
 
55126e6
 
 
bcc6dcf
55126e6
 
 
 
 
 
bcc6dcf
55126e6
bcc6dcf
55126e6
ab8f825
 
 
55126e6
ab8f825
 
55126e6
ab8f825
 
 
 
 
55126e6
ab8f825
 
 
 
55126e6
ab8f825
 
55126e6
9086500
 
 
 
ab8f825
9086500
 
ab8f825
55126e6
ab8f825
9086500
ab8f825
55126e6
9086500
 
55126e6
ab8f825
 
55126e6
ab8f825
 
 
 
55126e6
bf6d10b
 
9469c9b
bf6d10b
55126e6
bf6d10b
 
55126e6
bf6d10b
 
 
 
c8d687d
9469c9b
bf6d10b
 
 
55126e6
9469c9b
 
 
 
 
55126e6
9469c9b
 
 
55126e6
9469c9b
 
 
55126e6
9469c9b
 
 
 
 
55126e6
9469c9b
 
 
55126e6
9469c9b
 
 
 
 
55126e6
9469c9b
55126e6
9469c9b
 
 
 
55126e6
9469c9b
 
 
 
55126e6
9469c9b
bf6d10b
9469c9b
55126e6
c8d687d
bf6d10b
 
55126e6
bf6d10b
 
55126e6
ab8f825
 
55126e6
ab8f825
 
 
 
 
 
55126e6
ab8f825
 
 
 
55126e6
9086500
ab8f825
 
 
55126e6
ab8f825
55126e6
ab8f825
 
55126e6
ab8f825
 
55126e6
9086500
 
 
 
ab8f825
9086500
 
ab8f825
55126e6
9086500
ab8f825
 
55126e6
ab8f825
 
 
 
 
 
bf6d10b
ab8f825
bf6d10b
ab8f825
 
7c9bc5f
55126e6
 
31243f4
 
 
 
7d65c66
55126e6
3c4371f
7e4a06b
55126e6
3c4371f
7e4a06b
3c4371f
7d65c66
3c4371f
7e4a06b
31243f4
 
e80aab9
b177367
31243f4
 
 
3c4371f
31243f4
b177367
36ed51a
c1fd3d2
3c4371f
bcc6dcf
 
7d65c66
31243f4
eccf8e4
31243f4
7d65c66
31243f4
 
55126e6
 
31243f4
e80aab9
31243f4
 
3c4371f
55126e6
 
 
7d65c66
31243f4
 
e80aab9
b177367
7d65c66
 
3c4371f
31243f4
 
 
 
 
 
 
7d65c66
 
 
31243f4
55126e6
 
31243f4
99fefaa
 
31243f4
3c4371f
31243f4
 
55126e6
7d65c66
3c4371f
31243f4
e80aab9
7d65c66
31243f4
e80aab9
7d65c66
e80aab9
 
31243f4
e80aab9
 
3c4371f
 
 
e80aab9
 
31243f4
 
e80aab9
3c4371f
e80aab9
 
3c4371f
e80aab9
7d65c66
3c4371f
31243f4
7d65c66
31243f4
3c4371f
 
 
 
 
e80aab9
31243f4
 
 
 
7d65c66
31243f4
 
 
 
e80aab9
 
 
 
55126e6
0ee0419
e514fd7
 
 
81917a3
e514fd7
 
 
 
 
 
 
 
e80aab9
 
7e4a06b
e80aab9
31243f4
e80aab9
9088b99
7d65c66
 
e80aab9
31243f4
 
 
e80aab9
 
 
55126e6
7d65c66
3c4371f
55126e6
7d65c66
3c4371f
 
7d65c66
3c4371f
7d65c66
 
55126e6
7d65c66
 
 
 
 
 
55126e6
3c4371f
31243f4
55126e6
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
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
import os
import gradio as gr
import requests
import inspect
import pandas as pd
from huggingface_hub import InferenceClient  # Import Hugging Face InferenceClient

# (Keep Constants as is)
# --- Constants ---
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"

# --- Basic Agent Definition ---
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------


class BasicAgent:
    def __init__(self):
        print("BasicAgent initialized.")
        # Use Qwen2.5-7B-Instruct model
        self.model_name = "Qwen/Qwen2.5-7B-Instruct"
        self.hf_token = os.getenv("HF_TOKEN")  # Get token from environment if available

        try:
            print(f"Initializing model: {self.model_name}")
            self.hf_client = InferenceClient(
                model=self.model_name,
                token=self.hf_token
            )
            print(f"Model initialized successfully: {self.model_name}")
        except Exception as e:
            print(f"Error initializing model ({self.model_name}): {e}")
            self.hf_client = None
            print("WARNING: Model initialization failed. Agent may not function properly.")

    def break_down_question(self, question: str) -> list:
        """
        Use an LLM to break down a complex question into key search terms or sub-questions.

        Args:
            question (str): The original question

        Returns:
            list: A list of key search terms or sub-questions
        """
        try:
            print(f"Breaking down question with LLM: {question[:50]}...")

