728x90
반응형

전체 글 60

후쿠오카 여행

후쿠오카 3대 라멘집이라고 하는 하카타 신신라멘 집에 갔다. https://maps.app.goo.gl/uuUkVjgBYMMSMGAv5 웨이팅이 있다고 해서 오픈 30분 전쯤 도착해서 기다렸는데 오픈 할 때쯤 되니 줄이 꽤 많이 생겼다. 하지만 오픈때 바로 들어갈 수 있었다. 역시 맛있긴 했지만 느끼해서 김치가 생각나긴 했다. 그 후 다도를 할 수 있는 곳이 있다 하여 간 곳이 라쿠이스엔이다. https://maps.app.goo.gl/hZwK7tDjKufLsJFc7 라쿠스이엔 · 2 Chome-10-7 Sumiyoshi, Hakata Ward, Fukuoka, 812-0018 일본 ★★★★☆ · 정원 www.google.co.kr 일본식 정원이 있는 차 집인데 말차와 전통과자를 같이 판매한다. 방에서 ..

여행 2024.02.05

손으로 마우스 조종 with Mediapipe

Mediapipe 딥러닝을 이용한 라이브러리 중에 영상 정보를 통한 다양한 기능을 가진 Mediapipe가 있다. https://developers.google.com/mediapipe/solutions MediaPipe | Google for Developers Attention: This MediaPipe Solutions Preview is an early release. Learn more Stay organized with collections Save and categorize content based on your preferences. Compose on-device ML in minutes Supercharge your machine learning development and creat..

Deep Learning 2024.01.10

Recurrent Neural Network in Tensorflow

1. RNN RNN은 hidden state를 통해서 전 데이터가 이후의 데이터에 영향을 주어 연속적인 input이나 연속적인 output을 가능하게 해준다. 아래는 RNN many to one의 기본적인 구조이다. 여기서 h_t = tanh(W_hh * h_(t-1) + W_xh * x_t)이 hidden layer의 기본적인 구조이다. 2. in Tensorflow 1 2 3 4 5 6 7 8 9 10 11 12 import tensorflow as tf model = tf.keras.Sequential() model.add(tf.keras.layers.Embedding(input_dim=input_dim, #Embedding layer output_dim=output_dim, trainable=Fa..

Deep Learning 2024.01.07

Convolution Neural Network in Tensorflow

1. Convolution Neural Network convolution neural network는 이미지 분류에 가장 많이 쓰이는 인공지능 설계 방식이다. 특이한 점으로는 convolution layer와 pooling layer을 가지고 있다는 것이다. convolution layer는 이미지의 각각의 부분을 나누어 분석하면서 각 부분의 특징을 분류한다고 할 수 있다. pooling layer는 convolution layer에서 분류한 요소 중 유의미한 정보를 뽑아내는 것 같고 그렇기에 max pooling을 많이 사용하는 거 같다. 2. 예제 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 3..

Deep Learning 2023.12.15

Gradient descent algorithm in Tensor flow 경사하강법

1. gradient descent gradient descent는 함수의 최소값을 찾는 알고리즘이다. 그림과 같이 무작위 값에 기울기 값 * learning_rate를 곱해서 빼는 과정을 반복하면서 점점 주변 값보다 작은 값을 찾아간다. 2. gradient descent in tensorflow 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 import tensorflow as tf tf.random.set_seed(0) W = tf.Variable(tf.random.normal([1], -100., 100.)) for step in range(300) : learning_rate = 0.01 with tf.GradientTape() as tape : func = tf.s..

Deep Learning 2023.11.05

Notification with Python (FCM) in Android Studio

python을 이용해서 Firebase Cloud Messaging을 통해서 안드로이드 폰에 알람을 띄우는 것을 구현해봤다. android에 firebase를 연결하는 거나 python의 모듈을 다운 받는 것은 생략한다. 1. FirebaseInstanceIDService.java (android part) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 package com.example.smarttrashcanapplication; import ... pub..

Android App 2021.12.21

Room Database in Android Studio

https://youtu.be/jXzVrK4DXwU 1. 라이브러리 1 2 3 4 5 dependencies { ... implementation 'androidx.room:room-runtime:2.2.6' annotationProcessor 'androidx.room:room-compiler:2.2.6' } Colored by Color Scripter cs 위와 같이 room에 대한 라이브러리를 추가해준다. 사실 Alt + Enter를 통해서 자동으로 연결할 수도 있다. 2. User.java (Entity) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 package com.example.roomdatabaseexample; import androidx.room.Ent..

Android App 2021.11.27

RecyclerView with Firebase in Android Studio

https://youtu.be/HcPWt69d_wY 1. 라이브러리 추가 1) https://github.com/bumptech/glide GitHub - bumptech/glide: An image loading and caching library for Android focused on smooth scrolling An image loading and caching library for Android focused on smooth scrolling - GitHub - bumptech/glide: An image loading and caching library for Android focused on smooth scrolling github.com 위 페이지에 나와있는 설명처럼 아래 두 줄을 추..

Android App 2021.11.27
728x90
반응형