Published inPyTorch Lightning Developer BlogLightning Transformers 0.2 — New 🤗Tasks, Community Features and Big Model Training & InferencePairing 🤗 Transformers and Lightning has become increasingly popular, leveraging Lightning to hide away the boilerplate of your training…Nov 29, 2022Nov 29, 2022
Published inPyTorch Lightning Developer BlogPart I: Simplifying Transformer Research with xFormers & LightningRecently we’ve seen a large growth in variations of the Transformer model (Efficient Transformers: A Survey). However, leveraging improved…Dec 9, 20211Dec 9, 20211
Published inPyTorch Lightning Developer BlogLeverage Sparsity for Faster Inference with Lightning Flash and SparseMLSparseML brings GPU inference speeds to the CPU. This means substantial cost-saving, efficiency, and more options when it comes to…Sep 21, 2021Sep 21, 2021
Published inPyTorch Lightning Developer BlogFine-tune Transformers Faster with Lightning Flash and Torch ORTTorch ORT uses the ONNX Runtime to improve training and inference times for PyTorch models.Aug 26, 2021Aug 26, 2021
Published inPyTorch Lightning Developer BlogFine-tuning Wav2Vec for Speech Recognition with Lightning FlashAs a result of our recent Lightning Flash Taskathon, we introduced a new fine-tuning task backed by HuggingFace Wav2Vec, powered by PyTorch…Jul 26, 20214Jul 26, 20214
Published inPyTorchIntroducing PyTorch Lightning Sharded: Train SOTA Models, With Half The MemoryLightning 1.1 reveals Sharded Training — train deep learning models on multiple GPUs saving over 50% on memory, with no performance loss…Dec 10, 2020Dec 10, 2020
Published inTowards Data ScienceTrain Conversational AI in 3 lines of code with NeMo and LightningTrain state-of-the-art speech recognition, NLP and TTS models at scale with NeMo and LightningOct 28, 2020Oct 28, 2020
Published inPyTorchTraining DeepSpeech using TorchElasticReduce cost and horizontally scale deepspeech.pytorch using TorchElastic with Kubernetes.Aug 10, 2020Aug 10, 2020
CORD-19-ANN: Semantic Search Engine Using S-BERTAllen Institute as a part of their open research efforts released a data dump of scholarly articles as an initiative to aid efforts in…Apr 20, 20201Apr 20, 20201
Scaling DeepSpeech using Mixed Precision and KubeFlowOver the past few years at Digital Reasoning we have been developing audio analytics software to be highly effective at processing the…Jun 13, 20191Jun 13, 20191