            # Create a prompt that asks the LLM to break down the question
            prompt = f"""
            Please break down this question into 2-3 key search queries that would help find information to answer it.
            Return ONLY the search queries, one per line, with no additional text or explanations.

            Question: {question}
            """

            # Call the Hugging Face model to get the breakdown
            response = self.hf_client.text_generation(
                prompt=prompt,
                max_new_tokens=150,
                temperature=0.3,
                repetition_penalty=1.1,
                do_sample=True
            )

            # Extract the search terms from the response
            search_terms = response.strip().split('\n')
            search_terms = [term.strip() for term in search_terms if term.strip()]

            # Limit to 3 search terms maximum
            search_terms = search_terms[:3]

            print(f"Question broken down into {len(search_terms)} search terms: {search_terms}")
            return search_terms

        except Exception as e:
            print(f"Error breaking down question: {e}")
            # If there's an error, return the original question as a fallback
            return [question]

    def search_internet(self, query: str) -> str:
        """
        Search the internet for information using Wikipedia's API.
        This is a simple implementation that returns search results as text.

        Args:
            query (str): The search query

        Returns:
            str: Search results as text
        """
        print(f"Searching internet for: {query}")
        try:
            # Use Wikipedia API to search for information
            headers = {
                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
            }

            # Step 1: Search for relevant articles
            search_url = f"https://en.wikipedia.org/w/api.php?action=query&list=search&srsearch={query}&format=json"
            search_response = requests.get(search_url, headers=headers, timeout=10)
            search_response.raise_for_status()
            search_data = search_response.json()

            # Check if we found any search results
            if 'query' not in search_data or 'search' not in search_data['query'] or not search_data['query']['search']:
                return "No relevant information found."

            # Get the title of the first (most relevant) result
            first_result = search_data['query']['search'][0]
            page_title = first_result['title']

            # Step 2: Fetch the content of the most relevant article
            content_url = f"https://en.wikipedia.org/w/api.php?action=query&prop=extracts&exintro=1&explaintext=1&titles={page_title}&format=json"
            content_response = requests.get(content_url, headers=headers, timeout=10)
            content_response.raise_for_status()
            content_data = content_response.json()

            # Extract the page content
            pages = content_data['query']['pages']
            page_id = list(pages.keys())[0]

            if 'extract' in pages[page_id]:
                extract = pages[page_id]['extract']
                # Limit extract length to avoid very long responses
                if len(extract) > 1000:
                    extract = extract[:1000] + "..."

                result = f"Wikipedia article: {page_title}\n\n{extract}"

                # Also get a few more related article titles
                related_titles = []
                for item in search_data['query']['search'][1:4]:  # Get next 3 results
                    related_titles.append(item['title'])

                if related_titles:
                    result += "\n\nRelated topics:\n"
                    for title in related_titles:
                        result += f"- {title}\n"

                return result
            else:
                return "Found a relevant page, but couldn't extract its content."

        except Exception as e:
            print(f"Error searching internet: {e}")
            return f"Error performing internet search: {str(e)}"

    def __call__(self, question: str) -> str:
        print(f"Agent received question (first 50 chars): {question[:50]}...")

        # Use LLM to break down the question into key search terms
        search_terms = self.break_down_question(question)

        # Search for information using each search term
        all_results = []
        for term in search_terms:
            result = self.search_internet(term)
            if result and result != "No relevant information found." and not result.startswith("Error"):
                all_results.append(result)

        # Create a response based on collected search results
        if all_results:
            # Join the results with clear separation
            combined_results = "\n\n--- Next Search Result ---\n\n".join(all_results)

            # Use Hugging Face model to synthesize a coherent answer from the search results
            try:
                synthesis_prompt = f"""
                Based on the following search results, please provide a comprehensive answer to this question:

                Question: {question}

                Search Results:
                {combined_results}

                Answer:
                """

                # Call the Hugging Face model to synthesize an answer
                response = self.hf_client.text_generation(
                    prompt=synthesis_prompt,
                    max_new_tokens=500,
                    temperature=0.5,
                    repetition_penalty=1.05,
                    do_sample=True
                )

                answer = response.strip()
                print("Agent returning synthesized answer from search results.")
                return answer

            except Exception as e:
                print(f"Error synthesizing answer: {e}")
                # Fallback to returning the raw search results
                answer = f"Based on my searches, I found this information:\n\n{combined_results}"
                print("Agent returning raw search results due to synthesis error.")
                return answer
        else:
            # Fallback to default answer if all searches fail
            answer = "I couldn't find specific information about that question."
            print("Agent returning default answer as searches found no useful information.")
            return answer


def run_and_submit_all(profile: gr.OAuthProfile | None):
    """
    Fetches all questions, runs the BasicAgent on them, submits all answers,
    and displays the results.
    """
    # --- Determine HF Space Runtime URL and Repo URL ---
    space_id = os.getenv("SPACE_ID")  # Get the SPACE_ID for sending link to the code

    if profile:
        username = f"{profile.username}"
        print(f"User logged in: {username}")
    else:
        print("User not logged in.")
        return "Please Login to Hugging Face with the button.", None

    api_url = DEFAULT_API_URL
    questions_url = f"{api_url}/questions"
    submit_url = f"{api_url}/submit"

