[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10,"_11":12},"actionData","errors","root",{"_88":251,"_252":253,"_254":255,"_256":257,"_258":259,"_260":31,"_261":262,"_263":264,"_265":31,"_266":31,"_267":31,"_268":31,"_269":270,"_271":-5,"_272":273,"_274":275,"_276":277,"_278":279,"_280":115,"_281":117,"_282":283,"_109":-5,"_284":31,"_285":-5,"_286":287,"_288":289,"_138":290,"_291":292,"_293":294,"_295":296,"_297":298,"_299":31,"_26":27,"_28":300,"_30":31,"_32":301,"_34":302,"_36":303},"routes/orgs/$orgName/$teamName/$resourceType/$resourceId/$version/route",{"_124":125,"_126":127,"_128":129,"_130":16,"_131":31,"_132":133,"_134":31,"_135":31,"_136":137,"_138":21,"_139":-5,"_140":141,"_142":-5,"_143":144,"_145":146,"_147":31,"_148":149,"_26":27,"_28":150,"_30":31,"_32":151,"_34":152,"_36":153},"routes/orgs/$orgName/$teamName/$resourceType/$resourceId/$version/_access/route",{"_85":86,"_87":-5,"_88":89,"_90":91,"_26":27,"_28":92,"_30":31,"_32":93,"_34":94,"_36":95},"routes/orgs/$orgName/$teamName/$resourceType/$resourceId/$version/file-browser/route",{"_13":14,"_15":16,"_17":16,"_18":19,"_20":21,"_22":23,"_24":25,"_26":27,"_28":29,"_30":31,"_32":33,"_34":35,"_36":37},"fileTree",[72],"totalFiles",1,"page","pageSize",50,"unresolvedVersion","-","resolvedVersion","20.07.0","expandedFolders",[],"traceId","3411db8ca2f05b5e12046709237c4927","traceStartTime",1786122320976,"isStaleResponse",false,"notifications",[],"toasts",[],"apiCalls",[38,39,40,41,42,43,44],{"_45":59,"_47":70,"_49":61,"_51":62,"_53":71,"_55":56},{"_45":59,"_47":68,"_49":61,"_51":62,"_53":69,"_55":56},{"_45":59,"_47":66,"_49":61,"_51":62,"_53":67,"_55":56},{"_45":59,"_47":64,"_49":61,"_51":62,"_53":65,"_55":56},{"_45":59,"_47":60,"_49":61,"_51":62,"_53":63,"_55":56},{"_45":57,"_47":58,"_49":50,"_51":52,"_53":54,"_55":56},{"_45":46,"_47":48,"_49":50,"_51":52,"_53":54,"_55":56},"apiTag","subscription-api","path","/v2/users/me/subscriptions","status","error","httpStatus",401,"durationMs",3,"retryCount",0,"oms-api","/v2/users/me","models-api","/v2/resources/org/nvidia/team/dle/electra_tf2/collections?page-number=0&page-size=1","success",200,52,"/v2/resources/nvidia/dle/electra_tf2/versions/20.07.0/files?page-size=50&page-number=0",61,"/v2/resources/nvidia/dle/electra_tf2/versions/20.07.0",58,"/v2/resources/nvidia/dle/electra_tf2/versions?page-size=100&page-number=0&sort-order=CREATED_DATE_DESC",73,"/v2/resources/nvidia/dle/electra_tf2?resolve-labels=true&remove-unresolved-labels=true",83,{"_73":74,"_74":75},"type","file",{"_76":77,"_78":77,"_47":77,"_79":80,"_81":82,"_83":84},"id","electra_tf2.md","name","size",64150,"formattedSize","62.65 KB","updatedAt","2022-09-22T14:33:54.305Z","canDownload",true,"accessDeniedReason","appConfig",{"_107":108,"_109":110},"debugAccessInfo",{"_103":86,"_104":105,"_106":31},1786122320904,[],[],[96,97,98,99,100,101],{"_45":59,"_47":70,"_49":61,"_51":62,"_53":71,"_55":56},{"_45":59,"_47":68,"_49":61,"_51":62,"_53":69,"_55":56},{"_45":59,"_47":66,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":59,"_47":60,"_49":61,"_51":62,"_53":63,"_55":56},{"_45":57,"_47":58,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":46,"_47":48,"_49":50,"_51":52,"_53":54,"_55":56},"pending","canGuestAccess","requiredProducts",[],"hasLicenseRequirement","urls",{"_112":113,"_114":115,"_116":117,"_118":119,"_120":121,"_122":123},"user",{"_111":31},"loggedIn","build","https://build.nvidia.com","ngcApi","https://api.ngc.nvidia.com","ngcHelm","https://helm.ngc.nvidia.com","ngcRegistry","https://registry.ngc.nvidia.com","orgUi","https://org.ngc.nvidia.com","ngcUi","https://ngc.nvidia.com","resource",{"_76":199,"_200":201,"_202":203,"_204":205,"_161":162,"_163":164,"_206":207,"_167":208,"_209":210,"_172":211,"_212":213,"_214":23,"_215":-7,"_216":-7,"_217":218,"_219":220,"_221":86,"_222":80,"_223":-5,"_224":-7,"_225":-7,"_226":-7,"_227":-7,"_228":-7,"_229":-7,"_230":231,"_232":233,"_234":235,"_236":237,"_238":239,"_240":31},"html","

This resource is using open-source code maintained in github (see the quick-start-guide section) and available for download from NGC

Electra (Efficiently Learning an Encoder that Classifies Token Replacements Accurately), is a novel pre-training method for language representations which outperforms existing techniques, given the same compute budget on a wide array of Natural Language Processing (NLP) tasks. This model is based on the ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators paper. NVIDIA's implementation of ELECTRA is an optimized version of the Hugging Face implementation, leveraging mixed precision arithmetic and Tensor Cores on Volta, Turing, and the NVIDIA Ampere GPU architectures for faster training times with state-of-the-art accuracy.

This repository contains the scripts to interactively launch data download, training, benchmarking and inference routines in a Docker container for pre-training on your own dataset (Wikipedia and BookCorpus shown as an example), and fine-tuning for tasks such as question answering. The major differences between the original implementation as described in the paper and this version of ELECTRA are as follows:

Scripts to download Wikipedia and BookCorpus datasets
Scripts to preprocess downloaded data or a custom corpus into inputs and targets for pre-training in a modular fashion
Automatic mixed precision (AMP) support and optimized for performance
Multi-GPU and Multi-node training support with push-button scripts to reach state-of-the-art accuracy and performance.

Other publicly available implementations of Electra include:

Hugging Face
Google's implementation

This model is trained with mixed precision using Tensor Cores on Volta, Turing, and the NVIDIA Ampere GPU architectures. Additionally, this model provides push-button solutions to pre-training, fine-tuning and inference and on a corpus of choice. As a result, researchers can get results up to 4x faster than training without Tensor Cores. This model is tested against each NGC monthly container release to ensure consistent accuracy and performance over time.

