Pip Install Driven Deep Learning
summary
Practical guide to assembling deep-learning systems using pip-installable tools: OpenCV cascades for fast multi-scale face detection (~230 ms on 1024×768), dlib/face_recognition for identity matching (~500 ms, 300‑dim embeddings), pretrained Keras image models for quick inference, YOLO/darknet for object localization, spaCy for tokenization/NER and 300‑dim word/sentence vectors, and DVC for dataset/version pipelines.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 2:50ML without training
- 6:28Face recognition OpenCV
- 10:23Face detection face_recognition
- 12:32Embeddings in neural nets
- 16:53Face comparison embeddings
- 18:19Object detection Keras
- 29:09YOLO localization
- 32:02Text processing spaCy
- 37:08Summary Q&A
related talks
39mBuilding Highly Scalable Facial Recognition Pipelines
30mThe Art and Science of Deep Face Recognition in Python
01h 12mBuilding a live face recognition system
54mDeep Learning with Apache Spark
34mDeep Learning at Scale
47mDeep Learning in Computer Vision
22mBringing Deep Learning into the Big Data Analytics
37mData Science for Digital Humanities: Extracting meaning from Images and Text