Kursus
Deep Learning dengan PyTorch
Dari satu neuron hingga transformer: bangun, latih, dan debug jaringan saraf dengan PyTorch, setiap pelajaran disertai kode yang berjalan di laptop.
- Pelajaran
- 14
- Waktu baca
- 3 jam 25 menit
- Level
- Menengah
- Bahasa pengantar
- Inggris
- Format
- Artikel dengan kode yang bisa dijalankan
Kursus
Yang akan Anda pelajari
Empat belas pelajaran artikel dalam empat bagian. Fondasi: tensor dan autograd, jaringan saraf tiruan, dan melatihnya dengan baik menggunakan optimizer, dropout, dan batch normalization. Visi: jaringan konvolusi, arsitektur klasik, dan transfer learning. Sekuens dan bahasa: jaringan rekuren, LSTM dan GRU, word embedding, attention, dan transformer. Model generatif: autoencoder, GAN, dan neural style transfer. Setiap contoh cukup kecil untuk dilatih di CPU.
- Foundations
- Computer vision
- Sequences and language
- Generative models
Sebelum mulai
Komputer yang bisa menjalankan Python. Setiap pelajaran menyebutkan apa yang menjadi dasarnya.
Silabus
Silabus
Setiap pelajaran menjelaskan satu gagasan, lalu menjalankannya dalam kode beserta output-nya. Pelajaran pertama terbuka untuk semua orang.
01 · Foundations
- 01Tensors and autogradPyTorch tensors, shapes and devices, and automatic differentiation: how gradients are computed and a model learns by gradient descent.Pratinjau gratis15 mnt
- 02Artificial neural networksFrom a neuron to a multilayer perceptron: layers, activations, the forward pass, backpropagation and the training loop.14 mnt
- 03Training neural networks wellLosses, SGD, momentum and Adam, learning-rate schedules, initialisation, dropout, batch normalisation, weight decay and early stopping.15 mnt
02 · Computer vision
- 04Convolutional neural networksConvolutions, filters, padding, stride and pooling: why CNNs see images better than dense layers, built from scratch on small images.13 mnt
- 05CNN architecturesLeNet, AlexNet, VGG, Inception and ResNet: the ideas each one added, residual connections, and data augmentation.14 mnt
- 06Transfer learningReuse a network trained on millions of images: feature extraction, fine-tuning, freezing layers and adapting to a small dataset.15 mnt
03 · Sequences and language
- 07Recurrent neural networksProcess sequences step by step with a hidden state, unroll through time, and see why plain RNNs forget.15 mnt
- 08LSTM and GRUGates that decide what to keep and forget: LSTM and GRU cells, stacking and bidirectional layers, and forecasting a time series.15 mnt
- 09Word embeddingsTurn words into vectors: tokenisation, one-hot versus dense embeddings, word2vec, GloVe, similarity and analogies.15 mnt
- 10Sequence-to-sequence and attentionEncoder–decoder models for translation, the bottleneck they hit, and attention that lets the decoder look back at every input.14 mnt
- 11The transformerSelf-attention, multi-head attention, positional encoding and the transformer block, built in PyTorch, then BERT and GPT in context.15 mnt
04 · Generative models
- 12AutoencodersCompress and reconstruct: undercomplete, denoising and variational autoencoders, the latent space, and anomaly detection.15 mnt
- 13Generative adversarial networksA generator and a discriminator in a contest: the GAN objective, DCGAN, training instability and mode collapse.15 mnt
- 14Neural style transferPaint one image in the style of another: content and style losses from a pretrained CNN, Gram matrices, and optimising the pixels.15 mnt
Masukan
Bantu kami menyempurnakan kursus ini
Usulkan pelajaran, tunjukkan bagian yang kurang jelas, atau laporkan kesalahan pada kode. Setiap pesan dibaca oleh tim kami dan dibalas lewat email.