Model architecture

ELECTRA is a combination of two Transformer models: a generator and a discriminator. The generator's role is to replace tokens in a sequence, and is therefore trained as a masked language model. The discriminator, which is the model we are interested in, tries to identify which tokens were replaced by the generator in the sequence. Both generator and discriminator use the same architecture as the encoder of the Transformer. The encoder is simply a stack of Transformer blocks, which consist of a multi-head attention layer followed by successive stages of feed-forward networks and layer normalization. The multi-head attention layer performs self-attention on multiple input representations.

$\"Figure$

Default configuration

ELECTRA uses a new pre-training task called replaced token detection (RTD), that trains a bidirectional model (like a MLM) while learning from all input positions (like a LM). Inspired by generative adversarial networks (GANs), instead of corrupting the input by replacing tokens with \"[MASK]\" as in BERT, the generator is trained to corrupt the input by replacing some input tokens with incorrect, but somewhat plausible, fakes. On the other hand, the discriminator is trained to distinguish between \"real\" and \"fake\" input data.

The Google ELECTRA repository reports the results for three configurations of ELECTRA, each corresponding to a unique model size. This implementation provides the same configurations by default, which are described in the table below.

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n

tr:nth-child(odd)]:bg-surface-sunken\">

Model	Hidden layers	Hidden unit size	Parameters
ELECTRA_SMALL	12 encoder	256	14M
ELECTRA_BASE	12 encoder	768	110M
ELECTRA_LARGE	24 encoder	1024	335M

The following features were implemented in this model:

\n
General:
\n
- Mixed precision support with TensorFlow Automatic Mixed Precision (TF-AMP)
- Multi-GPU support using Horovod
- XLA support
- Multi-Node support
\n
\n
Training
\n
- Pre-training support
- Fine-tuning example
\n
\n
Inference:
\n
- Joint predictions with beam search.
\n

Feature support matrix

The following features are supported by this model.

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n

tr:nth-child(odd)]:bg-surface-sunken\">

Feature	ELECTRA
LAMB	Yes
Automatic mixed precision (AMP)	Yes
XLA	Yes
Horovod Multi-GPU	Yes
Multi-node	Yes

Features

Automatic Mixed Precision (AMP)

This implementation of ELECTRA uses AMP to implement mixed precision training. It allows us to use FP16 training with FP32 master weights by modifying just a few lines of code.

Horovod

Horovod is a distributed training framework for TensorFlow, Keras, PyTorch, and MXNet. The goal of Horovod is to make distributed deep learning fast and easy to use. For more information about how to get started with Horovod, see the Horovod: Official repository.

Multi-GPU training with Horovod

Our model uses Horovod to implement efficient multi-GPU training with NCCL. For details, see example sources in this repository or see the TensorFlow tutorial.

XLA support (experimental)

XLA is a domain-specific compiler for linear algebra that can accelerate TensorFlow models with potentially no source code changes. The results are improvements in speed and memory usage: most internal benchmarks run ~1.1-1.5x faster after XLA is enabled.\nAMP is an abbreviation used for automatic mixed precision training.

Multi-node Training

Supported on a Pyxis/Enroot Slurm cluster.

Mixed precision training

Mixed precision is the combined use of different numerical precisions in a computational method. Mixed precision training offers significant computational speedup by performing operations in half-precision format, while storing minimal information in single-precision to retain as much information as possible in critical parts of the network. Since the introduction of Tensor Cores in Volta, and following with both the Turing and Ampere architectures, significant training speedups are experienced by switching to mixed precision -- up to 3x overall speedup on the most arithmetically intense model architectures. Using mixed precision training requires two steps:

Porting the model to use the FP16 data type where appropriate.
Adding loss scaling to preserve small gradient values.

This can now be achieved using Automatic Mixed Precision (AMP) for TensorFlow to enable the full mixed precision methodology in your existing TensorFlow model code. AMP enables mixed precision training on Volta, Turing, and NVIDIA Ampere GPU architectures automatically. The TensorFlow framework code makes all necessary model changes internally.

In TF-AMP, the computational graph is optimized to use as few casts as necessary and maximize the use of FP16, and the loss scaling is automatically applied inside of supported optimizers. AMP can be configured to work with the existing tf.contrib loss scaling manager by disabling the AMP scaling with a single environment variable to perform only the automatic mixed-precision optimization. It accomplishes this by automatically rewriting all computation graphs with the necessary operations to enable mixed precision training and automatic loss scaling.

For information about:

How to train using mixed precision, see the Mixed Precision Training paper and Training With Mixed Precision documentation.
Techniques used for mixed precision training, see the Mixed-Precision Training of Deep Neural Networks blog.
How to access and enable AMP for TensorFlow, see Using TF-AMP from the TensorFlow User Guide.

Enabling mixed precision

This implementation exploits the TensorFlow Automatic Mixed Precision feature. To enable AMP, you simply need to supply the --amp flag to the run_pretraining.py or run_tf_squad.py script. For reference, enabling AMP required us to apply the following changes to the code:

Set the Keras mixed precision policy:

if config.amp:\n    policy = tf.keras.mixed_precision.experimental.Policy(\"mixed_float16\", loss_scale=\"dynamic\")\n    tf.keras.mixed_precision.experimental.set_policy(policy)\n

Use the loss scaling wrapper on the optimizer:

if config.amp:\n     optimizer = tf.keras.mixed_precision.experimental.LossScaleOptimizer(optimizer, \"dynamic\")\n

Use scaled loss to calculate the gradients:

#Scale loss\nif config.amp:\n    total_loss = optimizer.get_scaled_loss(total_loss)\ngradients = tape.gradient(total_loss, model.trainable_variables)\n#Get unscaled gradients if AMP\nif config.amp:\n    gradients = optimizer.get_unscaled_gradients(gradients)\n

Enabling TF32

TensorFloat-32 (TF32) is the new math mode in NVIDIA A100 GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs.

TF32 Tensor Cores can speed up networks using FP32, typically with no loss of accuracy. It is more robust than FP16 for models which require high dynamic range for weights or activations.

For more information, refer to the TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x blog post.

TF32 is supported in the NVIDIA Ampere GPU architecture and is enabled by default.

Glossary

Fine-tuning
\nTraining an already pretrained model further using a task specific dataset for subject-specific refinements, by adding task-specific layers on top if required.

Language Model
\nAssigns a probability distribution over a sequence of words. Given a sequence of words, it assigns a probability to the whole sequence.

Pre-training
\nTraining a model on vast amounts of data on the same (or different) task to build general understandings.

Transformer
\nThe paper Attention Is All You Need introduces a novel architecture called Transformer that uses an attention mechanism and transforms one sequence into another.

Phase 1\nPretraining on samples of sequence length 128 and at most 15% masked predictions per sequence.

