Natural Language Processing With PyTorch: Build Intelligent Language Applications Using Deep Learning
Description
mentions TF-IDF, interestingly (ref:4.33-6)
Downside is that this book doesn't abide by the KISS principle, with all its object-oriented line-noise…
This book has the best diagrams explaining convolution (CNNs).
Natural Language Processing (NLP) provides boundless opportunities for solving problems in artificial intelligence, making products such as Amazon Alexa and Google Translate possible. If you’re a developer or data scientist new to NLP and deep learning, this practical guide shows you how to apply these methods using PyTorch, a Python-based deep learning library.
Authors Delip Rao and Brian McMahon provide you with a solid grounding in NLP and deep learning algorithms and demonstrate how to use PyTorch to build applications involving rich representations of text specific to the problems you face. Each chapter includes several code examples and illustrations.
Explore computational graphs and the supervised learning paradigmMaster the basics of the PyTorch optimized tensor manipulation libraryGet an overview of traditional NLP concepts and methodsLearn the basic ideas involved in building neural networksUse embeddings to represent words, sentences, documents, and other featuresExplore sequence prediction and generate sequence-to-sequence modelsLearn design patterns for building production NLP systems
KOReader notes:
Categorical Cross-Entropy Loss
01/19/25Categorical Cross-Entropy Loss : 80
Example 3-8. Cross-entropy loss
Input[0]
import torch
import torch.nn as nn
ce_loss = nn.CrossEntropyLoss()
outputs = torch.randn(3, 5, requires_grad=True)
targets = torch.tensor([1, 0, 3], dtype=torch.int64)
loss = ce_loss(outputs, targets)
print(loss)
Output[0]
tensor(2.7256)
The Vocabulary, the Vectorizer, and the DataLoader
01/02/25The Vocabulary, the Vectorizer, and the DataLoader : 108
Example 3-17, we wrap the DataLoader in a generate_batches() function, which is a generator to conveniently switch the data between the CPU and the GPU.
The Practical Use of Pretrained Word Embeddings
01/09/25The Practical Use of Pretrained Word Embeddings : 197
subtract Word2> from Word1. This difference vector encodes the relationship between Word1 and Word2. This difference can then be added to Word3 to produce a vector that’s close to the fourth word, where the blank symbol is
Design Patterns for Production NLP Systems
01/22/25Design Patterns for Production NLP Systems : 336
multimodal systems
analogous to the sensus communis? 🤔