Skip to content

Instantly share code, notes, and snippets.

@celsowm
celsowm / megaten_demons_fandom.json
Last active June 9, 2024 15:14
megaten_demons_fandom.json
This file has been truncated, but you can view the full file.
[
{
"title": "\nAscomid ",
"Japanese Name": "アスコモイド",
"Romaji": "Asukomoido",
"Origin": "American Mythology and Popular Culture",
"First Appearance": "Megami Tensei II",
"Race(s)": "Foul",
"appearances": [
"Megami Tensei II: Foul Race",
@celsowm
celsowm / megaten_fandom_orpo_dpo_english.json
Created June 1, 2024 13:31
megaten_fandom_orpo_dpo_english.json
[
{
"prompt": "What is Aegis: The First Mission ?",
"chosen": "An action RPG game set in May 1999, 10 years before the events of Persona 3, following the story of Aegis, an Anti-Shadow weapon.",
"rejected": "A first-person shooter game set in a futuristic world where you play as a human character fighting against aliens."
},
{
"prompt": "Summarize the game Aegis: The First Mission",
"chosen": "Aegis: The First Mission is an action RPG that follows the story of Aegis, an anti-Shadow weapon, as she awakens on Yakushima Island in 1999. She participates in training exercises and battles against Shadows, while also developing emotions and relationships with the researchers around her, particularly Yuu Kimijima. As the game progresses, Aegis uncovers the truth about her existence and the source of the Shadows, leading to a final confrontation that will determine her fate.",
"rejected": "Aegis: The First Mission is a puzzle game set in ancient Egypt, where
@celsowm
celsowm / megaten_games_fandom.json
Created May 31, 2024 13:47
megaten_games_fandom.json
[
{
"title": "Aegis: The First Mission",
"link": "https://megamitensei.fandom.com/wiki/Aegis:_The_First_Mission",
"article_text": "\n\nAegis: The First Mission\n\n\n\n\n\n\nGame Information\n\nDeveloper\nAtlus & BBMF\n\n\nPublisher\nBBMFG-Mode (Nintendo Switch and Steam)\n\n\nGenre\nAction RPG\n\n\nPlatform\nNTT DoCoMo Mobile, Softbank MobileNintendo SwitchSteam\n\n\n\nRelease Dates\n\nJapan \nNTT DoCoMo MobileOctober 29, 2007Softbank MobileNovember 14, 2007Nintendo Switch/SteamJune 6, 2024\n\n\n\nAegis: The First Mission (アイギス THE FIRST MISSION, Aigisu Za Fāsuto Misshon) is a game designed for Japanese feature phones that stars Aegis set in May of 1999, 10 years before the events of Persona 3.\n\n\nOn April 18th, 2024 it was announced a Japanese-exclusive port of Aegis: The First Mission would be released as part of G-Mode Archives+ on June 6th 2024[1].\n\n\n\n\nGameplay\n\n\n\nControls\n\nAegis is controlled using the phone's directional pad. 0 is used to attack, and enter causes
@celsowm
celsowm / auryn.txt
Last active May 25, 2024 15:08
Auryn
Auryn: A Jóia do Leste
História
A cidade de Auryn, fundada no início do século XVIII, nasceu como um pequeno posto de comércio situado ao longo do rio Elara. Os primeiros colonizadores, atraídos pelas ricas terras férteis e pelo acesso ao rio, estabeleceram-se ali com o intuito de cultivar e comercializar produtos agrícolas. Durante o século XIX, Auryn experimentou um rápido crescimento devido à descoberta de jazidas de ouro nas proximidades. Essa descoberta atraiu uma grande quantidade de imigrantes, transformando a cidade em um próspero centro econômico.
Geografia
Auryn está localizada em um vale pitoresco, rodeada por montanhas majestosas e florestas densas. O rio Elara serpenteia pela cidade, proporcionando uma fonte vital de água e um belo cenário natural. O clima é temperado, com verões amenos e invernos frios, tornando Auryn um local atraente para viver durante todo o ano.
Economia
A economia de Auryn é diversificada e robusta. Inicialmente impulsionada pela mineração de ouro, a cidade rapidamente se
@celsowm
celsowm / install.sh
Created April 30, 2024 01:35
Install llama-cpp-python with spack gcc load
source spack/share/spack/setup-env.sh
spack load gcc@12.3.0
CMAKE_ARGS="-DLLAMA_CUBLAS=on -DCMAKE_CUDA_ARCHITECTURES=all-major" FORCE_CMAKE=1 pip install llama-cpp-python --force-reinstall --upgrade --no-cache-dir
@celsowm
celsowm / fine_tunning_llama2_bode_alpaca.py
Created February 19, 2024 01:47
fine_tunning_llama2_bode_alpaca.py
from datasets import load_dataset, Dataset
from trl import SFTTrainer
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig, TrainingArguments
from peft import LoraConfig
import torch, sys
dataset = load_dataset("celsowm/auryn", split="train"
#, download_mode="force_redownload"
)
@celsowm
celsowm / advocacia_publica_estadual.json
Last active November 26, 2023 19:36
advocacia_publica_estadual.json
{
"perguntas_respostas": [
{
"pergunta": "O que é uma procuradoria estadual e qual é o seu papel no sistema jurídico?",
"resposta": "Uma procuradoria estadual é um órgão responsável pela representação jurídica do Estado em questões legais. Seu papel inclui a defesa dos interesses e direitos do Estado nas esferas judicial e extrajudicial."
},
{
"pergunta": "Quais são as principais funções desempenhadas por uma procuradoria estadual?",
"resposta": "As principais funções de uma procuradoria estadual incluem a representação judicial e extrajudicial do Estado, a elaboração de pareceres jurídicos, a condução de processos administrativos e a assessoria aos órgãos públicos estaduais em questões legais."
},
@celsowm
celsowm / neural_network_plot_dinamico.py
Created November 12, 2023 02:01
neural_network_plot_dinamico.py
import networkx as nx
import matplotlib.pyplot as plt
def plot_neural_network(entrada, oculta, saida):
G = nx.DiGraph()
G.add_nodes_from([f'Entrada{i+1}' for i in range(entrada)])
G.add_nodes_from([f'Oculta{i+1}' for i in range(oculta)])
G.add_nodes_from([f'Saida{i+1}' for i in range(saida)])
@celsowm
celsowm / neural_network_pytorch.py
Created November 12, 2023 01:59
neural_network_pytorch.py
import torch
import torch.nn as nn
import pandas as pd
# Ler o arquivo CSV
df = pd.read_csv('diabetes.csv')
# Extrair os dados para feature_set e labels
feature_set = torch.tensor(df.drop('diabetico', axis=1).values, dtype=torch.float32)
labels = torch.tensor(df['diabetico'].values, dtype=torch.float32)
@celsowm
celsowm / neural_network_tensorflow.py
Created November 12, 2023 01:59
neural_network_tensorflow.py
import tensorflow as tf
import pandas as pd
# Ler o arquivo CSV
df = pd.read_csv('diabetes.csv')
# Extrair os dados para feature_set e labels
feature_set = df.drop('diabetico', axis=1).values
labels = df['diabetico'].values