Phase 2\nPretraining on samples of sequence length 512 and at most 15% masked predictions per sequence.

","resourceVersions",[195,196],"resourceVersionsTotal","hasMoreVersions","versionMetadata",{"_166":23,"_167":-5,"_49":168,"_169":-5,"_170":171,"_172":-5,"_173":80,"_174":16,"_175":31,"_176":-5,"_177":31,"_178":179,"_180":86,"_181":182,"_183":86,"_184":185,"_186":86,"_187":188,"_189":86,"_190":191,"_192":86,"_193":194},"versionNotFound","isLoggedIn","latestVersionPath","/orgs/nvidia/dle/resources/electra_tf2/-","version","imageReference","cliDownloadCommand","ngc registry resource download-version nvidia/dle/electra_tf2:20.07.0","helmFetchCommand","viewAllVersionsPath","/orgs/nvidia/dle/resources/electra_tf2/-/version-history","accessCheckResult",{"_165":86},"hasRelatedCollections","context",{"_161":162,"_163":164},1786122320966,[],[],[154,155,156,157,158,159,160],{"_45":59,"_47":70,"_49":61,"_51":62,"_53":71,"_55":56},{"_45":59,"_47":68,"_49":61,"_51":62,"_53":69,"_55":56},{"_45":59,"_47":66,"_49":61,"_51":62,"_53":67,"_55":56},{"_45":59,"_47":64,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":59,"_47":60,"_49":61,"_51":62,"_53":63,"_55":56},{"_45":57,"_47":58,"_49":50,"_51":52,"_53":54,"_55":56},{"_45":46,"_47":48,"_49":50,"_51":52,"_53":54,"_55":56},"orgName","nvidia","teamName","dle","allowed","versionId","description","UPLOAD_COMPLETE","releaseType","createdDate",["D",1663857232912],"updatedDate","totalSizeInBytes","totalFileCount","isSigned","malwareScanStatus","isScanned","releaseNotes","### Changelog\n \nJuly 2020\n- Initial release.\n \nOctober 2020\n- Data preparation scripts for pre-training.\n- Pre-training support.\n- Mixed precision support with Keras AMP policy.\n- Update beam size in SQuAD fine-tuning from 4 to 5 for higher accuracy.\n- T4 inference performance.\n \n### Known issues\n \nThere are no known issues with this model.","hasReleaseNotes","setup","The following section lists the requirements that you need to meet in order to start training the ELECTRA model.\n \n### Requirements\n \nThis repository contains Dockerfile which extends the TensorFlow2 NGC container and encapsulates some dependencies. Aside from these dependencies, ensure you have the following components:\n \n- [NVIDIA Docker](https://github.com/NVIDIA/nvidia-docker)\n- [TensorFlow2 20.07-py3 NGC container or later](https://ngc.nvidia.com/registry/nvidia-tensorflow)\n- Supported GPUs:\n - [NVIDIA Volta architecture](https://www.nvidia.com/en-us/data-center/volta-gpu-architecture/)\n - [NVIDIA Turing architecture](https://www.nvidia.com/en-us/geforce/turing/)\n - [NVIDIA Ampere architecture](https://www.nvidia.com/en-us/data-center/nvidia-ampere-gpu-architecture/)\n \nFor more information about how to get started with NGC containers, see the following sections from the NVIDIA GPU Cloud Documentation and the Deep Learning Documentation:\n- [Getting Started Using NVIDIA GPU Cloud](https://docs.nvidia.com/ngc/ngc-getting-started-guide/index.html)\n- [Accessing And Pulling From The NGC Container Registry](https://docs.nvidia.com/deeplearning/dgx/user-guide/index.html#accessing_registry)\n- [Running TensorFlow2](https://docs.nvidia.com/deeplearning/frameworks/tensorflow-release-notes/running.html#running)\n \nFor those unable to use the TensorFlow 2 NGC container, to set up the required environment or create your own container, see the versioned [NVIDIA Container Support Matrix](https://docs.nvidia.com/deeplearning/dgx/support-matrix/index.html).\n \nFor multi-node, the sample provided in this repository requires [Enroot](https://github.com/NVIDIA/enroot) and [Pyxis](https://github.com/NVIDIA/pyxis) set up on a [SLURM](https://slurm.schedmd.com) cluster.\n \nMore information on how to set up and launch can be found in the [Multi-node Documentation](https://docs.nvidia.com/ngc/multi-node-bert-user-guide).","hasSetup","quickStartGuide","To train your model using mixed precision or TF32 precision with Tensor Cores or using FP32, perform the following steps using the default parameters of the ELECTRA model. The default parameters for pre-training have been set to run on both 8x A100 40G and 8 x V100 32G GPUs. For the specifics concerning training and inference, see the [Advanced](#advanced) section.\n \n1. Clone the repository.\n \n```\ngit clone https://github.com/NVIDIA/DeepLearningExamples.git\ncd DeepLearningExamples/TensorFlow2/LanguageModeling/ELECTRA\n```\n \n2. Build ELECTRA on top of the NGC container.\n```\nbash scripts/docker/build.sh\n```\n \n3. Start an interactive session in the NGC container to run data download, training and inference.\n```\nbash scripts/docker/launch.sh\n```\n \nResultant logs of pre-training and fine-tuning routines are stored in the `results/` folder. Checkpoints are stored in the `results//` folder.\n \nRequired data is downloaded into the `data/` directory by default.\n \n4. Download and preprocess the dataset.\n \nThis repository provides scripts to download, verify, and extract the following datasets:\n \n- [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) (fine-tuning for question answering)\n- Wikipedia (pre-training)\n- BookCorpus (pre-training)\n \nTo download, verify, extract the datasets, and create the shards in `tfrecord` format, run:\n```\n/workspace/electra/data/create_datasets_from_start.sh\n```\n \nNote: For fine-tuning only, Wikipedia and Bookscorpus dataset download and preprocessing can be skipped by commenting it out.\n \n- Download Wikipedia only for pretraining\n \nThe pre-training dataset is 170GB+ and takes 15+ hours to download. The BookCorpus server most of the time gets overloaded and also contains broken links resulting in HTTP 403 and 503 errors. Hence, it is recommended to skip downloading BookCorpus data by running:\n \n```\n/workspace/electra/data/create_datasets_from_start.sh wiki_only\n```\n \n- Download Wikipedia and BookCorpus\n \nUsers are welcome to download the BookCorpus from other sources to match our accuracy, or repeatedly try our script until the required number of files are downloaded by running the following:\n```\n/workspace/electra/data/create_datasets_from_start.sh wiki_books\n```\n \nNote: Not using the BookCorpus can potentially change the final accuracy on a few downstream tasks.\n \n5. Start pretraining.\n \nTo run on a single node 8 x V100 32G, from within the container, you can use the following script to run pre-training.\n```\nbash scripts/run_pretraining.sh\n```\n \nThe default hyperparameters are set to run on both 8 x A100 40G and 8 x V100 32G.