Free AI web copilot to create summaries, insights and extended knowledge, download it at here
7654
Abstract
-string">"product"</span>],
template=<span class="hljs-string">"What is a good name for a company that makes {product}?"</span>,
<span class="hljs-built_in">print</span>(prompt.<span class="hljs-built_in">format</span>(product=<span class="hljs-string">"typing machines"</span>))
llm = OpenAI(temperature=<span class="hljs-number">0.9</span>)
text = (prompt.<span class="hljs-built_in">format</span>(product=<span class="hljs-string">"typing machines"</span>))
<span class="hljs-built_in">print</span>(llm(text))</pre></div><p id="107e">⬇️ Notebook view:</p><figure id="8121"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*sNd-kSpCkaenSz159KjRuQ.png"><figcaption></figcaption></figure><p id="0e77">The following prototype is a good example of how LangChain can be used to automate prompt engineering, and leverage LLMs for conversational context:</p><div id="e533"><pre><span class="hljs-keyword">from</span> langchain.llms <span class="hljs-keyword">import</span> OpenAI
<span class="hljs-keyword">from</span> langchain.chains <span class="hljs-keyword">import</span> ConversationChain
<span class="hljs-keyword">from</span> langchain.chains.conversation.memory <span class="hljs-keyword">import</span> ConversationBufferMemory
llm = OpenAI(temperature=<span class="hljs-number">0</span>)
conversation = ConversationChain(
llm=llm,
verbose=<span class="hljs-literal">True</span>,
memory=ConversationBufferMemory()
)
conversation.predict(<span class="hljs-built_in">input</span>=<span class="hljs-string">"Hi there!"</span>)</pre></div><p id="e3e8">Response:</p><div id="f542"><pre>> Entering new ConversationChain chain...
Prompt after formatting:
The following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context. If the AI does not know the answer to a question, it truthfully says it does not know.
Current conversation:
Human: Hi there!
AI:
> Finished chain.
Hi there! It's nice to meet you. My name is AI. What's your name?</pre></div><p id="31b3">Input:</p><div id="3efb"><pre>conversation.predict(<span class="hljs-built_in">input</span>=<span class="hljs-string">"My name is Cobus"</span>)</pre></div><p id="1b3e">Response:</p><div id="8fca"><pre>> Entering new ConversationChain chain...
Prompt after formatting:
The following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context. If the AI does not know the answer to a question, it truthfully says it does not know.
Current conversation:
Human: Hi there!
AI: Hi there! It's nice to meet you. My name is AI. What's your name?
Human: My name is Cobus
AI:
> Finished chain.
Nice to meet you, Cobus! What can I do for you today?</pre></div><p id="bcbb">This approach is not optimal for longer conversations as the prompt size sent to the LLM will just grow too big.</p><p id="fdcf">In this final example, the user is prompted for input and this is used to generate the response, it also illustrates how two chains are combined.</p><div id="f1bf"><pre>title = <span class="hljs-built_in">input</span>(<span class="hljs-string">'What is your title? '</span>)
<span class="hljs-built_in">print</span> (<span class="hljs-string">'I have your title as '</span> + title)
era = <span class="hljs-built_in">input</span>(<span class="hljs-string">'Lasty, from what era should the text be? '</span>)
<span class="hljs-built_in">print</span> (<span class="hljs-string">'I have your era as '</span> + era)
<span class="hljs-comment"># This is an LLMChain to write a synopsis given a title of a play and the era it is set in.</span>
llm = OpenAI(temperature=<span class="hljs-number">.7</span>)
template = <span class="hljs-string">"""You are a playwright. Given the title of play and the era it is set in, it is your job to write a synopsis for that title.
Title: {title}
Era: {era}
Playwright: This is a synopsis for the above play:"""</span>
prompt_template = PromptTemplate(input_variables=[<span class="hljs-string">"title"</span>, <span class="hljs-string">'era'</span>], template=template)
synopsis_chain = LLMChain(llm=llm, prompt=prompt_template, output_key=<span class="hljs-string">"synopsis"</span>)
<span class="hljs-comment"># This is an LLMChain to write a review of a play given a synopsis.</span>
llm = OpenAI(temperature=<span class="hljs-number">.7</span>)
template = <span class="hljs-string">"""You are a play critic from the New York Times. Given the synopsis of play, it is your job to write a review for that play.
