d2l-en · Issues· 196 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #2734
Add a readable original-paper link to the BERT chapter?
Updated Aug 30, 2026 - #2733
register link not arriving in gmail email
Updated Aug 25, 2026 - #2669
conv-layer.ipynb
Updated Aug 1, 2026 - #2678
Upgrade `numpy` (v1.23.5) to prevent depricated `distutils` error in Python 3.12+
Updated Jul 24, 2026 - #2728
17.2.2 - Value Function - Clarify value function decomposition in reinforcement learning
Updated Jul 4, 2026 - #2076
MaskedSoftmaxCELoss code wrong
Updated Jun 24, 2026 - #2714
Tar-Slip Arbitrary File Write via HTTP Download in d2l
Updated Jun 10, 2026 - #2649
Error in chapter 17, section 3 "Q-learning", equation 17.3.3 and 17.3.4.
Updated Jun 1, 2026 - #2681
Epub version
Updated May 6, 2026 - #2692
Can we have a lock file for d2l?
Updated Apr 28, 2026 - #2684
LBD
Updated Mar 30, 2026 - #2573
Incorrect Use of torch.no_grad() in fit_epoch Method in d2l/torch.py::Trainer::fit_epoch
Updated Feb 24, 2026 - #2668
Question about the <unk> in the subsample function of chapter_natural-language-processing-pretraining
Updated Jan 29, 2026 - #2679
German translation: object-before-schon sounds more natural
Updated Jan 27, 2026 - #2603
Addition of dark mode
Updated Jan 17, 2026