GPU-optimized AI, Machine Learning, & HPC Software

NVIDIA

ASR + BERT QA interactive chatbot demo for Jetson

Container

NVIDIA

Text Classification with BERT and NeMo. This NeMo application trains text classification models using single-GPU or multi-GPU. We log performance metrics and visualize them with TensorBoard. We show how to do inference with NeMo, and we visualize BERT embeddings before and after fine-tuning.

Container

NVIDIA

BERT for TensorFlow Jupyter Notebook

Jupyter Notebook example for Question Answering with BERT for TensorFlow

Resource

NVIDIA

DLI NLP Course - Base Environment with NeMo

Base environment used in the NVIDIA NeMo projects of the NVIDIA Deep Learning Institute (DLI) course, "Building Transformer-Based Natural Language Processing Applications". This container also includes a "Next Steps" project.

Container

NVIDIA

Fine-Tune and Optimize BERT

Jupyter Notebooks for BERT Pre-training, Fine-Tuning and Inference profiling and optimization via TensorFlow, AMP, XLA, DLProf, TF-TRT and Triton.

Container

DeepPavlov

DeepPavlov is an open-source conversational AI library built on TensorFlow and Keras. DeepPavlov is designed for development of production ready chatbots and complex conversational systems, research in the area of NLP and, particularly, of dialog systems.

Container

NVIDIA

Domain Classification English Bert

Domain classification of the query for weather chat bot.

Model

NVIDIA

BERT on Google Cloud AI Platform

Fine-tune a pre-trained BERT model with the SQuAD dataset, optimize for inference using TensorRT and deploy with Triton Inference Server on Google Cloud AI Platform using Custom Containers

Resource

NVIDIA

Bertlargeuncased

BERT Large Uncased trained on English Wikipedia and BookCorpus

Model

NVIDIA

BioMegatron345mUncased

Megatron pretrained on uncased biomedical dataset PubMed with 345 million parameters.

Model

NVIDIA

Bertbaseuncased

BERT Base Uncased trained on English Wikipedia and BookCorpus

Model

NVIDIA

Question Answering Notebook

End to End workflow for question answering starting with training in TAO Toolkit and deployment using Riva.

Resource

NVIDIA

RIVA Punctuation and Capitalization for English

For each word in the input text, the model: 1) predicts a punctuation mark that should follow the word (if any), the model supports commas, periods and question marks) and 2) predicts if the word should be capitalized or not.

Model

NVIDIA

Text Classification Notebook

End to End sample workflow for Text Classification starting with training in TAO Toolkit and deployment using Riva.

Resource

NVIDIA

RIVA Punctuation

For each word in the input text, the model: 1) predicts a punctuation mark that should follow the word (if any), the model supports commas, periods and question marks) and 2) predicts if the word should be capitalized or not.

Model

NVIDIA

Punctuation En Bert

Punctuation and Capitalization model with BERT

Model

NVIDIA

BertBaseUncasedForNemo

BERT Base Model trained on uncased Wikipedia and BookCorpus dataset on a sequence length of 512.

Model

NVIDIA

NER En Bert

Named Entity Recognition model with BERT

Model

—

BioMegatron345mCased

Megatron pretrained on cased biomedical dataset PubMed with 345 million parameters.

Model

NVIDIA

BioBERTLargeCasedForNeMo

BioBERT-large cased model checkpoint for NeMo.

Model

NVIDIA

BertLargeUncasedForNemo

BERT Large Model trained with NeMo on uncased Wikipedia and Bookcorpus on a sequence length 512.

Model

NVIDIA

Question Answering SQUAD2.0 Bert

Question Answering Bert Base uncased model for extractive question answering on any provided content.

Model

NVIDIA

Joint Intent and Slot Classification Bert

Intent and Slot classification of the qeuries for the weather chat bot (trained on weather chat bot data).

Model

NVIDIA

BioMegatron345mUncasedSQuADv1

BioMegatron 345M uncased model for Question Answering finetuned with NeMo on SQuAD v1.1 dataset.

Model