PyTorch實作NLP中的Self-Attention

我們準備了兩個句子來進行這次實驗

sentences = ['helo attention','have a nice day']

一開始先建立詞表與對應的單詞one-hot encoding

vocabs = ' '.join(sentences).split()
vocabs = list(set(vocabs))

RainJayTsai / tw_address_analysis.py

Last active November 29, 2023 02:15

regex to analysis tw address

	try:
	import re2 as re
	except:
	import re


	pattern = '(?P<code>\d{5}\|\d{3})?\s*(?P<country>[台臺]灣)?(?P<city>\D+?[縣市])(?P<district>\D+?(市區\|鎮區\|鎮市\|[鄉鎮市區]))?(?P<village>\D+?[村里](?!路))?(?P<neighbor>\d+[鄰])?(?P<road>\D+?(村路\|[路街道段]))?(?P<section>(\d\|\D)?段)?(?P<lane>\d+巷)?(?P<alley>\d+弄)?(?P<no>\d+號?)?(?P<seq>[-之]\d+?(號))?(?P<floor>(((\D\|\d)+樓)\|((B\|地下室?)\d+)))?(?P<other>.+)?'
	addr_regex = re.compile(pattern)
	cn_num = '一二三四五六七八九十'
	cn_mapping = {k: str(v) for k, v in zip(list(cn_num),

RainJayTsai / .env

Created October 29, 2021 09:41 — forked from FaKleiser/.env

Run GitLab behind Traefik. https://www.fabian-keller.de/blog/migrating-a-gitlab-omnibus-deployment-to-docker

	S3_ACCESS_KEY_ID_GITLAB_BACKUP=<ACCESS_KEY>
	S3_SECRET_ACCESS_KEY_GITLAB_BACKUP=<SECRET_KEY>

RainJayTsai / convert_voc2createml.py

Created September 9, 2021 14:13 — forked from bezineb5/convert_voc2createml.py

Pascal VOC annotation to Apple createML annotations conversion tool

	import argparse
	import json
	import logging
	import pathlib
	from xml.etree import ElementTree

	log = logging.getLogger(__name__)


	def _parse_arguments() -> argparse.Namespace:

RainJayTsai / server.conf

Created June 7, 2021 03:04 — forked from laurenorsini/server.conf

OpenVPN configuration for /etc/openvpn/server.conf

	local 192.168.2.0 # SWAP THIS NUMBER WITH YOUR RASPBERRY PI IP ADDRESS
	dev tun
	proto udp #Some people prefer to use tcp. Don't change it if you don't know.
	port 1194
	ca /etc/openvpn/easy-rsa/keys/ca.crt
	cert /etc/openvpn/easy-rsa/keys/Server.crt # SWAP WITH YOUR CRT NAME
	key /etc/openvpn/easy-rsa/keys/Server.key # SWAP WITH YOUR KEY NAME
	dh /etc/openvpn/easy-rsa/keys/dh1024.pem # If you changed to 2048, change that here!
	server 10.8.0.0 255.255.255.0
	# server and remote endpoints

RainJayTsai / Makefile

Created January 13, 2020 03:25 — forked from doubleyou/Makefile

grpc-gateway python example

	GATEWAY_FLAGS := -I. -I/usr/local/include -I$(GOPATH)/src/github.com/grpc-ecosystem/grpc-gateway/third_party/googleapis -I/usr/local/include

	GRPC_FLAGS := --python_out=. --grpc_python_out=.

	code:
	python -m grpc_tools.protoc $(GRPC_FLAGS) $(GATEWAY_FLAGS) *.proto

	gw:
	protoc $(GATEWAY_FLAGS) \
	--go_out=Mgoogle/api/annotations.proto=github.com/grpc-ecosystem/grpc-gateway/third_party/googleapis/google/api,plugins=grpc:. \

RainJayTsai / build_tensorflow_in_docker

Created April 15, 2019 01:56

build tensorflow in docker the image is python:3.6 ( debian )

	FROM python:3.6

	# install bazel
	# check the bazel version what you want to build tensroflw and change the VERSION
	# https://www.tensorflow.org/install/source#linux
	ARG BAZEL_VERSION 0.15.0
	ARG TENSORFLOW_VERSION 1.12.0
	RUN apt-get update; apt-get install openjdk-8-jdk
	RUN set -ex && wget https://github.com/bazelbuild/bazel/releases/download/${BAZEL_VERSION}/bazel-${BAZEL_VERSION}-installer-linux-x86_64.sh \
	&& chmod +x bazel-${BAZEL_VERSION}-installer-linux-x86_64.sh ./bazel-${BAZEL_VERSION}-installer-linux-x86_64.sh \

RainJayTsai / minimize_docker_python_application

Last active May 26, 2022 06:25

using docker BUILDKIT example, cache pip and mount temp whl to install. tensorflow python3.6

	# syntax = docker/dockerfile:experimental

	FROM python:3.6 as base

	######################################
	FROM base as builder

	ENV PYTHONPATH=$PYTHONPATH:/install/lib/python3.6/site-packages/

	# bind a wheel and install it

RainJayTsai / compress.py

Created March 18, 2019 03:45

python sanic compress

	#!/usr/bin/env python3
	# -- coding: utf-8 --

	### fork from https://github.com/subyraman/sanic_compress

	import gzip

	DEFAULT_MIME_TYPES = frozenset([
	'text/html', 'text/css', 'text/xml',
	'application/json',

RainJayTsai / gmmhmm.py

Created February 20, 2019 06:29 — forked from kastnerkyle/gmmhmm.py

GMM-HMM (Hidden markov model with Gaussian mixture emissions) implementation for speech recognition and other uses

	# (C) Kyle Kastner, June 2014
	# License: BSD 3 clause

	import scipy.stats as st
	import numpy as np

	class gmmhmm:
	#This class converted with modifications from https://code.google.com/p/hmm-speech-recognition/source/browse/Word.m
	def __init__(self, n_states):
	self.n_states = n_states

RainJay RainJayTsai