File size: 11,685 Bytes
0af0679
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Welcome to the Second Lab - Week 1, Day 3\n",
    "\n",
    "Today we will work with lots of models! This is a way to get comfortable with APIs."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<table style=\"margin: 0; text-align: left; width:100%\">\n",
    "    <tr>\n",
    "        <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
    "            <img src=\"../assets/stop.png\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
    "        </td>\n",
    "        <td>\n",
    "            <h2 style=\"color:#ff7800;\">Important point - please read</h2>\n",
    "            <span style=\"color:#ff7800;\">The way I collaborate with you may be different to other courses you've taken. I prefer not to type code while you watch. Rather, I execute Jupyter Labs, like this, and give you an intuition for what's going on. My suggestion is that you carefully execute this yourself, <b>after</b> watching the lecture. Add print statements to understand what's going on, and then come up with your own variations.<br/><br/>If you have time, I'd love it if you submit a PR for changes in the community_contributions folder - instructions in the resources. Also, if you have a Github account, use this to showcase your variations. Not only is this essential practice, but it demonstrates your skills to others, including perhaps future clients or employers...\n",
    "            </span>\n",
    "        </td>\n",
    "    </tr>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This version adds Reflection pattern where we ask each model to critique and improve its own answer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Start with imports - ask ChatGPT to explain any package that you don't know\n",
    "\n",
    "import os\n",
    "import json\n",
    "from dotenv import load_dotenv\n",
    "from openai import OpenAI\n",
    "from anthropic import Anthropic\n",
    "from IPython.display import Markdown, display"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "request = \"Please come up with a challenging, nuanced question that I can ask a number of LLMs to evaluate their intelligence. \"\n",
    "request += \"Answer only with the question, no explanation.\"\n",
    "messages = [{\"role\": \"user\", \"content\": request}]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "messages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "competitors = []\n",
    "answers = []\n",
    "messages = [{\"role\": \"user\", \"content\": question}]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gemini = OpenAI(api_key=google_api_key, base_url=\"https://generativelanguage.googleapis.com/v1beta/openai/\")\n",
    "model_name = \"gemini-2.0-flash\"\n",
    "\n",
    "response = gemini.chat.completions.create(model=model_name, messages=messages)\n",
    "answer = response.choices[0].message.content\n",
    "\n",
    "display(Markdown(answer))\n",
    "competitors.append(model_name)\n",
    "answers.append(answer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "deepseek = OpenAI(api_key=deepseek_api_key, base_url=\"https://api.deepseek.com/v1\")\n",
    "model_name = \"deepseek-chat\"\n",
    "\n",
    "response = deepseek.chat.completions.create(model=model_name, messages=messages)\n",
    "answer = response.choices[0].message.content\n",
    "\n",
    "display(Markdown(answer))\n",
    "competitors.append(model_name)\n",
    "answers.append(answer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "groq = OpenAI(api_key=groq_api_key, base_url=\"https://api.groq.com/openai/v1\")\n",
    "model_name = \"llama-3.3-70b-versatile\"\n",
    "\n",
    "response = groq.chat.completions.create(model=model_name, messages=messages)\n",
    "answer = response.choices[0].message.content\n",
    "\n",
    "display(Markdown(answer))\n",
    "competitors.append(model_name)\n",
    "answers.append(answer)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<table style=\"margin: 0; text-align: left; width:100%\">\n",
    "    <tr>\n",
    "        <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
    "            <img src=\"../assets/stop.png\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
    "        </td>\n",
    "        <td>\n",
    "            <h2 style=\"color:#ff7800;\">Super important - ignore me at your peril!</h2>\n",
    "            <span style=\"color:#ff7800;\">The model called <b>llama3.3</b> is FAR too large for home computers - it's not intended for personal computing and will consume all your resources! Stick with the nicely sized <b>llama3.2</b> or <b>llama3.2:1b</b> and if you want larger, try llama3.1 or smaller variants of Qwen, Gemma, Phi or DeepSeek. See the <A href=\"https://ollama.com/models\">the Ollama models page</a> for a full list of models and sizes.\n",
    "            </span>\n",
    "        </td>\n",
    "    </tr>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!ollama pull llama3.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's bring this together - note the use of \"enumerate\"\n",