\n \nFor the other platforms, the configs present in `scripts/configs/pretrain_config.sh` can be used as shown below:\n```\nbash scripts/run_pretraining.sh $(source scripts/configs/pretrain_config.sh && dgxa100_8gpu_amp)\n```\n \nTo run pre-training on multiple nodes, see the [Multi-node](#multi-node) section.\n \n6. Postprocess pretrained checkpoint and fine-tune on SQuAD dataset\n \nThe above pretrained ELECTRA model representations can be fine-tuned with just one additional output layer for a state-of-the-art question answering system. Running the following script extracts and saves the discriminator and generator from the pretrained checkpoint and fine-tunes the discriminator on SQuAD:\n \n```\ncheckpoints=results/base/checkpoints bash scripts/finetune_ckpts_on_squad.sh\n```\n \nIt internally runs `postprocess_pretrained_ckpt.py` which extracts and saves the discriminator and the generator from the pretrained checkpoint.\n \nThe default hyperparameters are set to run on 8 x V100 16G.\n \nTo run fine-tuning with the SQuAD dataset on Google's pretrained checkpoints, do the following.\n```\nbash scripts/run_squad.sh\n```\n \nFor other platforms, configs present in `scripts/configs/squad_config.sh` can be used as shown below:\n```\nbash scripts/run_squad.sh $(source scripts/configs/squad_config.sh && dgxa100_8gpu_amp) train_eval\n```\n \n7. Start validation/evaluation.\n \nValidation can be performed by running:\n```\nbash scripts/run_squad.sh $(source scripts/configs/squad_config.sh && dgxa100_8gpu_amp) eval\n```\nRunning training first is required to generate needed checkpoints.\n \n8. Start inference/predictions.\n \nInference can be performed by running:\n```\nbash scripts/run_squad.sh $(source scripts/configs/squad_config.sh && dgxa100_8gpu_amp) prediction\n```\nInference predictions are saved to `/predictions.json`.","hasQuickStartGuide","advanced","The following sections provide greater details of the datasets, running training and inference, and the training results.\n \n### Scripts and sample code\n \nDescriptions of the key scripts and folders are provided below.\n \n- `data/` - Contains scripts for downloading and preparing individual datasets, and will contain downloaded and processed datasets.\n- `scripts/` - Contains shell scripts to launch the Docker container, data download, pre-training, fine-tuning and inference.\n- `results/` - Folder where all training and inference results get stored by default.\n- `run_squad.sh` - Interface for launching question answering fine-tuning with `run_tf_squad.py`.\n- `run_pretraining.sh` - Interface for launching ELECTRA pre-training with `run_pretraining.py`.\n- `finetune_ckpts_on_squad.sh` - Interface for extracting and saving discriminator and generator from the pretrained checkpoint and run SQuAD fine-tuning on discriminator.\n- `build_pretraining_dataset.py` - Creates `tfrecord` files from shared text files in the final step of dataset creation.\n- `postprocess_pretrained_ckpt.py` - Converts pretrained checkpoint to discriminator checkpoint and generator checkpoint which can be fed into `run_tf_squad.py`.\n- `modeling.py` - Implements the ELECTRA pre-training and fine-tuning model architectures with TensorFlow2.\n- `optimization.py` - Implements the Adam optimizer, LAMB and the learning rate schedule with TensorFlow2.\n- `configuration.py` - Implements parent class for model config.\n- `tokenization.py` - Implements the ELECTRA tokenizer.\n- `run_pretraining.py` - Implements ELECTRA pre-training.\n- `pretrain_utils.py` - Utilities required for pre-training such as dynamic masking etc.,\n- `run_tf_squad.py` - Implements fine-tuning training and evaluation for question answering on the [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) dataset.\n- `inference.py` - Implements interactive question answering.\n- `postprocess_pretrained_ckpt.py` - Implements extracting and saving the discriminator and the generator from the pretrained checkpoint.\n \n### Parameters\n \n#### Pre-training parameters\n \nELECTRA is designed to pre-train deep bidirectional networks for language representations. The following scripts replicate pre-training on Wikipedia + BookCorpus from this [paper](https://openreview.net/forum?id=r1xMH1BtvB). These scripts are general and can be used for pre-training language representations on any corpus of choice.\n \nIn the parameters expected by `scripts/run_pretraining.sh`, `p1` stands for phase 1 whereas `p2` stands for phase 2 training. They are as follows:\n \n- `` is per-GPU batch size used for training. Larger batch sizes run more efficiently, but require more GPU memory. Default is 176.\n- `` is the base learning rate for training. Default is 6e-3.\n- `` is the type of math in your model, can be either `fp32` or `amp`. Default is `amp`. The options mean:\n - FP32: 32-bit IEEE single precision float format.\n - AMP: Automatic mixed precision 16 and 32-bit float format.\n- `` is the number of GPUs to use for training. Must be equal to or smaller than the number of GPUs attached to your node. Default is 8.\n- `` is the percentage of training steps used for warm-up at the start of training. Default is 2000.\n- `` is the total number of training steps. Default is 10000.\n- `` controls how often checkpoints are saved. Default is 500.\n- `` if set to `true`, training should resume from the latest model in `/results/checkpoints`. Default is `false`.\n- `` a flag indicating whether a larger batch should be simulated with gradient accumulation. Default is `true`.\n- `` an integer indicating the number of steps to accumulate gradients over. Effective batch size / GPU = `training_batch_size` x `gradient_accumulation_steps`. Default is 48.\n- `` random seed for the run.\n \n- `` is per-GPU batch size used for training in phase 2. Larger batch sizes run more efficiently, but require more memory. Default is 24.\n- `` is the base learning rate for training phase 2. Default is 4e-3.\n- `` is the percentage of training steps used for warm-up at the start of training. Default is 200.\n- `` is the total number of training steps for phase 2, to be continued in addition to phase 1. Default is 930.\n- `` an integer indicating the number of steps to accumulate gradients over in phase 2. Effective batch size / GPU = `training_batch_size_p2` * `gradient_accumulation_steps_p2`. Default is 144.\n- `` A checkpoint to start the pre-training routine on (Usually a ELECTRA pretrained checkpoint). Default is `None`.\n \nThe complete list of the available parameters for the `run_pretraining.py` script are:\n \n```\n --model_name MODEL_NAME\n - Model name, used to define the name of the results folder.\n \n --pretrain_tfrecords PRETRAIN_TFRECORDS\n - Specifies tfrecord files used for pretraining.