brad/LLM_Engineering_OLD
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Updated explanations and tips

Browse Source
This commit is contained in:
Edward Donner
2024-11-18 16:57:26 -05:00
parent 7aeb18a3df
commit f4206653c3
13 changed files with 430 additions and 73 deletions
Download Patch File Download Diff File Expand all files Collapse all files

77
week2/day1.ipynb
View File

@@ -14,6 +14,31 @@
"Today we'll connect with the APIs for Anthropic and Google, as well as OpenAI."
]
},
{
"cell_type": "markdown",
"id": "2b268b6e-0ba4-461e-af86-74a41f4d681f",
"metadata": {},
"source": [
"<table style=\"margin: 0; text-align: left;\">\n",
" <tr>\n",
" <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
" <img src=\"../important.jpg\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
" </td>\n",
" <td>\n",
" <h2 style=\"color:#900;\">Important Note - Please read me</h2>\n",
" <span style=\"color:#900;\">I'm continually improving these labs, adding more examples and exercises.\n",
" At the start of each week, it's worth checking you have the latest code.<br/>\n",
" First do a <a href=\"https://chatgpt.com/share/6734e705-3270-8012-a074-421661af6ba9\">git pull and merge your changes as needed</a>. Any problems? Try asking ChatGPT to clarify how to merge - or contact me!<br/><br/>\n",
" After you've pulled the code, from the llm_engineering directory, in an Anaconda prompt (PC) or Terminal (Mac), run:<br/>\n",
" <code>conda env update --f environment.yml --prune</code><br/>\n",
" Or if you used virtualenv rather than Anaconda, then run this from your activated environment in a Powershell (PC) or Terminal (Mac):<br/>\n",
" <code>pip install -r requirements.txt</code>\n",
" </span>\n",
" </td>\n",
" </tr>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"id": "85cfe275-4705-4d30-abea-643fbddf1db0",
@@ -465,30 +490,64 @@
" claude_messages.append(claude_next)"
]
},
{
"cell_type": "markdown",
"id": "1d10e705-db48-4290-9dc8-9efdb4e31323",
"metadata": {},
"source": [
"<table style=\"margin: 0; text-align: left;\">\n",
" <tr>\n",
" <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
" <img src=\"../important.jpg\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
" </td>\n",
" <td>\n",
" <h2 style=\"color:#900;\">Before you continue</h2>\n",
" <span style=\"color:#900;\">\n",
" Be sure you understand how the conversation above is working, and in particular how the <code>messages</code> list is being populated. Add print statements as needed. Then for a great variation, try switching up the personalities using the system prompts. Perhaps one can be pessimistic, and one optimistic?<br/>\n",
" </span>\n",
" </td>\n",
" </tr>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"id": "3637910d-2c6f-4f19-b1fb-2f916d23f9ac",
"metadata": {},
"source": [
"# See the community-contributions folder\n",
"# More advanced exercises\n",
"\n",
"For a great variation with a 3-way bringing Gemini into the conversation!\n",
"Try creating a 3-way, perhaps bringing Gemini into the conversation! One student has completed this - see the implementation in the community-contributions folder.\n",
"\n",
"Try doing this yourself before you look in the folder.\n",
"Try doing this yourself before you look at the solutions.\n",
"\n",
"## Additional exercise\n",
"\n",
"Try adding in an Ollama model in to the conversation.\n",
"\n",
"## Business relevance\n",
"\n",
"This structure of a conversation, as a list of messages, is fundamental to the way we build conversational AI assistants and how they are able to keep the context during a conversation. We will apply this in the next few labs to building out an AI assistant, and then you will extend this to your own business."
"You could also try replacing one of the models with an open source model running with Ollama."
]
},
{
"cell_type": "markdown",
"id": "446c81e3-b67e-4cd9-8113-bc3092b93063",
"metadata": {},
"source": [
"<table style=\"margin: 0; text-align: left;\">\n",
" <tr>\n",
" <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
" <img src=\"../business.jpg\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
" </td>\n",
" <td>\n",
" <h2 style=\"color:#181;\">Business relevance</h2>\n",
" <span style=\"color:#181;\">This structure of a conversation, as a list of messages, is fundamental to the way we build conversational AI assistants and how they are able to keep the context during a conversation. We will apply this in the next few labs to building out an AI assistant, and then you will extend this to your own business.</span>\n",
" </td>\n",
" </tr>\n",
"</table>"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "0d86790a-3a6f-4b18-ab0a-bc6107945a27",
"id": "c23224f6-7008-44ed-a57f-718975f4e291",
"metadata": {},
"outputs": [],
"source": []

