일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
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 |
- NMF
- 경사하강법
- SOMs
- TFX
- Python
- Clustering
- MLOps
- Binary classification
- Attention
- BERT
- Logistic Regression
- NER
- tensorflow
- ResNet
- RNN
- Transfer Learning
- Ann
- LSTM
- stemming
- cross domain
- AI 윤리
- nlp
- textmining
- Gradient Descent
- 자기조직화지도
- Support Vector Machine
- gaze estimation
- VGGNet
- Generative model
- 군집화
- Today
- Total
목록Deep Learning Study (13)
juooo1117

이미지에서 객체를 검출하도록 학습된 YOLOV3이라는 모델을 배포해 보자. 학습된 모델이 작동하는 모습을 우선 확인해 보기 위해서 아래의 함수를 정의했고, 이미에서 객체를 올바르게 검출하는 것을 확인했다. # Read the image into a numpy array # Perform the object detection # Print detected objects with confidence level # Create a new image that includes the bounding boxes # Save the image in the directory images_with_boxes # Display the image with bounding boxes import cv2 import cvlib..
During this week you saw how to create a model that will predict the next word in a text sequence, now you will implement such model and train it using a corpus of Shakespeare's sonnets, while also creating some helper functions to pre-process the data. Generating n_grams 적합한 토크나이저와 말뭉치(문자열 목록)를 수신하고 말뭉치의 각 줄에 대한 n_gram 시퀀스를 포함하는 목록을 반환한다. # texts_to_sequences : converting the text into sequence..
Transfer Learning : utilize an already trained network to help you solve a similar problem to the one it was originally trained to solve → 이미 훈련된 네트워크를 활용하여 훈련된 모델과 유사한 문제를 해결하는 데 도움이 되는 기술 horse & human 이미지를 가지고 image classifier를 만들어보자. Dataset As expected, the sample image has a resolution of 300x300 and the last dimension is used for each one of the RGB channels to represent color. → 데이터셋의 사이..

AutoEncoder? 입력이 들어왔을 때, 해당 입력 데이터를 최대한 compression(압축)시킨 후, compressed data를 다시 본래의 입력 형태로 복원 시키는 신경망으로 비지도 학습 방법이다. 이때, 데이터를 압축하는 부분을 Encoder(recognition network)라고 하고, 복원하는 부분을 Decoder(generative network)라고 부른다. 압축 과정에서 추출한 의미 있는 데이터 Z를 보통 latent vector라고 부르고, 이 외에도도 ‘latent variable’, ‘latent vector’, ‘code’, ‘feature’로 부른다. 인코딩 - 디코딩 과정을 거치면서, reconstruction error 가 원래의 입력 데이터와 복원 결과물의 차이를 ..

What are autoencoders? "Autoencoding" is a data compression algorithm where the compression and decompression functions are 1) data-specific 2) lossy 3) learned automatically from examples rather than engineered by a human. Additionally, in almost all contexts where the term "autoencoder" is used, the compression and decompression functions are implemented with neural networks. - Autoencoders ar..

Boltzmann MachineWhat makes the Boltzmann Machine stand out?Botlzmann machine has connections too but it does not operate in a direction → 볼츠만 머신은 undirected model 이다.아래 그림에서 볼 수 있듯이, the connections of the Boltzmann Machine has no arrows → so it operates on a random directionouput layer가 따로없고 모든것은 연결되어 있으며 방향성은 존재하지 않는다.Generative Model이다. ⇒ Deterministic Model들이 타겟과 가설 간의 차이를 줄여서 오차를 줄이는 것이 목표..

이전에 Self Organized Maps 을 이용해서 frauds를 찾아냈는데, 이 결과를 이용해서 고객 정보를 확인해서 어떤 고객이 frauds가 될 위험이 있는지 알아내는 task를 수행하고자 한다. 즉, SOM이라는 unsupervised deep learning의 결과를 supervised deep learning에 활용하는 방법을 연습했다. Creating the matrix of features # importing the dataset dataset = pd.read_csv('/content/MyDrive/MyDrive/Credit_Card_Applications.csv') som_frauds = pd.read_csv('/content/MyDrive/MyDrive/som_frauds.csv..

Finding Frauds using Credit card applications dataset Data Load & Preprocessing credit card applications dataset 을 이용해서 누가 사기꾼인지 알아내는 task 를 수행해 보자. dataset 을 확인해보니, 690명의 고객(row)과 특징 A1~A14(column)이 존재하고, 마지막 'class' column에는 이 고객이 승인되었는지 아닌지의 여부가 '0'(not approved), '1'(approved) 로 기재되어 있다. # importing the dataset dataset = pd.read_csv('/content/MyDrive/MyDrive/Credit_Card_Applications.csv') # ..

Self-organizing maps are even often referred to as Kohonen maps.Un-supervised deep learning algorithmSOMs의 목적reducing dimensionality (reduce amount of columns → 2-dimension으로 출력한다.)매우 많은 columns와 dimensions 을 가진 complex dataset을 simplified map으로 만들어 준다. The map provides you with a two-dimensional representation of the exact same dataset → 읽기가 더 쉬워진다. How do SOMs Learn?input vector - 3개의 feature ..

Predict stock price of google Part 1 - Data Preprocessing Importing the training set Google_Stock_Price_Train.csv의 파일에서 .iloc을 이용해서 'Open' column(index → 1)만 가져온다. 단, simple vector가 아니라 np array를 사용해야 하므로 [:, 1] 이 아니라, [:, 1:2]를 입력해 주어야 한다. (1:2를 입력해도, 파이썬에서는 '마지막숫자-1'로 계산되기 때문에, index1만 가져오게 되기때문!) dataset_train = pd.read_csv('/content/drive/MyDrive/Google_Stock_Price_Train.csv') training_set =..