uv-instructions.txt

GlOBAL INSTALLATION OF UV
https://github.com/astral-sh/uv
https://medium.com/bitgrit-data-science-publication/forget-pip-install-use-this-instead-754863c58f1e

➜ pip install uv

# On macOS and Linux.
curl -LsSf https://astral.sh/uv/install.sh | sh

# On Windows.
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

# With pip.
pip install uv

# With pipx.
pipx install uv

# With Homebrew.
brew install uv

# With Pacman.
pacman -S uv


CREAtE A VENV
uv venv  # Create a virtual environment at .venv.

➜ uv venv  # Create a virtual environment at .venv.
Using Python 3.10.0rc2 interpreter at: C:\Users\FabioMatricardi\AppData\Local\Programs\Python\Python310\python.exe
Creating virtualenv at: .venv
Activate with: .venv\Scripts\activate
➜ cpuORPO ⚡


INSTALL PACKAGES
uv pip install flask                # Install Flask.
uv pip install -r requirements.txt  # Install from a requirements.txt file.
uv pip install -e .                 # Install the current project in editable mode.
uv pip install "package @ ."        # Install the current project from disk
uv pip install "flask[dotenv]"      # Install Flask with "dotenv" extra.

uv pip install llama-cpp-python[server]==0.2.60
uv pip install openai
uv pip install streamlit
uv pip install langchain
uv pip install tiktoken
uv pip install sentence_transformers
uv pip install rich
uv pip install --upgrade pymupdf


!wget https://github.com/fabiomatricardi/cdQnA/raw/main/2024-04-09%2007.05.35%20Review%20of%20Abraham%20Flexner%20s%20The%20Usefulness%20of%20Use.txt
!wget https://github.com/fabiomatricardi/cdQnA/raw/main/2024-04-11%2012.52.28%20Kaggle%20s%20wrong%20turn%20when%20AI%20becomes%20a%20teacher%20and.txt
!wget https://github.com/fabiomatricardi/cdQnA/raw/main/28884E00-%20SYSTEM%20OPERATIONAL%20TEST%20PROCEDURE%20PREPARATION%20CUIDELINE.pdf
!wget https://github.com/fabiomatricardi/cdQnA/raw/main/Dark%20Psychology%20and%20Manipulation%20-%20The%20Ultimate%20Guide%20To%20Master%20The%20Art%20Of%20Persuasion.pdf
!wget https://github.com/fabiomatricardi/cdQnA/raw/main/UsefulnessHarpers.pdf
!wget https://github.com/fabiomatricardi/cdQnA/raw/main/2024-04-12%2013.16.08%20Kaggle%20s%20wrong%20turn%20when%20AI%20becomes%20a%20teacher%20and.txt



PREIMPORT
```
from tqdm.rich import trange, tqdm
from rich.markdown import Markdown
import warnings
warnings.filterwarnings(action='ignore')
import datetime
from rich.console import Console
console = Console(width=110)
from llama_cpp import Llama
import tiktoken
encoding = tiktoken.get_encoding("r50k_base")
```

LOAD PDF
```
#Load PDF Function
import os
import fitz #pyMuPDF
miofile = "/content/28884E00- SYSTEM OPERATIONAL TEST PROCEDURE PREPARATION CUIDELINE.pdf"
def LoadPDFandWork(filepath,chunks, overlap):
  from langchain.document_loaders import TextLoader
  from langchain.text_splitter import TokenTextSplitter
  TOKENtext_splitter = TokenTextSplitter(chunk_size=chunks, chunk_overlap=overlap)
  #splitted_text = TOKENtext_splitter.split_text(fulltext) #create a list
  from langchain_community.document_loaders import PyMuPDFLoader
  import datetime
  start = datetime.datetime.now()
  console.print('1. loading pdf')
  loader = PyMuPDFLoader(filepath) #on Win local simply 'stl-0000011.pdf'
  data = loader.load_and_split(TOKENtext_splitter)
  delta = datetime.datetime.now() - start
  console.print(f'2. Loaded in {delta}')
  console.print(f'3. Number of items: {len(data)}')
  console.print('---')
  its = 0
  chars = 0
  solotesto = ''
  for items in data:
      testo = len(items.page_content)
      solotesto = solotesto + ' ' + items.page_content
      #console.print(f"Number of CHAR in Document {its}: {testo}")
      its = its + 1
      chars += testo

  console.print('---')
  console.print(f'> Total lenght of text in characthers: {chars}')
  console.print('---')
  return data,solotesto
d,article =  LoadPDFandWork(miofile, 300,50)
import tiktoken
encoding = tiktoken.get_encoding("r50k_base")
context_count = len(encoding.encode(article))
console.print(f"Number of Tokens in the Article: {context_count}")
docs = []
for items in d:
  docs.append(items.page_content)
```


lOAD A TXT FILE
```
#FOR TXT
with open('/content/2024-04-11 12.52.28 Kaggle s wrong turn when AI becomes a teacher and.txt', encoding='utf-8') as f:
  article = f.read()
f.close()
import tiktoken
encoding = tiktoken.get_encoding("r50k_base")
context_count = len(encoding.encode(article))
console.print(f"Number of Tokens in the Article: {context_count}")  
from langchain.document_loaders import TextLoader
from langchain.text_splitter import TokenTextSplitter
TOKENtext_splitter = TokenTextSplitter(chunk_size=700, chunk_overlap=50)
d = TOKENtext_splitter.split_text(article) #create a list
console.print(f"Number of Document Chunks in the Article: {len(d)}") 
```