Alfred Increment alfredplpl

Research Scientist. Interests: data science, machine learning, robotics, neuroscience

alfredplpl / pagoda.html

Created August 1, 2026 08:48

pagoda voxel art

	<!DOCTYPE html>
	<html lang="ja">
	<head>
	<meta charset="UTF-8">
	<title>🌸 Voxel Pagoda Garden</title>
	<style>
	* {
	margin: 0;
	padding: 0;
	}

alfredplpl / Qwen-Image.py

Created August 5, 2025 06:36

	# Ref: https://huggingface.co/Qwen/Qwen-Image/discussions/7
	from diffusers import DiffusionPipeline, PipelineQuantizationConfig
	from diffusers import BitsAndBytesConfig
	import torch

	quant_config = PipelineQuantizationConfig(
	quant_mapping={
	"transformer": BitsAndBytesConfig(
	load_in_4bit=True,
	bnb_4bit_compute_dtype=torch.bfloat16,

alfredplpl / qwen-image.ipynb

Created August 5, 2025 02:45

Qwen-Image.ipynb

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

alfredplpl / calm2-7b-jmmlu.py

Last active April 2, 2024 01:59

CALM2をJMMLUで評価してみたい人用のスクリプト

	import torch
	from transformers import AutoTokenizer,AutoModelForCausalLM
	import pandas

	model_name_or_path = "cyberagent/calm2-7b-chat"
	tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
	model = AutoModelForCausalLM.from_pretrained(model_name_or_path, device_map="cpu", torch_dtype=torch.float32)

	# https://github.com/nlp-waseda/JMMLU/blob/main/JMMLU/college_computer_science.csv
	df=pandas.read_csv("college_computer_science.csv",header=None)

alfredplpl / gemma_finetune_lora.py

Created February 24, 2024 04:23

Gemma初心者ファインチューニングコードです。HFの設定などはよしなにやってください。

	# Reference #1: https://note.com/npaka/n/nc55e44e407ff
	# Reference #2: https://huggingface.co/blog/gemma-peft
	# Licence: MIT

	from peft import LoraConfig

	lora_config = LoraConfig(
	r=8,
	target_modules=["q_proj", "o_proj", "k_proj", "v_proj", "gate_proj", "up_proj", "down_proj"],
	task_type="CAUSAL_LM",

alfredplpl / convert_ckpt_to_diffusers.py

Created November 10, 2023 07:11

	# MIT
	from diffusers import StableDiffusionXLPipeline
	import torch

	pipe = StableDiffusionXLPipeline.from_single_file('/path/to/checkpoint.safetensors', torch_dtype=torch.float16)
	pipe.save_pretrained('/path/to/diffusers_version')

alfredplpl / lcm_with_consistency_decoder.py

Last active November 7, 2023 08:46

	from diffusers import DiffusionPipeline
	import torch
	from consistencydecoder import ConsistencyDecoder
	from PIL import Image
	import numpy as np

	pipe = DiffusionPipeline.from_pretrained("SimianLuo/LCM_Dreamshaper_v7", torch_dtype=torch.float32)

	decoder_consistency = ConsistencyDecoder(device="cuda:0") # Model size: 2.49 GB

alfredplpl / calm2_gptq_colab.ipynb

Created November 5, 2023 06:28

calm2_gptq_colab.ipynb

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

alfredplpl / collectDALLE3fromLAION.py

Created October 9, 2023 06:15

	from datasets import load_dataset
	import requests
	from PIL import Image
	from tqdm import tqdm

	dataset = load_dataset("laion/dalle-3-dataset",split="train")


	for i,row in enumerate(tqdm(dataset)):
	with open(f"dalle3/{i:06}.txt","w") as f:

alfredplpl / whole_paper_read_llm.py

Created September 1, 2023 19:47

	# MIT License
	from transformers import AutoTokenizer
	import transformers
	from langchain.document_loaders import PyPDFLoader
	import torch

	model = "NousResearch/Yarn-Llama-2-13b-128k"

	tokenizer = AutoTokenizer.from_pretrained(model)
	pipeline = transformers.pipeline(