Play Synopsis:
{synopsis}
Review from a New York Times play critic of the above play:"""</span>
prompt_template = PromptTemplate(input_variables=[<span class="hljs-string">"synopsis"</span>], template=template)
review_chain = LLMChain(llm=llm, prompt=prompt_template, output_key=<span class="hljs-string">"review"</span>)
<span class="hljs-comment"># This is the overall chain where we run these two chains in sequence.</span>
<span class="hljs-keyword">from</span> langchain.chains <span class="hljs-keyword">import</span> SequentialChain
overall_chain = SequentialChain(
chains=[synopsis_chain, review_chain],
input_variables=[<span class="hljs-string">"era"</span>, <span class="hljs-string">"title"</span>],
<span class="hljs-comment"># Here we return multiple variables</span>
output_variables=[<span class="hljs-string">"synopsis"</span>, <span class="hljs-string">"review"</span>],
verbose=<span class="hljs-literal">True</span>)
review = overall_chain({<span class="hljs-string">"title"</span>:title, <span class="hljs-string">"era"</span>: era})
<span class="hljs-built_in">print</span> (review)</pre></div><p id="bc8d">And the output:</p><div id="b537"><pre>> Entering new SequentialChain chain...
> Finished chain.
{'title': 'Tragedy at sunset on the beach', 'era': 'Victorian ', 'synopsis': "\n\nTragedy at Sunset on the Beach is set in the Victorian era and tells the story of a young girl, Amelia, whose search for love and happiness leads her to a fateful encounter with a mysterious stranger on the beach at sunset. While walking along the shore, Amelia meets a man named William, who presents her with a rose as a symbol of his affection. Though she initially resists his advances, Amelia eventually finds herself drawn to William and the two quickly grow close. \n\nHowever, as the sun sets, a secret from William's past is revealed that changes everything. It turns out that William is a wanted criminal and is being pursued by the police. After a tense confrontation, William is arrested and taken away, leaving Amelia heartbroken and alone.\n\nIn the aftermath of their chance encounter, Amelia is forced to grapple with the tragedy of her unrequited love, and the harsh realities of life. Her journey serves as a reminder that love isn't always as simple and straightforward as it may seem.", 'review': '\n\nTragedy at Sunset on the Beach is a captivating tale of love, loss, and the power of the human spirit. The set in the Victorian era allows for a unique perspective on the themes of the story, and the performance by the cast is truly remarkable.\n\nThe story follows Amelia, a young girl whose search for love and happiness leads her to a fateful encounter with a mysterious stranger on the beach at sunset. We watch as Amelia’s relationship with the stranger, William, develops and evolves, only to be cut short when a secret from his past is revealed. This revelation forces Amelia to confront the harsh realities of life and the tragedy of her unrequited love.\n\nThe play is a thought-p
Options
rovoking exploration of love, sacrifice, and the power of human connection. It’s a story that is sure to resonate with audiences who are looking for a story of love, loss, and hope. Highly recommended.'}</pre></div><h1 id="e5a3">In Conclusion</h1><p id="2a02">In the recent past, I have written a number of times on the coming fragmentation and that a single contained Conversational AI framework will not suffice.</p><p id="43a6">Already we are seeing that LLMs are forcing traditional chatbot framework providers to look out ward to technologies they should make provision for.</p><p id="e321">LangChain is software that fills a critical voice which currently exists to make LLMs more conversational.</p><p id="b240"><b><i>⭐️ Please follow me on <a href="https://www.linkedin.com/in/cobusgreyling/">LinkedIn</a> for updates on Conversational AI ⭐️</i></b></p><figure id="10d2"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*lIm_TXh6TC9uGn63lOjZtQ.png"><figcaption></figcaption></figure><p id="42a0"><i>I’m currently the <a href="https://www.linkedin.com/in/cobusgreyling">Chief Evangelist</a> @ <a href="https://www.humanfirst.ai">HumanFirst</a>. I explore and write about all things at the intersection of AI and language; ranging from <a href="https://cobusgreyling.medium.com/the-large-language-model-landscape-9da7ee17710b">LLMs</a>, <a href="https://cobusgreyling.medium.com/the-five-categories-of-conversational-ai-f2410beeaf2f">Chatbots</a>, <a href="https://cobusgreyling.medium.com/three-key-voicebot-design-considerations-7bf25444dfec">Voicebots</a>, Development Frameworks, <a href="https://cobusgreyling.medium.com/testing-complex-utterances-with-the-co-here-humanfirst-integration-145ab4eedd84">Data-Centric latent spaces</a> and more.</i></p><div id="02f7" class="link-block">
