Where to Start with Machine Learning
summary
vTime demonstrates a working pipeline that turns a photo into a vtag avatar using an image-to-image deep model trained on freely available datasets and GitHub code, producing usable results in about a day. Training needs GPUs with 8–10 GB minimum and the recommended cards are RTX 3070/3080/3090, a home build costs ~£2k–£2.5k versus AWS becoming costlier after 1–4 months. Notable data warning: the CelebA 200k dataset contains biased labels (e.g. an “attractive” tag) that alter outputs like skin tone.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:50What is Machine Learning
- 3:54Machine Learning in Gaming
- 6:25Hardware vs Cloud
- 10:41Free Resources
- 13:18What to Expect
- 17:56Legal Ethical Concerns
related talks
38mReal-world Machine Learning with TensorFlow and Cloud AI
28mChallenges & future trends of Deep Learning
40mHow to Implement a Production Ready Deep Learning Model in 12 Weeks
33mThe Role of Software Developers in Machine Learning
40mMachines are Learning
57mBuilding smarter solutions with Machine Learning, from magic to reality
44mGetting started with Azure Machine Learning services
21mPowerAI Vision & Other Solutions on an AI Box