    "\n",
    "together = \"\"\n",
    "for index, answer in enumerate(answers):\n",
    "    together += f\"# Response from competitor {index+1}\\n\\n\"\n",
    "    together += answer + \"\\n\\n\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "judge = f\"\"\"You are judging a competition between {len(competitors)} competitors.\n",
    "Each model has been given this question:\n",
    "\n",
    "{question}\n",
    "\n",
    "Your job is to evaluate each response for clarity and strength of argument, and rank them in order of best to worst.\n",
    "Respond with JSON, and only JSON, with the following format:\n",
    "{{\"results\": [\"best competitor number\", \"second best competitor number\", \"third best competitor number\", ...]}}\n",
    "\n",
    "Here are the responses from each competitor:\n",
    "\n",
    "{together}\n",
    "\n",
    "Now respond with the JSON with the ranked order of the competitors, nothing else. Do not include markdown formatting or code blocks.\"\"\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "judge_messages = [{\"role\": \"user\", \"content\": judge}]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<table style=\"margin: 0; text-align: left; width:100%\">\n",
    "    <tr>\n",
    "        <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
    "            <img src=\"../assets/exercise.png\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
    "        </td>\n",
    "        <td>\n",
    "            <h2 style=\"color:#ff7800;\">Exercise</h2>\n",
    "            <span style=\"color:#ff7800;\">Which pattern(s) did this use? Try updating this to add another Agentic design pattern.\n",
    "            </span>\n",
    "        </td>\n",
    "    </tr>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Ensemble (Model Competition) Pattern\n",
    "Description: The same prompt/question is sent to multiple different LLMs (OpenAI, Anthropic, Ollama, etc.).\n",
    "Purpose: To compare the quality, style, and content of responses from different models.\n",
    "Where in notebook:\n",
    "The code sends the same question to several models and collects their answers in the competitors and answers lists.\n",
    "\n",
    "2. Judging/Evaluator Pattern\n",
    "Description: After collecting responses from all models, another LLM is used as a “judge” to evaluate and rank the responses.\n",
    "Purpose: To automate the assessment of which model gave the best answer, based on clarity and strength of argument.\n",
    "Where in notebook:\n",
    "The judge prompt is constructed, and an LLM is asked to rank the responses in JSON format.\n",
    "\n",
    "3. Self-Improvement/Meta-Reasoning Pattern\n",
    "Description: The system not only generates answers but also reflects on and evaluates its own outputs (or those of its peers).\n",
    "Purpose: To iteratively improve or select the best output, often used in advanced agentic systems.\n",
    "Where in notebook:\n",
    "The “judge” LLM is an example of meta-reasoning, as it reasons about the quality of other LLMs’ outputs.\n",
    "\n",
    "4. Chain-of-Thought/Decomposition Pattern (to a lesser extent)\n",
    "Description: Breaking down a complex task into subtasks (e.g., generate question → get answers → evaluate answers).\n",
    "Purpose: To improve reliability and interpretability by structuring the workflow.\n",
    "Where in notebook:\n",
    "The workflow is decomposed into:\n",
    "Generating a challenging question\n",
    "Getting answers from multiple models\n",
    "Judging the answers\n",
    "\n",
    "In short:\n",
    "This notebook uses the Ensemble/Competition, Judging/Evaluator, and Meta-Reasoning agentic patterns, and also demonstrates a simple form of Decomposition by structuring the workflow into clear stages.\n",
    "If you want to add more agentic patterns, you could try things like:\n",
    "Reflexion (let models critique and revise their own answers)\n",
    "Tool Use (let models call external tools or APIs)\n",
    "Planning (let a model plan the steps before answering)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<table style=\"margin: 0; text-align: left; width:100%\">\n",
    "    <tr>\n",
    "        <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
    "            <img src=\"../assets/business.png\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
    "        </td>\n",
    "        <td>\n",
    "            <h2 style=\"color:#00bfff;\">Commercial implications</h2>\n",
    "            <span style=\"color:#00bfff;\">These kinds of patterns - to send a task to multiple models, and evaluate results,\n",
    "            are common where you need to improve the quality of your LLM response. This approach can be universally applied\n",
    "            to business projects where accuracy is critical.\n",
    "            </span>\n",
    "        </td>\n",
    "    </tr>\n",
    "</table>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}