83
week2/day2.ipynb
View File

@@ -172,6 +172,8 @@
"metadata": {},
"outputs": [],
"source": [
"# The simplicty of gradio. This might appear in \"light mode\" - I'll show you how to make this in dark mode later.\n",
"\n",
"gr.Interface(fn=shout, inputs=\"textbox\", outputs=\"textbox\").launch()"
]
},
@@ -182,9 +184,55 @@
"metadata": {},
"outputs": [],
"source": [
"# Adding share=True means that it can be accessed publically\n",
"# A more permanent hosting is available using a platform called Spaces from HuggingFace, which we will touch on next week\n",
"\n",
"gr.Interface(fn=shout, inputs=\"textbox\", outputs=\"textbox\", flagging_mode=\"never\").launch(share=True)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "cd87533a-ff3a-4188-8998-5bedd5ba2da3",
"metadata": {},
"outputs": [],
"source": [
"# Adding inbrowser=True opens up a new browser window automatically\n",
"\n",
"gr.Interface(fn=shout, inputs=\"textbox\", outputs=\"textbox\", flagging_mode=\"never\").launch(inbrowser=True)"
]
},
{
"cell_type": "markdown",
"id": "b42ec007-0314-48bf-84a4-a65943649215",
"metadata": {},
"source": [
"## Forcing dark mode\n",
"\n",
"Gradio appears in light mode or dark mode depending on the settings of the browser and computer. There is a way to force gradio to appear in dark mode, but Gradio recommends against this as it should be a user preference (particularly for accessibility reasons). But if you wish to force dark mode for your screens, below is how to do it."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e8129afa-532b-4b15-b93c-aa9cca23a546",
"metadata": {},
"outputs": [],
"source": [
"# Define this variable and then pass js=force_dark_mode when creating the Interface\n",
"\n",
"force_dark_mode = \"\"\"\n",
"function refresh() {\n",
" const url = new URL(window.location);\n",
" if (url.searchParams.get('__theme') !== 'dark') {\n",
" url.searchParams.set('__theme', 'dark');\n",
" window.location.href = url.href;\n",
" }\n",
"}\n",
"\"\"\"\n",
"gr.Interface(fn=shout, inputs=\"textbox\", outputs=\"textbox\", flagging_mode=\"never\", js=force_dark_mode).launch()"
]
},
{
"cell_type": "code",
"execution_count": null,
@@ -192,6 +240,8 @@
"metadata": {},
"outputs": [],
"source": [
"# Inputs and Outputs\n",
"\n",
"view = gr.Interface(\n",
" fn=shout,\n",
" inputs=[gr.Textbox(label=\"Your message:\", lines=6)],\n",
@@ -208,6 +258,8 @@
"metadata": {},
"outputs": [],
"source": [
"# And now - changing the function from \"shout\" to \"message_gpt\"\n",
"\n",
"view = gr.Interface(\n",
" fn=message_gpt,\n",
" inputs=[gr.Textbox(label=\"Your message:\", lines=6)],\n",
@@ -224,6 +276,11 @@
"metadata": {},
"outputs": [],
"source": [
"# Let's use Markdown\n",
"# Are you wondering why it makes any difference to set system_message when it's not referred to in the code below it?\n",
"# I'm taking advantage of system_message being a global variable, used back in the message_gpt function (go take a look)\n",
"# Not a great software engineering practice, but quite sommon during Jupyter Lab R&D!\n",
"\n",
"system_message = \"You are a helpful assistant that responds in markdown\"\n",
"\n",
"view = gr.Interface(\n",
@@ -243,6 +300,8 @@
"outputs": [],
"source": [
"# Let's create a call that streams back results\n",
"# If you'd like a refresher on Generators (the \"yield\" keyword),\n",
"# Please take a look at the Intermediate Python notebook in week1 folder.\n",
"\n",
"def stream_gpt(prompt):\n",
" messages = [\n",
@@ -334,7 +393,9 @@
"\n",
"There's actually a more elegant way to achieve this (which Python people might call more 'Pythonic'):\n",
"\n",
"`yield from result`"
"`yield from result`\n",
"\n",
"I cover this in more detail in the Intermediate Python notebook in the week1 folder - take a look if you'd like more."
]
},
{
@@ -380,6 +441,26 @@
"Now you know how - it's simple!"
]
},
{
"cell_type": "markdown",
"id": "92d7c49b-2e0e-45b3-92ce-93ca9f962ef4",
"metadata": {},
"source": [
"<table style=\"margin: 0; text-align: left;\">\n",
" <tr>\n",
" <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
" <img src=\"../important.jpg\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
" </td>\n",
" <td>\n",
" <h2 style=\"color:#900;\">Before you read the next few cells</h2>\n",
" <span style=\"color:#900;\">\n",
" Try to do this yourself - go back to the company brochure in week1, day5 and add a Gradio UI to the end. Then come and look at the solution.\n",
" </span>\n",
" </td>\n",
" </tr>\n",
"</table>"
]
},
{
"cell_type": "code",
"execution_count": null,

18
week2/day3.ipynb
View File

@@ -256,11 +256,19 @@
"id": "82a57ee0-b945-48a7-a024-01b56a5d4b3e",
"metadata": {},
"source": [
"# Business Applications\n",
"\n",
"Conversational Assistants are of course a hugely common use case for Gen AI, and the latest frontier models are remarkably good at nuanced conversation. And Gradio makes it easy to have a user interface. Another crucial skill we covered is how to use prompting to provide context, information and examples.\n",
"\n",
"Consider how you could apply an AI Assistant to your business, and make yourself a prototype. Use the system prompt to give context on your business, and set the tone for the LLM."
"<table style=\"margin: 0; text-align: left;\">\n",
" <tr>\n",
" <td style=\"width: 150px; height: 150px; vertical-align: middle;\">\n",
" <img src=\"../business.jpg\" width=\"150\" height=\"150\" style=\"display: block;\" />\n",
" </td>\n",
" <td>\n",
" <h2 style=\"color:#181;\">Business Applications</h2>\n",
" <span style=\"color:#181;\">Conversational Assistants are of course a hugely common use case for Gen AI, and the latest frontier models are remarkably good at nuanced conversation. And Gradio makes it easy to have a user interface. Another crucial skill we covered is how to use prompting to provide context, information and examples.\n",
"<br/><br/>\n",
"Consider how you could apply an AI Assistant to your business, and make yourself a prototype. Use the system prompt to give context on your business, and set the tone for the LLM.</span>\n",
" </td>\n",
" </tr>\n",
"</table>"
]
},
{