\n \n --max_seq_length MAX_SEQ_LENGTH\n - The maximum total input sequence length after\n WordPiece tokenization. Sequences longer than\n this will be truncated, and sequences shorter\n than this will be padded.\n \n --mask_prob MASK_PROB - Percentage of input tokens to mask out / replace.\n \n --disc_weight DISC_WEIGHT\n - Ratio of discriminator loss over generator loss.\n \n --generator_hidden_size GENERATOR_HIDDEN_SIZE\n - Fraction of discriminator hidden size for generator.\n \n --train_batch_size TRAIN_BATCH_SIZE\n - Batch size per GPU for training.\n \n --learning_rate LEARNING_RATE\n - The initial learning rate for the optimizer.\n \n --num_train_steps NUM_TRAIN_STEPS\n - Total number of training steps to perform.\n \n --num_warmup_steps NUM_WARMUP_STEPS\n - Number of steps of training to perform linear learning\n rate warmup for. For example, 0.1 = 10% of training.\n \n --seed SEED - Sets the seed to use for random number generation.\n \n --gradient_accumulation_steps GRADIENT_ACCUMULATION_STEPS\n - Number of update steps to accumulate before\n performing a backward/update pass.\n \n --fp16_compression - Whether to use 16-bit all reduce\n \n --amp - If set, will perform computations using\n automatic mixed precision.\n \n --log_freq LOG_FREQ - If set, the script will output the training\n loss every LOG_FREQ steps.\n \n --save_checkpoints_steps SAVE_CHECKPOINTS_STEPS\n - Checkpoints saving frequency.\n \n --keep_checkpoint_max KEEP_CHECKPOINT_MAX\n - Maximum number of checkpoints to keep.\n \n --restore_checkpoint RESTORE_CHECKPOINT\n - Whether to restore from a checkpoint; if specified,\n set to `path-to-checkpoint` or `latest`\n \n --phase2 - Specified if training on phase 2 only. \n If not specified, default pre-training is on phase 1.\n \n --optimizer OPTIMIZER - Specifies optimizer, `adam` or `lamb`.\n \n --skip_adaptive - Whether to apply adaptive learning rate on LayerNorm and biases.\n \n --gradient_accumulation_steps GRADIENT_ACCUMULATION_STEPS\n - Number of steps to accumulate gradients across before\n performing an update.\n \n --lr_decay_power LR_DECAY_POWER\n - Learning rate polynomial decay power.\n \n --opt_beta_1 OPT_BETA_1 - beta2 of optimizer.\n \n --opt_beta_2 OPT_BETA_2 - beta2 of optimizer.\n \n --end_lr END_LR - Ending learning rate.\n \n```\n \n#### Fine-tuning parameters\n \nDefault arguments are listed below in the order `scripts/run_squad.sh` expects:\n \n- ELECTRA MODEL - The default is `\"google/electra-base-discriminator\"`.\n- Number of training Epochs - The default is `2`.\n- Batch size - The default is `16`.\n- Learning rate - The default is `4e-4`.\n- Precision (either `amp`, `tf32` or `fp32`) - The default is `amp`.\n- Number of GPUs - The default is `8`.\n- Seed - The default is `1`.\n- SQuAD version - The default is `1.1`\n- SQuAD directory - The default is `/workspace/electra/data/download/squad/v$SQUAD_VERSION`.\n- Output directory for result - The default is `results/`.\n- Initialize checkpoint - The default is `\"None\"`\n- Mode (`train`, `eval`, `train_eval`, `prediction`) - The default is `train_eval`.\n \nThe script saves the checkpoint at the end of each epoch to the `checkpoints/` folder.\n \nThe main script `run_tf_squad.py` specific parameters are:\n \n```\n --electra_model ELECTRA_MODEL - Specifies the type of ELECTRA model to use;\n should be the discriminator of a pretrained checkpoint(output of postprocess_pretrained_ckpt.py)\n or one of the following:\n google/electra-small-generator\n google/electra-base-generator\n google/electra-large-generator\n google/electra-small-discriminator\n google/electra-base-discriminator\n google/electra-large-discriminator\n \n --amp - If set, will perform computations using\n automatic mixed precision.\n \n --data_dir DATA_DIR - Path to the SQuAD json for training and evaluation.\n \n --max_seq_length MAX_SEQ_LENGTH\n - The maximum total input sequence length\n after WordPiece tokenization.\n Sequences longer than this will be truncated,\n and sequences shorter than this will be padded.\n \n --doc_stride DOC_STRIDE - When splitting up a long document into chunks\n this parameters sets how much stride to take\n between chunks of tokens.\n \n --max_query_length MAX_QUERY_LENGTH\n - The maximum number of tokens for the question.\n Questions longer than \n will be truncated to the value specified.\n \n --n_best_size N_BEST_SIZE - The total number of n-best predictions to\n generate in the nbest_predictions.json\n output file.\n \n --max_answer_length MAX_ANSWER_LENGTH\n - The maximum length of an answer that can be\n generated. This is needed because the start and\n end predictions are not conditioned on one another.\n \n --joint_head - If true, beam search will be used to jointly predict\n the start and end positions. Default is True.\n \n --beam_size BEAM_SIZE - The beam size used to do joint predictions. The default value is 5.\n \n --verbose_logging - If true, all the warnings related to data\n processing will be printed. A number of warnings\n are expected for a normal SQuAD evaluation.\n \n --do_lower_case - Whether to lower case the input text. Set to\n true for uncased models and false for cased models.\n \n --version_2_with_negative - If true, the SQuAD examples contain questions\n that do not have an answer.\n \n --null_score_diff_threshold NULL_SCORE_DIFF_THRES HOLD\n - A null answer will be predicted if null_score\n is greater than NULL_SCORE_DIFF_THRESHOLD.\n```\n \n### Command-line options\n \nTo see the full list of available options and their descriptions, use the `-h` or `--help` command line option, for example:\n \n`python run_pretraining.py --help`\n \n`python run_tf_squad.py --help`\n \nDetailed descriptions of command-line options can be found in the [Parameters](#parameters) section.\n \n### Getting the data\n \nFor pre-training ELECTRA, we use the concatenation of Wikipedia (2500M words) as well as BookCorpus (800M words). For Wikipedia, we extract only the text passages and ignore headers, lists, and tables. ELECTRA requires that datasets are structured as a document level corpus rather than a shuffled sentence level corpus because it is critical to extract long contiguous sentences.\n \nThe preparation of the pre-training dataset is described in the `dataPrep.py` script found in the `data/` folder. The component steps in the automated scripts to prepare the datasets are as follows:\n \n1. Data download and extract - the dataset is downloaded and extracted.\n \n2. Clean and format - document tags, etc. are removed from the dataset.\n \n3. Sentence segmentation - the corpus text file is processed into separate sentences.\n \n4. Sharding - the sentence segmented corpus file is split into a number of uniformly distributed smaller text documents.