All Good

Deep Learning5

👔 Tuning fashionMNIST Classifier I: Adjusting the Number of epochs 👔 유명한 fashionMNIST Classifier에 Convolution과 Pooling을 적용한 후 성능 증가를 직접 경험해보았다. 이제 다양한 model tuning과 epochs를 조절하면서 소폭의 성능 증가에 기여할 수 있는 지 알아보도록 해보자! 👔 모든 코드는 GitHub 클릭! 👔 default fashionMNIST Classifier 모델: 2개 연산 → Flatten() → Dense(128, relu) → Dense(10, softmax)① Adjusting the Number of Epochs👔 epochs란 주어진 batch_size에 맞는 여러 batch를 미리 만들고 난 후, 동일 batch training을 반복적으로 몇 번 수행하는 가이다. 예를 들어 5 epo.. Deep Learning/Experiments 2024. 8. 2.
👔Improving fashionMNIST Classifier using Convolutions 👔 앞서 만들었던 DNN fashionMNIST Classifier의 정확도를 Convolution & Pooling을 사용해서 높이려 한다 👔 fashionMNIST Classifierbefore building a classification model ① check the version & load fashionMNIST data & load the training, test split of the fashionMNIST dataset : fashionMNIST dataset is a collection of grayscale 28x28 pixel clothing images. Each image is associated withsh-avid-learner.tistory.com① load the .. Deep Learning/Experiments 2024. 6. 2.
🌸Model Analysis - Convolution / number of Pooling params + shape size ① case 1# Define the modelmodel = tf.keras.models.Sequential([ # Add convolutions and max pooling tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(28, 28, 1)), tf.keras.layers.MaxPooling2D(2, 2), tf.keras.layers.Conv2D(32, (3,3), activation='relu'), tf.keras.layers.MaxPooling2D(2,2), # Add the same layers as before .. Deep Learning/Fundamentals 2024. 5. 12.
👔 fashionMNIST Classifier before building a classification model ① check the version & load fashionMNIST data & load the training, test split of the fashionMNIST dataset : fashionMNIST dataset is a collection of grayscale 28x28 pixel clothing images. Each image is associated with a label as shown in this table import tensorflow as tf print(tf.__version__) #2.15.0 # Load the Fashion MNIST dataset fmnist = tf.keras.dataset.. Deep Learning/Experiments 2024. 1. 15.
C1W2 - Logistic Regression as a Neural Network 1. Binary Classification ★ Logistic Regression is an algorithm for binary classification ex) an input of an image(represented in 3 RGB matrices: 3 x 64 x 64; if 64 x 64 pixels) → output 1 (cat) vs 0 (non cat) ★ in binary classification, our goal is to learn a classifier(x → y), to predict whether this is a cat image or non-cat image(whether y is 1 or 0) ★ Notation: ① A single training example is.. Deep Learning/Fundamentals 2023. 9. 6.

티스토리툴바