Lädt...

🔧 Demystifying loss.backward(): How PyTorch Autograd Actually Works


Nachrichtenbereich: 🔧 Programmierung
🔗 Quelle: dev.to

If you use PyTorch, there’s one line of code you probably type out of sheer muscle memory: loss.backward().

When you first learn neural networks, you spend hours deriving backpropagation equations... [Weiterlesen]

💾 trunk/46931e0e655d691efc0bf431ab0fe920e7db871a: Remove plain asserts in torch/_dynamo/source.py (#182171)


📈 577.43 Punkte
💾 Downloads

🔧 Demystifying loss.backward(): How PyTorch Autograd Actually Works


📈 500.31 Punkte
🔧 Programmierung

💾 ciflow/trunk/182326: Update on "Add tearDown checks for torch function state leaks"


📈 412.45 Punkte
💾 Downloads

🔧 Shai-Hulud Malware in PyTorch Lightning: What Actually Happened and How to Check Your Environment


📈 316.21 Punkte
🔧 Programmierung

💾 ciflow/torchtitan/186020: Reintroduce ptxas bundling for CUDA 13.0 binaries


📈 288.71 Punkte
💾 Downloads

💾 trunk/826ab493be5714d267a00badb3773e0d9b78e00f: [Dynamo] Remove has_unpack_var_sequence (#185153)


📈 261.22 Punkte
💾 Downloads

🔧 Convert Tensorflow to Pytorch Model. in 2025?


📈 261.22 Punkte
🔧 Programmierung

🔧 How I Achieved 2.9 PyTorch Training Speedup with One Line of Code


📈 261.22 Punkte
🔧 Programmierung

💾 trunk/9c77ff51b6b2a30dc0964289e54c55b3f1bcdf13: Fix functional accumulate_grad state modeling (#184082)


📈 240.86 Punkte
💾 Downloads

🔧 AWS re:Invent 2025 - Break through AI performance and cost barriers with AWS Trainium (AIM201)


📈 233.72 Punkte
🔧 Programmierung

🔧 Mastering PyTorch: Tensors, Autograd & the Power of GPU Training


📈 225.17 Punkte
🔧 Programmierung

🔧 autograd-cpp


📈 221.03 Punkte
🔧 Programmierung

🔧 TensorFlow vs PyTorch: The Real Difference Isn't Accuracy


📈 213.1 Punkte
🔧 Programmierung

💾 viable/strict/1780178955: Use SVE `exp_u20`/`fexp_u20` for SVE128 (#184341)


📈 206.22 Punkte
💾 Downloads

💾 trunk/eaa777900b14e1aa6ffc1959cbcdb055d5de3e75: Update third_party/kineto submodule to 7739225 (#182973)


📈 206.22 Punkte
💾 Downloads

🔧 PyTorch vs TensorFlow 2025: Which one wins after 72 hours?


📈 199.35 Punkte
🔧 Programmierung

🔧 Terkuak! Pertarungan Sengit TensorFlow vs PyTorch: Siapa Sebenarnya Raja AI yang Wajib Anda Kuasai?


📈 199.35 Punkte
🔧 Programmierung

🔧 numr 0.5.0: The Rust numerical computing library that doesn't make you choose


📈 197.94 Punkte
🔧 Programmierung

🔧 The Great AI Convergence: PyTorch vs. TensorFlow in 2026


📈 185.6 Punkte
🔧 Programmierung

🔧 Qlib - 模型训练和预测(Model Training & Prediction)


📈 178.73 Punkte
🔧 Programmierung

🔧 Project of the Week: PyTorch


📈 178.73 Punkte
🔧 Programmierung

🔧 AxonML -- A PyTorch-equivalent ML framework written in Rust


📈 174.58 Punkte
🔧 Programmierung

💾 trunk/60137ad76df228e3e8636d67162109c1ef4a7e04: [ROCm] Enable external events in CUDA graphs (#178264)


📈 171.85 Punkte
💾 Downloads

🔧 How to Prune Networks in Pytorch in 2025?


📈 171.85 Punkte
🔧 Programmierung

💾 trunk/1b392e4a7dde975322e568643c639e7f71df6594: [xpu][test] Fix XPU CI failure (#185895)


📈 164.98 Punkte
💾 Downloads

💾 trunk/6578ad55c85f53c677f28c46021b4e93483c482b: Re-enable triton XPU Python 3.15 Linux build (#185094)


📈 164.98 Punkte
💾 Downloads