王悄悄 wfsg00098

📚

Studying

wfsg00098 / llm_compression.py

Created June 13, 2024 06:36 — forked from 0xDigest/llm_compression.py

A demonstration of https://o565.com/llm-text-compression/

	import llama_cpp
	import re
	import json

	# Model configuration
	# tested with mistral, llama2, llama3, and phi3
	model_path = "/path/to/model"

	base_llm = llama_cpp.Llama(model_path, seed=42, n_gpu_layers=-1, n_ctx=4096, verbose=False, temperature=0.0)

wfsg00098 / ela.py

Created May 24, 2018 08:43 — forked from ewencp/ela.py

Quick, simple implementation of Error Level Analysis

	#!/usr/bin/env python

	# This is a really simple implementation of ELA as described in
	# http://blackhat.com/presentations/bh-dc-08/Krawetz/Whitepaper/bh-dc-08-krawetz-WP.pdf
	# You shouldn't actually use it, or at least read the paper carefully
	# and implement more of the techniques before drawing any conclusions.

	from PIL import Image, ImageChops, ImageEnhance
	import sys, os.path