Browse Source

remove client.close() to allow multiple llm runs

pull/231/head
Octavio Ortiz-Bosch 2 months ago
parent
commit
be59cb5378
  1. 54
      week1/community-contributions/Week_1-Day 2-Article_Title_Generator.ipynb

54
week1/community-contributions/Week_1-Day 2-Article_Title_Generator.ipynb

@ -9,7 +9,7 @@
"\n",
"Summarization use-case in which the user provides an article, which the LLM will analyze to suggest an SEO-optimized title.\n",
"\n",
"NOTES:\n",
"**NOTES**:\n",
"\n",
"1. This version does NOT support website scrapping. You must copy and paste the required article.\n",
"2. The following models were configured:\n",
@ -17,7 +17,21 @@
" b. Llama llama3.2\n",
" c. Deepseek deepseek-r1:1.5b\n",
" It is possible to configure additional models by adding the new model to the MODELS dictionary and its\n",
" initialization to the CLIENTS dictionary."
" initialization to the CLIENTS dictionary. Then, call the model with --> ***answer =\n",
" get_answer('NEW_MODEL')***.\n",
"3. Users are encouraged to assess and rank the suggested titles using any headline analyzer tool online.\n",
" Example: https://www.isitwp.com/headline-analyzer/. "
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e773daa6-d05e-49bf-ad8e-a8ed4882b77e",
"metadata": {},
"outputs": [],
"source": [
"# Confirming Llama is loaded\n",
"!ollama pull llama3.2"
]
},
{
@ -43,18 +57,11 @@
"source": [
"# set environment variables for OpenAi\n",
"load_dotenv(override=True)\n",
"api_key = os.getenv('OPENAI_API_KEY')\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e773daa6-d05e-49bf-ad8e-a8ed4882b77e",
"metadata": {},
"outputs": [],
"source": [
"# Confirming Llama is loaded\n",
"!ollama pull llama3.2"
"api_key = os.getenv('OPENAI_API_KEY')\n",
"\n",
"# validate API Key\n",
"if not api_key:\n",
" raise ValueError(\"No API key was found! Please check the .env file.\")"
]
},
{
@ -154,9 +161,6 @@
" messages=messages\n",
" )\n",
" \n",
" # closing LLM client connection\n",
" client.close()\n",
" \n",
" # return answer\n",
" return response.choices[0].message.content\n",
" "
@ -199,10 +203,10 @@
"metadata": {},
"outputs": [],
"source": [
"# get openAi answer\n",
"# get Llama answer\n",
"answer = get_answer('LLAMA')\n",
"\n",
"# display openAi answer\n",
"# display Llama answer\n",
"display(Markdown(f\"### {MODELS['LLAMA']} Answer\\n\\n{answer}\" ))"
]
},
@ -221,10 +225,10 @@
"metadata": {},
"outputs": [],
"source": [
"# get openAi answer\n",
"# get Deepseek answer\n",
"answer = get_answer('DEEPSEEK')\n",
"\n",
"# display openAi answer\n",
"# display Deepseek answer\n",
"display(Markdown(f\"### {MODELS['DEEPSEEK']} Answer\\n\\n{answer}\" ))"
]
},
@ -235,7 +239,7 @@
"source": [
"### Suggested future improvements\n",
"\n",
"1. Add support for website scrapping to replace copy/pasting of articles.\n",
"1. Add website scrapping support to replace copy/pasting of articles.\n",
"2. Improve the system_prompt to provide specific SEO best practices to adopt during the title generation.\n",
"3. Rephrase the system_prompt to ensure the model provides a single Title (not a list of suggestions). \n",
"4. Add the logic that would allow each model to assess the recommendations from the different models and \n",
@ -245,12 +249,10 @@
{
"cell_type": "code",
"execution_count": null,
"id": "1af8260b-5ba1-4eeb-acd0-02de537b1bf4",
"id": "cf7403ac-d43b-4493-98bb-6fee94950cb0",
"metadata": {},
"outputs": [],
"source": [
"S"
]
"source": []
}
],
"metadata": {

Loading…
Cancel
Save