Skip to content

Instantly share code, notes, and snippets.

View admariner's full-sized avatar

Periklis Papanikolaou admariner

  • admariner
  • Thessaloniki
View GitHub Profile
@admariner
admariner / positive-words.txt
Last active June 6, 2024 01:08 — forked from mkulakowski2/positive-words.txt
Opinion Lexicon: Positive
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;
; Opinion Lexicon: Positive
;
; This file contains a list of POSITIVE opinion words (or sentiment words).
;
; This file and the papers can all be downloaded from
; http://www.cs.uic.edu/~liub/FBS/sentiment-analysis.html
;
; If you use this list, please cite one of the following two papers:
@admariner
admariner / negative-words.txt
Last active June 6, 2024 01:06 — forked from mkulakowski2/negative-words.txt
Opinion Lexicon: Negative
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;
; Opinion Lexicon: Negative
;
; This file contains a list of NEGATIVE opinion words (or sentiment words).
;
; This file and the papers can all be downloaded from
; http://www.cs.uic.edu/~liub/FBS/sentiment-analysis.html
;
; If you use this list, please cite one of the following two papers:
@admariner
admariner / HLC.md
Created April 2, 2024 04:48 — forked from ruvnet/HLC.md
f763620dbb895ea6410aed952bfa4cf5

Incorporating a Hypergraph Lambda Calculus (HLC) based model as part of a larger mixture of experts system could provide several benefits and enhance the overall capabilities of the model:

  1. Improved Reasoning Capabilities:

    • HLC's higher-order logic and lambda calculus foundations enable more sophisticated reasoning capabilities.
    • The model can handle complex dependencies, abstractions, and quantification, allowing it to perform advanced inference and deduction tasks.
    • This can complement other experts in the mixture that may focus on pattern recognition, data-driven learning, or specialized domain knowledge.
  2. Enhanced Expressiveness:

    • HLC's hypergraph-based representation allows modeling complex structures and relationships that may be difficult to capture with traditional graph-based or vector-based representations.
  • The model can express and reason about intricate domain knowledge, logical rules, and constraints.
@admariner
admariner / gist:83359d6f9800cda5896b84c522923b4d
Created April 2, 2024 04:44 — forked from manchumahara/gist:dc88a6b9b157ada5f02cb8408653b80f
Google My Business Categories (Json formatted)
[
"Abbey",
"Aboriginal and Torres Strait Islander organization",
"Aboriginal art gallery",
"Abortion clinic",
"Abrasives supplier",
"Abundant Life church",
"Accountant",
"Accounting firm",
"Accounting school",
@admariner
admariner / gist:a37ebf708e7c1c97e83b8575d575fe96
Created March 21, 2024 08:28 — forked from boydnorwood/gist:e19086c77c477b8ad32f00d0c1247add
SQL for generating a basic rankings report in Data Studio with data from your Nozzle workspace
--Top Ranking URLs Report for Nozzle.io
--Data Studio Template can be found here: https://datastudio.google.com/u/1/reporting/359d4414-0cd6-4da1-8df5-2c6908e0ddec/page/pyxcB
WITH
-- find the latest versioned keyword data
-- this can also be used to pin a query to an older version, good for static reports
latest_keyword_source_versions AS (
SELECT keyword_source_id, MAX(keyword_source_version_id) AS keyword_source_version_id
FROM nozzledata.nozzle_nozzleofficial.keywords
WHERE keyword_source_id=930701976723823
@admariner
admariner / finetune_llama2.py
Created March 20, 2024 17:41 — forked from mlabonne/finetune_llama2.py
Easy Llama 2 fine-tuning script (📝 Article: https://tinyurl.com/finetunellama2)
# Based on younesbelkada/finetune_llama_v2.py
# Install the following libraries:
# pip install accelerate==0.21.0 peft==0.4.0 bitsandbytes==0.40.2 transformers==4.31.0 trl==0.4.7 scipy
from dataclasses import dataclass, field
from typing import Optional
import torch
from datasets import load_dataset
from transformers import (
@admariner
admariner / cron.route.ts
Created March 16, 2024 23:20 — forked from adrianhajdin/cron.route.ts
Web Scraping Full Course 2023 | Build and Deploy eCommerce Price Tracker
import { NextResponse } from "next/server";
import { getLowestPrice, getHighestPrice, getAveragePrice, getEmailNotifType } from "@/lib/utils";
import { connectToDB } from "@/lib/mongoose";
import Product from "@/lib/models/product.model";
import { scrapeAmazonProduct } from "@/lib/scraper";
import { generateEmailBody, sendEmail } from "@/lib/nodemailer";
export const maxDuration = 300; // This function can run for a maximum of 300 seconds
export const dynamic = "force-dynamic";
# Must have conda installed
# It costs approximately $0.2 (in GPT-4 API fees) to generate one example with analysis and design, and around $2.0 for a full project.
conda create -n metagpt python=3.11.4
conda activate metagpt
npm --version # to check you have npm installed
# optional: install node if you don't have it
npm install -g @mermaid-js/mermaid-cli
git clone https://github.com/geekan/metagpt
cd metagpt
{
"basics": {
"name": "Pavlos Hatziapostolou",
"label": "Architectural Engineer",
"email": "phatziapostolou@gmail.com",
"phone": "6946507797",
"summary": "Creative and results-driven Architectural Engineer with over 20 years of comprehensive experience in designing, supervising, and executing a wide range of architectural projects from residential renovations to healthcare facilities construction. ",
"location": {
"address": "Filikis Etairias 30",
"postalCode": "54621",
/*
# Chat with ChatGPT
## <span class="text-primary">👉 Note: LangChain is still in development. This script will keep updating to use the latest APIs</span>
Use `Kit` -> `Manage npm Packages` -> `Update a Package` -> `langchain` to update to install the latest version.
- Opens the `chat` component
- Type a message and press `enter` to send
- The message is sent to the OpenAI API