\n \n5. `tfrecord` file creation - each text file shard is processed by the `build_pretraining_dataset.py` script to produce a corresponding `tfrecord` file. The script generates input data for the input text shard.\n \nThe tools used for preparing the BookCorpus and Wikipedia datasets can be applied to prepare an arbitrary corpus. The `create_datasets_from_start.sh` script in the `data/` directory applies sentence segmentation, sharding, and `tfrecord` file creation given an arbitrary text file containing a document-separated text corpus.\n \nFor fine-tuning a pre-trained ELECTRA model for specific tasks, by default this repository prepares the following dataset:\n \n- [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/): for question answering\n \nDepending on the speed of your internet connection, this process takes about a day to complete. The BookCorpus server could sometimes get overloaded and also contain broken links resulting in HTTP 403 and 503 errors. You can either skip the missing files or retry downloading at a later time.\n \n#### Multi-dataset\n \nThis repository provides functionality to combine multiple datasets into a single dataset for pre-training on a diverse text corpus at the shard level. Currently Wikipedia and BookCorpus get merged in `data/create_datasets_from_start.sh`. Snippets to download and format more text corpuses can be added to `data/dataPrep.py`. The sharding scheme combines multiple corpuses together and splits them into the required number of training(90%) and testing(10%) shards. Once the data is sharded, the `build_pretraining_dataset.py` converts raw text shards to tokenized segments and saves the dataset to the `data` directory in TFRecord format. This dataset can now be used to pre-train ELECTRA.\n \n### Training process\n \nThe training process consists of two steps: pre-training and fine-tuning.\n \n#### Pre-training\n \nPre-training is performed using `run_pretraining.py` along with parameters defined in `scripts/run_pretraining.sh` and `scripts/configs/pretrain_configs.sh`.\n \nThe `run_pretraining.sh` script runs a job on a single node that trains the ELECTRA-base model from scratch using Wikipedia and BookCorpus datasets as training data using the LAMB optimizer.\n \nPhase 1: (Maximum sequence length of 128)\n- Runs on 8 GPUs with training batch size of 176 per GPU\n- Uses a learning rate of 6e-3\n- Has FP16 precision enabled\n- Runs for 10000 steps, where the first 2000 are warm-up steps\n- Saves a checkpoint every 500 iterations (keeps only the latest 5 checkpoints) and at the end of training. All checkpoints, and training logs are saved to the `results/` directory.\n- Creates a log file containing all the output\n \nPhase 2: (Maximum sequence length of 512)\n- Runs on 8 GPUs with training batch size of 24 per GPU\n- Uses a learning rate of 4e-3\n- Has FP16 precision enabled\n- Runs for 930 steps, where the first 200 are warm-up steps\n- Saves a checkpoint every 500 iterations (keeps only the latest 5 checkpoints) and at the end of training. All checkpoints, and training logs are saved to the `results/` directory.\n- Creates a log file containing all the output\n \nSpecific configs available at `scripts/configs/pretrain_config.sh` can be run as follows:\n```\nbash scripts/run_pretraining.sh $(source scripts/configs/pretrain_config.sh && dgxa100_8gpu_amp)\nbash scripts/run_pretraining.sh $(source scripts/configs/pretrain_config.sh && dgx2_16gpu_amp)\nbash scripts/run_pretraining.sh $(source scripts/configs/pretrain_config.sh && dgx1_8gpu_amp)\n```\n \nThe above commands will train ELECTRA based on Wikipedia and BookCorpus to state-of-the-art accuracy on any DGX platform using FP16 arithmetic. Around 96% of the training sequences are of length 128 (phase 1 of training) and less than 4% of the training sequences are of length 512 (phase 2 of training).\n \nIn order to run pre-training routine on an initial checkpoint, perform the following in `scripts/run_pretraining.sh`:\n- set `restore_checkpoint=`\n- Note: The parameter value assigned to `--model_size` during training should remain unchanged. Also, to resume pre-training on your corpus of choice, the training dataset should be created using the same vocabulary file used in `data/create_datasets_from_start.sh`.\n \n#### Multi-node\n \nMulti-node runs can be launched on a Pyxis/enroot Slurm cluster (see [Requirements](#requirements)) with the `run.sub` script with the following command for a 48-node NVIDIA DGX A100 example for both phase 1 and phase 2:\n \n```\nBATCHSIZE=176 LR=6e-3 GRAD_ACCUM_STEPS=1 PHASE=1 STEPS=10000 WARMUP=2000 b1=0.878 b2=0.974 decay=0.5 skip_adaptive=yes end_lr=0.0 sbatch N48 --ntasks-per-node=8 run.sub\nBATCHSIZE=24 LR=4e-3 GRAD_ACCUM_STEPS=3 PHASE=2 STEPS=930 WARMUP=200 b1=0.878 b2=0.974 decay=0.5 skip_adaptive=yes end_lr=0.0 sbatch N48 --ntasks-per-node=8 run.sub\n```\n \nCheckpoint after phase 1 will be saved in `/models/`. The checkpoint will be automatically picked up to resume training on phase 2. Note that phase 2 should be run after phase 1.\n \nThe batch variables `BATCHSIZE`, `LR`, `GRAD_ACCUM_STEPS`, `PHASE`, `STEPS`, `WARMUP`, `b1`, `b2`, `decay`, `skip_adaptive` and `end_lr` refer to the Python arguments `train_batch_size`, `learning_rate`, `gradient_accumulation_steps`, `phase2`, `num_train_steps`, `num_warmup_steps`, `opt_beta_1`, `opt_beta_2`, `lr_decay_power`, `skip_adaptive` and `end_lr` in `run_pretraining.py` respectively.\n \nNote that the `run.sub` script is a starting point that has to be adapted depending on the environment. In particular, variables such as `docker_image` and `datadir` handle the location of the files for each phase.\n \nRefer to the files contents to see the full list of variables to adjust for your system.\n \n#### Fine-tuning\n \nFine-tuning is provided for a variety of tasks. The following tasks are included with this repository through the following scripts:\n \n- Question Answering (`scripts/run_squad.sh`)\n \nBy default, each Python script implements fine-tuning a pre-trained ELECTRA model for a specified number of training epochs as well as evaluation of the fine-tuned model. Each shell script invokes the associated Python script with the following default parameters:\n \n- Uses 8 GPUs\n- Has FP16 precision enabled\n- Has XLA enabled\n- Saves a checkpoint at the end of training to the `checkpoints/` folder\n \nSpecific configs available at `scripts/configs/squad_configs.sh` can be run as follows:\n```\nbash scripts/run_squad.sh $(source scripts/configs/squad_config.sh && dgxa100_8gpu_amp) train_eval\nbash scripts/run_squad.sh $(source scripts/configs/squad_config.sh && dgx2_16gpu_amp) train_eval\nbash scripts/run_squad.sh $(source scripts/configs/squad_config.sh && dgx1_8gpu_amp) train_eval\n```\n \nFine-tuning Python scripts implement support for mixed precision and multi-GPU training through [Horovod](https://github.com/horovod/horovod). For a full list of parameters and associated explanations, see the [Parameters](#parameters) section.