\n<\/div>\n
Name it something like \u201cDave\u2019s Awesome Articles.\u201d You know\u2026 easy to remember.<\/p>\n
Next, you\u2019ll enable the API by clicking Enable APIs and services<\/em>.<\/p>\n\n<\/div>\n
Find the Knowledge Graph Search API, and enable it.<\/p>\n
\n<\/div>\n
You\u2019ll then be taken back to the main API page, where you can create credentials:<\/p>\n
\n<\/div>\n
And we\u2019ll be creating a service account.<\/p>\n
\n<\/div>\n
Simply create a service account:<\/p>\n
\n<\/div>\n
Fill in the required information:<\/p>\n
\n<\/div>\n
(You\u2019ll need to give it a name and grant it owner privileges.)<\/p>\n
Now we have our service account. All that\u2019s left is to create our key.\u00a0<\/p>\n
Click the three dots under Actions<\/em> and click Manage keys<\/em>.<\/p>\n\n<\/div>\n
Click Add key<\/em> then Create new key<\/em>:<\/p>\n\n<\/div>\n
The key type will be JSON.<\/p>\n
Immediately, you\u2019ll see it download to your default download location.<\/p>\n
This key will give access to your APIs, so keep it safe, just like your OpenAI API.<\/p>\n
Alright\u2026 and we\u2019re back. Ready to continue with our script?<\/p>\n
Now that we have them, we need to define our API key and path to the downloaded file. The code to do this is:<\/p>\n
os.environ[‘GOOGLE_APPLICATION_CREDENTIALS’] = ‘\/PATH-TO-FILE\/FILENAME.JSON’
\n%env OPENAI_API_KEY=YOUR_OPENAI_API_KEY
\nopenai.api_key = os.environ.get(“OPENAI_API_KEY”)<\/p>\n
You will replace YOUR_OPENAI_API_KEY with your own key.<\/p>\n
You will also replace \/PATH-TO-FILE\/FILENAME.JSON with the path to the service account key you just downloaded, including the file name.<\/p>\n
Run the cell and you\u2019re ready to move on.<\/p>\n
<\/span>Step 4: Create the functions<\/span><\/h3>\nNext, we\u2019ll create the functions to:<\/p>\n
Scrape the webpage we entered above.<\/p>\n
Analyze the content and extract the entities.<\/p>\n
Generate an article using GPT-4.<\/p>\n
#The function to scrape the web page
\ndef scrape_url(url):
\n response = requests.get(url)
\n soup = BeautifulSoup(response.content, “html.parser”)
\n paragraphs = soup.find_all(“p”)
\n text = ” “.join([p.get_text() for p in paragraphs])
\n return text<\/p>\n
#The function to pull and analyze the entities on the page using Google’s Knowledge Graph API
\ndef analyze_content(content):
\n client = language_v1.LanguageServiceClient()
\n response = client.analyze_entities(
\n request={“document”: language_v1.Document(content=content, type_=language_v1.Document.Type.PLAIN_TEXT), “encoding_type”: language_v1.EncodingType.UTF8}
\n )
\n top_entities = sorted(response.entities, key=lambda x: x.salience, reverse=True)[:10]
\n for entity in top_entities:
\n print(entity.name)
\n return top_entities<\/p>\n
#The function to generate the content
\ndef generate_article(content):
\n openai.api_key = os.environ[“OPENAI_API_KEY”]
\n response = openai.ChatCompletion.create(
\n messages = [{“role”: “system”, “content”: “You are a highly skilled writer, and you want to produce articles that will appeal to users and rank well.”},
\n {“role”: “user”, “content”: content}],
\n model=”gpt-4″,
\n max_tokens=1500, #The maximum with GPT-3 is 4096 including the prompt
\n n=1, #How many results to produce per prompt
\n #best_of=1 #When n>1 completions can be run server-side and the “best” used
\n stop=None,
\n temperature=0.8 #A number between 0 and 2, where higher numbers add randomness
\n)
\n return response.choices[0].message.content.strip()<\/p>\n
This is pretty much exactly what the comments describe. We\u2019re creating three functions for the purposes outlined above.<\/p>\n
Keen eyes will notice:<\/p>\n
messages = [{“role”: “system”, “content”: “You are a highly skilled writer, and you want to produce articles that will appeal to users and rank well.”},<\/p>\n
You can edit the content (You are a highly skilled writer, and you want to produce articles that will appeal to users and rank well.) and describe the role you want ChatGPT to take. You can also add tone (e.g., \u201cYou are a friendly writer \u2026\u201d).<\/p>\n