F8 Refresh | Deep Learning with PyTorch

on YouTube ↗17m

" summary

Live demo builds a model in PyTorch that trains on Fashion‑MNIST with SGD and NLLLoss, logs to TensorBoard in a Colab/Jupyter notebook and gets useful predictions after one epoch. A separate project uses PyTorch to train agents in Habitat 2.0, claiming a simulator that yields an 80 years equivalent of robot experience. Project stats: ~71,000 downstream GitHub projects, ~1,100 arXiv papers mentioning PyTorch and ~64% year‑over‑year growth, and over 60% share on Papers With Code.

AI-written summary · about

" chapters (5)

1. 0:00Introduction
2. 0:33Agenda
3. 4:40Community
4. 13:50Habitat
5. 15:58Conclusion

" related talks

• 01h 40mA Deep Dive into NLP with PyTorchJeffrey Hsu, Susannah Klanecek · PyData London 2019
• 01h 32mIntroduction to PyTorchRobert Alvarez · PyData Miami 2019
• 33mBuild Better AI with First-Party Data and Fine-TuningPaul Yang · AI By the Bay 2025
• 01h 05mDeep Learning with PyTorchSeth Juarez · NDC London 2019
• 43mPyTorch Technical Deep DivePeng Wu, Alban Desmaison, Mark Saroufim, Edward Yang · PyTorch Conference 2025
• 19mBeat GPT 4 & Build the Best AI For YouPaul Yang · PyTorch Day France 2025
• 47mDeep Learning with PyTorchSachin Solkhan · All Things Open 2020
• 54mKeynote: PyTorch 2.1 Technical Deep DiveJoe Spisak, Peng Wu, Mergen Nachin, Mark Saroufim, Yanan Cao, Will Constable, Mario Lezcano · PyTorch Conference 2023