\n \nAll fine-tuning shell scripts have the same positional arguments, outlined below:\n \n```bash scripts/run_squad.sh ```\n \nBy default, the mode positional argument is set to `train_eval`. See the [Fine-tuning parameters](#fine-tuning-parameters) for explanations of each positional argument.\n \nNote: The first positional argument (the path to the checkpoint to load) is required.\n \nEach fine-tuning script assumes that the corresponding dataset files exist in the `data/` directory or separate path can be a command-line input to `run_squad.sh`.\n \n### Inference process\n \n#### Fine-tuning inference\n \nEvaluation fine-tuning is enabled by the same scripts as training:\n \n- Question Answering (`scripts/run_squad.sh`)\n \nThe mode positional argument of the shell script is used to run in evaluation mode. The fine-tuned ELECTRA model will be run on the evaluation dataset, and the evaluation loss and accuracy will be displayed.\n \nEach inference shell script expects dataset files to exist in the same locations as the corresponding training scripts. The inference scripts can be run with default settings. By setting the `mode` variable in the script to either `eval` or `prediction` flag, you can choose between running predictions and evaluating them on a given dataset or just the former.\n \n`bash scripts/run_squad.sh `\n \nTo run inference interactively on question-context pairs, use the script `run_inference.py` as follows:\n \n`python run_inference.py --electra_model --init_checkpoint --question=\"What food does Harry like?\" --context=\"My name is Harry and I grew up in Canada. I love apples.\"`","hasAdvanced","performance","The performance measurements in this document were conducted at the time of publication and may not reflect the performance achieved from NVIDIA's latest software release. For the most up-to-date performance measurements, go to [NVIDIA Data Center Deep Learning Product Performance](https://developer.nvidia.com/deep-learning-performance-training-inference).\n \n### Benchmarking\n \nThe following section shows how to run benchmarks measuring the model performance in training and inference modes.\n \n#### Training performance benchmark\n \nTraining performance benchmarks for both pre-training phases can be obtained by running `scripts/benchmark_pretraining.sh`. Default parameters are set to run a few training steps for a converging NVIDIA DGX A100 system.\n \nTo benchmark training performance with other parameters, run:\n```\nbash scripts/benchmark_pretraining.sh \n```\n \nAn example call used to generate throughput numbers:\n```\nbash scripts/benchmark_pretraining.sh 88 amp xla 8 true 2 12 4 base\n```\n \nTraining performance benchmarks for fine-tuning can be obtained by running `scripts/benchmark_squad.sh`. The required parameters can be passed through the command-line as described in [Training process](#training-process). The performance information is printed after 200 training iterations.\n \nTo benchmark the training performance on a specific batch size, run:\n```\nbash scripts/benchmark_squad.sh train \n```\n \nAn example call used to generate throughput numbers:\n```\nbash scripts/benchmark_squad.sh train 8 16\n```\n \n#### Inference performance benchmark\n \nInference performance benchmarks fine-tuning can be obtained by running `scripts/benchmark_squad.sh`. The required parameters can be passed through the command-line as described in [Inference process](#inference-process). This script runs one epoch by default on the SQuAD v1.1 dataset and extracts the average performance for the given configuration. \n \nTo benchmark the training performance on a specific batch size, run:\n`bash scripts/benchmark_squad.sh train `\n \nAn example call used to generate throughput numbers:\n`bash scripts/benchmark_squad.sh eval 8 256`\n \n### Results\n \nThe following sections provide details on how we achieved our performance and accuracy in training and inference. All results are on ELECTRA-base model and on SQuAD v1.1 dataset with a sequence length of 384 unless otherwise mentioned.\n \n#### Training accuracy results\n \n##### Pre-training loss curves\n![Pretraining Loss Curves](https://raw.githubusercontent.com/NVIDIA/DeepLearningExamples/master/TensorFlow2/LanguageModeling/ELECTRA//images/total_loss.svg)\n\n \nPhase 1 is shown by the blue curve and Phase 2 by the grey. Y axis stands for the total loss and x axis for the total steps trained.\n \n##### Pre-training loss results\n \n| DGX System | GPUs | Batch size / GPU (Phase 1 and Phase 2) | Accumulation steps (Phase 1 and Phase 2) | Final Loss - TF32/FP32 | Final Loss - mixed precision | Time to train(hours) - TF32/FP32 | Time to train(hours) - mixed precision | Time to train speedup (TF32/FP32 to mixed precision)\n|---|---|---|---|---|---|---|---|---\n|48 x DGX A100 |8 |176 and 24 |1 and 3 |8.686|8.68|1.61 |1.126|1.43 \n|24 x DGX-2H |16|176 and 24 |1 and 3 |8.72 |8.67|5.58 |1.74 |3.20\n|1 x DGX A100 |8 |176 and 24 |48 and 144|- |- |54.84 |30.47|1.8\n|1 x DGX-1 16G |8 |88 and 12 |96 and 288|- |- |241.8 |65.1 |3.71\n|1 x DGX-2 32G |16|176 and 24 |24 and 72 |- |- |109.97|29.08|3.78\n \nIn the above table, FP32 and TF32 runs were made at half the batch per GPU and twice the gradient accumulation steps of a run with mixed precision in order to not run out of memory.\n \nThe SQuAD fine-tuning scripts by default train on [Google's ELECTRA++ base pretrained checkpoint](https://github.com/google-research/electra#released-models) which uses around 10x training dataset (dataset used by XLNet authors) and greater than 5x training steps compared to the training recipe in `scripts/run_pretraining.sh`. The latter trains and achieves state-of-the-art accuracy on Wikipedia and BookCorpus datasets only.\n \n##### Fine-tuning accuracy: NVIDIA DGX A100 (8x A100 40GB)\n \nOur results were obtained by running the `scripts/run_squad.sh` training script in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX A100 (8x A100 40GB) GPUs. \n \n*ELECTRA BASE++*\n \n| GPUs | Batch size / GPU | Accuracy / F1 - FP32 | Accuracy / F1 - mixed precision | Time to train - TF32 (sec) | Time to train - mixed precision (sec) | Time to train speedup (FP32 to mixed precision) | \n|---------|---------------------|------------------|-----------------------------|--------------------------|---------------------------------|-------------------------------------------------|\n| 1 | 32 | 87.19 / 92.85 | 87.19 / 92.84 | 1699 | 749 | 2.27 |\n| 8 | 32 | 86.84 / 92.57 | 86.83 / 92.56 | 263 | 201 | 1.30 |\n \n##### Fine-tuning accuracy: NVIDIA DGX-1 (8x V100 16GB)\n \nOur results were obtained by running the `scripts/run_squad.sh` training script in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX-1 with (8x V100 16GB) GPUs.