<a href="https://www.humanfirst.ai">
<div>
<div>
<h2>NLU design tooling</h2>
<div><h3>“Conversation Designer, Retail, 10k+ employees The tool that turned conversation designers, into NLU designers” ★★★★★…</h3></div>
<div><p>www.humanfirst.ai</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/0*lgMaRGTU1PRX4MB7)"></div>
</div>
</div>
</a>
</div><figure id="7278"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*qPfFI9uFl04n1ZUywxH38w.png"><figcaption></figcaption></figure><figure id="9789"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*mwQw4LOeZdWG1AD8RDheXw.jpeg"><figcaption><a href="https://www.linkedin.com/in/cobusgreyling">https://www.linkedin.com/in/cobusgreyling</a></figcaption></figure><figure id="bcbe"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*qPfFI9uFl04n1ZUywxH38w.png"><figcaption></figcaption></figure><div id="525b" class="link-block">
<a href="https://cobusgreyling.medium.com/subscribe">
<div>
<div>
<h2>Get an email whenever Cobus Greyling publishes.</h2>
<div><h3>Get an email whenever Cobus Greyling publishes. By signing up, you will create a Medium account if you don’t already…</h3></div>
<div><p>cobusgreyling.medium.com</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/0*jN8P0LRxgRrefLMK)"></div>
</div>
</div>
</a>
</div><div id="b7d5" class="link-block">
<a href="https://eliza.community">
<div>
<div>
<h2>Eliza Language Technology Community — Language Technology: Conversational AI, NLP/NLP, CCAI…</h2>
<div><h3>ELIZA — Where language technology enthusiasts unite.</h3></div>
<div><p>eliza.community</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/0*yM6mui4LoM4fDMPx)"></div>
</div>
</div>
</a>
</div><div id="63e6" class="link-block">
<a href="https://cobusgreyling.medium.com/the-cobus-quadrant-of-nlu-design-4b1654f21d70">
<div>
<div>
<h2>The Cobus Quadrant™ Of NLU Design</h2>
<div><h3>NLU design is vital to planning and continuously improving Conversational AI experiences.</h3></div>
<div><p>cobusgreyling.medium.com</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/1*yLwyX_HKYjwqFY0IkmesCw.png)"></div>
</div>
</div>
</a>
</div><div id="e342" class="link-block">
<a href="https://cobusgreyling.medium.com/the-cobus-quadrant-of-conversation-design-capabilities-9a28c23409cd">
<div>
<div>
<h2>The Cobus Quadrant™ Of Conversation Design Capabilities</h2>
<div><h3>∗ This is part one of a two part series, please also take a look part two, the Cobus Quadrant of NLU Design.</h3></div>
<div><p>cobusgreyling.medium.com</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/1*xdvS5qYcZsxll7FEUWRitA.png)"></div>
</div>
</div>
</a>
</div><div id="198e" class="link-block">
<a href="https://chat.langchain.dev/">
<div>
<div>
<h2>LangChain Chat</h2>
<div><h3>LangChain documentation chatbot</h3></div>
<div><p>chat.langchain.dev</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/)"></div>
</div>
</div>
</a>
</div><figure id="a944"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*qPfFI9uFl04n1ZUywxH38w.png"><figcaption></figcaption></figure><div id="8473" class="link-block">
<a href="https://github.com/hwchase17/langchain">
<div>
<div>
<h2>GitHub - hwchase17/langchain: ⚡ Building applications with LLMs through composability ⚡</h2>
<div><h3>⚡ Building applications with LLMs through composability ⚡ pip install langchain Large language models (LLMs) are…</h3></div>
<div><p>github.com</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/0*JB0ci66QVXXsDtm8)"></div>
</div>
</div>
</a>
</div><div id="1e65" class="link-block">
<a href="https://langchain.readthedocs.io/">
<div>
<div>
<h2>Welcome to LangChain</h2>
<div><h3>Large language models (LLMs) are emerging as a transformative technology, enabling developers to build applications…</h3></div>
<div><p>langchain.readthedocs.io</p></div>
</div>
<div>
<div style="background-image: url(https://miro.readmedium.com/v2/resize:fit:320/)"></div>
</div>
</div>
</a>
</div><figure id="8d81"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*qPfFI9uFl04n1ZUywxH38w.png"><figcaption></figcaption></figure></article></body>