    # 1. Instantiate Agent ( modify this part to create your agent)
    try:
        agent = BasicAgent()
    except Exception as e:
        print(f"Error instantiating agent: {e}")
        return f"Error initializing agent: {e}", None
    # 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)
    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
    print(agent_code)

    return

    # 2. Fetch Questions
    print(f"Fetching questions from: {questions_url}")
    try:
        response = requests.get(questions_url, timeout=15)
        response.raise_for_status()
        questions_data = response.json()
        if not questions_data:
            print("Fetched questions list is empty.")
            return "Fetched questions list is empty or invalid format.", None
        print(f"Fetched {len(questions_data)} questions.")
    except requests.exceptions.RequestException as e:
        print(f"Error fetching questions: {e}")
        return f"Error fetching questions: {e}", None
    except requests.exceptions.JSONDecodeError as e:
        print(f"Error decoding JSON response from questions endpoint: {e}")
        print(f"Response text: {response.text[:500]}")
        return f"Error decoding server response for questions: {e}", None
    except Exception as e:
        print(f"An unexpected error occurred fetching questions: {e}")
        return f"An unexpected error occurred fetching questions: {e}", None

    # 3. Run your Agent
    results_log = []
    answers_payload = []
    print(f"Running agent on {len(questions_data)} questions...")
    for item in questions_data:
        task_id = item.get("task_id")
        question_text = item.get("question")
        if not task_id or question_text is None:
            print(f"Skipping item with missing task_id or question: {item}")
            continue
        try:
            submitted_answer = agent(question_text)
            answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
        except Exception as e:
            print(f"Error running agent on task {task_id}: {e}")
            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})

        return

    if not answers_payload:
        print("Agent did not produce any answers to submit.")
        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)

    # 4. Prepare Submission
    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
    print(status_update)

    # 5. Submit
    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
    try:
        response = requests.post(submit_url, json=submission_data, timeout=60)
        response.raise_for_status()
        result_data = response.json()
        final_status = (
            f"Submission Successful!\n"
            f"User: {result_data.get('username')}\n"
            f"Overall Score: {result_data.get('score', 'N/A')}% "
            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
            f"Message: {result_data.get('message', 'No message received.')}"
        )
        print("Submission successful.")
        results_df = pd.DataFrame(results_log)
        return final_status, results_df
    except requests.exceptions.HTTPError as e:
        error_detail = f"Server responded with status {e.response.status_code}."
        try:
            error_json = e.response.json()
            error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
        except requests.exceptions.JSONDecodeError:
            error_detail += f" Response: {e.response.text[:500]}"
        status_message = f"Submission Failed: {error_detail}"
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df
    except requests.exceptions.Timeout:
        status_message = "Submission Failed: The request timed out."
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df
    except requests.exceptions.RequestException as e:
        status_message = f"Submission Failed: Network error - {e}"
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df
    except Exception as e:
        status_message = f"An unexpected error occurred during submission: {e}"
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df


# --- Build Gradio Interface using Blocks ---
with gr.Blocks() as demo:
    gr.Markdown("# Basic Agent Evaluation Runner (Attempt #3)")
    gr.Markdown(
        """
        **Instructions:**

        1.  Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
        2.  Log in to your Hugging Face account using the button below. This uses your HF username for submission.
        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.

        ---
        **Disclaimers:**
        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).
        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.
        """
    )

    gr.LoginButton()

    run_button = gr.Button("Run Evaluation & Submit All Answers")

    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
    # Removed max_rows=10 from DataFrame constructor
    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)

    run_button.click(
        fn=run_and_submit_all,
        outputs=[status_output, results_table]
    )

if __name__ == "__main__":
    print("\n" + "-" * 30 + " App Starting " + "-" * 30)
    # Check for SPACE_HOST and SPACE_ID at startup for information
    space_host_startup = os.getenv("SPACE_HOST")
    space_id_startup = os.getenv("SPACE_ID")  # Get SPACE_ID at startup

    if space_host_startup:
        print(f"✅ SPACE_HOST found: {space_host_startup}")
        print(f"   Runtime URL should be: https://{space_host_startup}.hf.space")
    else:
        print("ℹ️  SPACE_HOST environment variable not found (running locally?).")

    if space_id_startup:  # Print repo URLs if SPACE_ID is found
        print(f"✅ SPACE_ID found: {space_id_startup}")
        print(f"   Repo URL: https://huggingface.co/spaces/{space_id_startup}")
        print(f"   Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main")
    else:
        print("ℹ️  SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.")

    print("-" * (60 + len(" App Starting ")) + "\n")

    print("Launching Gradio Interface for Basic Agent Evaluation...")
    demo.launch(debug=True, share=False)