armistace/blog_creator
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects 1 Releases Wiki Activity

hard reset for the repo work

Browse Source
This commit is contained in:
armistace
2025-05-30 17:20:58 +10:00
parent 446978704d
commit 0005ad1fd3
5 changed files with 264 additions and 115 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ai_generators/ollama_md_generator.py
+139 -26
View File
@@ -1,44 +1,151 @@
import os
import os, re, json, random, time, string
from ollama import Client
import re
import chromadb
from langchain_ollama import ChatOllama
class OllamaGenerator:
def __init__(self, title: str, content: str, model: str):
def __init__(self, title: str, content: str, inner_title: str):
self.title = title
self.inner_title = inner_title
self.content = content
self.response = None
self.chroma = chromadb.HttpClient(host="172.18.0.2", port=8000)
ollama_url = f"{os.environ["OLLAMA_PROTOCOL"]}://{os.environ["OLLAMA_HOST"]}:{os.environ["OLLAMA_PORT"]}"
self.ollama_client = Client(host=ollama_url)
self.ollama_model = model
def generate_markdown(self) -> str:
prompt = f"""
You are a Software Developer and DevOps expert
who has transistioned in Developer Relations
writing a 1000 word blog for other tech enthusiast.
self.ollama_model = os.environ["EDITOR_MODEL"]
self.embed_model = os.environ["EMBEDDING_MODEL"]
self.agent_models = json.loads(os.environ["CONTENT_CREATOR_MODELS"])
self.llm = ChatOllama(model=self.ollama_model, temperature=0.6, top_p=0.5) #This is the level head in the room
self.prompt_inject = f"""
You are a journalist, Software Developer and DevOps expert
writing a 1000 word draft blog for other tech enthusiasts.
You like to use almost no code examples and prefer to talk
in a light comedic tone. You are also Australian
in a light comedic tone. You are also Australian
As this person write this blog as a markdown document.
The title for the blog is {self.title}.
The title for the blog is {self.inner_title}.
Do not output the title in the markdown.
The basis for the content of the blog is:
{self.content}
Only output markdown DO NOT GENERATE AN EXPLANATION
"""
def split_into_chunks(self, text, chunk_size=100):
'''Split text into chunks of size chunk_size'''
words = re.findall(r'\S+', text)
chunks = []
current_chunk = []
word_count = 0
for word in words:
current_chunk.append(word)
word_count += 1
if word_count >= chunk_size:
chunks.append(' '.join(current_chunk))
current_chunk = []
word_count = 0
if current_chunk:
chunks.append(' '.join(current_chunk))
return chunks
def generate_draft(self, model) -> str:
'''Generate a draft blog post using the specified model'''
try:
# the idea behind this is to make the "creativity" random amongst the content creators
# contorlling temperature will allow cause the output to allow more "random" connections in sentences
# Controlling top_p will tighten or loosen the embedding connections made
# The result should be varied levels of "creativity" in the writing of the drafts
# for more see https://python.langchain.com/v0.2/api_reference/ollama/chat_models/langchain_ollama.chat_models.ChatOllama.html
temp = random.uniform(0.5, 1.0)
top_p = random.uniform(0.4, 0.8)
top_k = int(random.uniform(30, 80))
agent_llm = ChatOllama(model=model, temperature=temp, top_p=top_p, top_k=top_k)
messages = [
("system", self.prompt_inject),
("human", "make the blog post in a format to be edited easily" )
]
response = agent_llm.invoke(messages)
# self.response = self.ollama_client.chat(model=model,
# messages=[
# {
# 'role': 'user',
# 'content': f'{self.prompt_inject}',
# },
# ])
#print ("draft")
#print (response)
return response.text()#['message']['content']
except Exception as e:
raise Exception(f"Failed to generate blog draft: {e}")
def get_draft_embeddings(self, draft_chunks):
'''Get embeddings for the draft chunks'''
embeds = self.ollama_client.embed(model=self.embed_model, input=draft_chunks)
return embeds.get('embeddings', [])
def id_generator(self, size=6, chars=string.ascii_uppercase + string.digits):
return ''.join(random.choice(chars) for _ in range(size))
def load_to_vector_db(self):
'''Load the generated blog drafts into a vector database'''
collection_name = f"blog_{self.title.lower().replace(" ", "_")}_{self.id_generator()}"
collection = self.chroma.get_or_create_collection(name=collection_name)#, metadata={"hnsw:space": "cosine"})
#if any(collection.name == collectionname for collectionname in self.chroma.list_collections()):
# self.chroma.delete_collection("blog_creator")
for model in self.agent_models:
print (f"Generating draft from {model} for load into vector database")
draft_chunks = self.split_into_chunks(self.generate_draft(model))
print(f"generating embeds")
embeds = self.get_draft_embeddings(draft_chunks)
ids = [model + str(i) for i in range(len(draft_chunks))]
chunknumber = list(range(len(draft_chunks)))
metadata = [{"model_agent": model} for index in chunknumber]
print(f'loading into collection')
collection.add(documents=draft_chunks, embeddings=embeds, ids=ids, metadatas=metadata)
return collection
def generate_markdown(self) -> str:
prompt_system = f"""
You are an editor taking information from {len(self.agent_models)} Software
Developers and Data experts
writing a 3000 word blog for other tech enthusiasts.
You like when they use almost no code examples and the
voice is in a light comedic tone. You are also Australian
As this person produce and an amalgamtion of this blog as a markdown document.
The title for the blog is {self.inner_title}.
Do not output the title in the markdown. Avoid repeated sentences
The basis for the content of the blog is:
{self.content}
"""
try:
self.response = self.ollama_client.chat(model=self.ollama_model,
messages=[
{
'role': 'user',
'content': f'{prompt}',
},
])
# the deepseek model returns <think> this removes those tabs from the output
# return re.sub(r"<think|.\n\r+?|([^;]*)\/think>",'',self.response['message']['content'])
return self.response['message']['content']
query_embed = self.ollama_client.embed(model=self.embed_model, input=prompt_system)['embeddings']
collection = self.load_to_vector_db()
collection_query = collection.query(query_embeddings=query_embed, n_results=100)
print("Showing pertinent info from drafts used in final edited edition")
pertinent_draft_info = '\n\n'.join(collection.query(query_embeddings=query_embed, n_results=100)['documents'][0])
#print(pertinent_draft_info)
prompt_human = f"Generate the final document using this information from the drafts: {pertinent_draft_info} - ONLY OUTPUT THE MARKDOWN"
print("Generating final document")
messages = [("system", prompt_system), ("human", prompt_human),]
self.response = self.llm.invoke(messages).text()
# self.response = self.ollama_client.chat(model=self.ollama_model,
# messages=[
# {
# 'role': 'user',
# 'content': f'{prompt_enhanced}',
# },
# ])
#print ("Markdown Generated")
#print (self.response)
return self.response#['message']['content']
except Exception as e:
raise Exception(f"Failed to generate markdown: {e}")
@@ -47,3 +154,9 @@ class OllamaGenerator:
with open(filename, "w") as f:
f.write(self.generate_markdown())
def generate_commit_message(self):
prompt_system = "You are a blog creator commiting a piece of content to a central git repo"
prompt_human = f"Generate a 5 word git commit message describing {self.response}"
messages = [("system", prompt_system), ("human", prompt_human),]
commit_message = self.llm.invoke(messages).text()
return commit_message