\n \n*ELECTRA BASE++*\n \n| GPUs | Batch size / GPU (FP32 : mixed precision) | Accuracy / F1 - FP32 | Accuracy / F1 - mixed precision | Time to train - FP32 (sec) | Time to train - mixed precision (sec) | Time to train speedup (FP32 to mixed precision) | \n|---------|---------------------|------------------|-----------------------------|--------------------------|---------------------------------|-------------------------------------------------|\n| 1 | 8 : 16 | 87.36 / 92.82 | 87.32 / 92.74 | 5136 | 1378 | 3.73 |\n| 8 | 8 : 16 | 87.02 / 92.73 | 87.02 / 92.72 | 730 | 334 | 2.18 |\n \n*ELECTRA BASE checkpoint Wikipedia and BookCorpus*\n \nGPUs | SQuAD version| Batch size / GPU (FP32 : mixed precision) | Accuracy / F1 - FP32 | Accuracy / F1 - mixed precision | Time to train - FP32 (sec) | Time to train - mixed precision (sec) | Time to train speedup (FP32 to mixed precision) | \n|---------|-----|----------------|------------------|-----------------------------|--------------------------|---------------------------------|-------------------------------------------------|\n| 8 | v1.1 | 8 : 16 | 85.00 / 90.94 | 85.04 / 90.96 | 5136 | 1378 | 3.73 |\n| 8 | v2.0 | 8 : 16 | 80.517 / 83.36 | 80.523 / 83.43 | 730 | 334 | 2.18 \n \n##### Fine-tuning accuracy: NVIDIA DGX-2 (16x V100 32GB)\n \nOur results were obtained by running the `scripts/run_squad.sh` training script in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX-2 (16x V100 32G) GPUs.\n \n*ELECTRA BASE++*\n \n| GPUs | Batch size / GPU | Accuracy / F1 - FP32 | Accuracy / F1 - mixed precision | Time to train - FP32 (sec) | Time to train - mixed precision (sec) | Time to train speedup (FP32 to mixed precision) | \n|---------|---------------------|------------------|-----------------------------|--------------------------|---------------------------------|-------------------------------------------------|\n| 1 | 32 | 87.14 / 92.69 | 86.95 / 92.69 | 4478 | 1162 | 3.85 |\n| 16 | 32 | 86.95 / 90.58 | 86.93 / 92.48 | 333 | 229 | 1.45 |\n \n##### Training stability test\n \n###### Pre-training stability test: NVIDIA DGX A100 (8x A100 40GB)\n \n*ELECTRA BASE Wikipedia and BookCorpus*\n \nTraining stability with 48 x DGX A100, TF32 computations and loss reported after Phase 2:\n \n| Accuracy Metric | Seed 1 | Seed 2 | Seed 3 | Seed 4 | Seed 5 | Mean | Standard Deviation\n|---|---|---|---|---|---|---|---\n|Final Loss| 8.72 | 8.69 | 8.71 | 8.7 | 8.68 | 8.7 | 0.015\n \n###### Fine-tuning stability test: NVIDIA DGX-1 (8x V100 16GB)\n \n*ELECTRA BASE++*\n \nTraining stability with 8 GPUs, FP16 computations, batch size of 16 on SQuAD v1.1:\n \n| Accuracy Metric | Seed 1 | Seed 2 | Seed 3 | Seed 4 | Seed 5 | Mean | Standard Deviation\n|---|---|---|---|---|---|---|---\n|Exact Match %| 86.99 | 86.81 | 86.95 | 87.10 | 87.26 | 87.02 | 0.17\n| f1 % | 92.7 | 92.66 | 92.65 | 92.61 | 92.97 | 92.72 | 0.14\n \n Training stability with 8 GPUs, FP16 computations, batch size of 16 on SQuAD v2.0:\n \n| Accuracy Metric | Seed 1 | Seed 2 | Seed 3 | Seed 4 | Seed 5 | Mean | Standard Deviation\n|---|---|---|---|---|---|---|---\n|Exact Match %| 83.00 | 82.84 | 83.11 | 82.70 | 82.94 | 82.91 | 0.15\n| f1 % | 85.63 | 85.48 | 85.69 | 85.31 | 85.57 | 85.54 | 0.15\n \n#### Training performance results\n \n##### Training performance: NVIDIA DGX A100 (8x A100 40GB)\n \nOur results were obtained by running the `scripts/benchmark_squad.sh` training script in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX A100 (8x A100 40GB) GPUs. Performance numbers (in items/images per second) were averaged over an entire training epoch.\n \n###### Pre-training NVIDIA DGX A100 (8x A100 40GB)\n \n| GPUs | Batch size / GPU (TF32 and FP16) | Accumulation steps (TF32 and FP16) | Sequence length | Throughput - TF32(sequences/sec) | Throughput - mixed precision(sequences/sec) | Throughput speedup (TF32 - mixed precision) | Weak scaling - TF32 | Weak scaling - mixed precision\n|------------------|----------------------|----------------------|-------------------|-----------------------------------------------|------------------------------------|---------------------------------|----------------------|----------------------------------------------\n|1 | 88 and 176| 768 and 384 | 128| 533 |955 |1.79|1.00| 1.00\n|8 | 88 and 176| 96 and 48 | 128| 4202|7512|1.79|7.88| 7.87\n|1 | 12 and 24 | 2304 and 1152| 512| 90 |171 |1.90|1.00| 1.00\n|8 | 12 and 24 | 288 and 144 | 512| 716 |1347|1.88|7.96| 7.88 \n \n###### Fine-tuning NVIDIA DGX A100 (8x A100 40GB)\n \n| GPUs | Batch size / GPU | Sequence length | Throughput - TF32 (sequences/sec) | Throughput - mixed precision (sequences/sec) | Throughput speedup (TF32 - mixed precision) | Weak scaling - TF32 | Weak scaling - mixed precision |\n|------------------|-----------|-----------|-----------------------------------------------|------------------------------------|---------------------------------|----------------------|----------------------------------------------\n| 1 | 32 | 384 | 107 | 317 | 2.96 | 1.00 | 1.00\n| 8 | 32 | 384 | 828 | 2221| 2.68 | 7.74 | 7.00\n \n##### Training performance: NVIDIA DGX-1 (8x V100 16GB)\n \nOur results were obtained by running the `scripts/benchmark_squad.sh` training scripts in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX-1 with (8x V100 16GB) GPUs. Performance numbers (in sequences per second) were averaged over an entire training epoch.\n \n###### Pre-training NVIDIA DGX-1 (8x V100 16GB)\n \n| GPUs | Batch size / GPU (FP32 and FP16) | Accumulation steps (FP32 and FP16) | Sequence length | Throughput - FP32(sequences/sec) | Throughput - mixed precision(sequences/sec) | Throughput speedup (FP32 - mixed precision) | Weak scaling - FP32 | Weak scaling - mixed precision\n|------------------|----------------------|----------------------|-------------------|-----------------------------------------------|------------------------------------|---------------------------------|----------------------|----------------------------------------------\n|1 | 40 and 88| 1689 and 768 | 128| 116 |444 |3.83 |1.00 | 1.00\n|8 | 40 and 88| 211 and 96 | 128| 920 |3475|3.77 |7.93 | 7.83\n|1 | 6 and 12 | 4608 and 2304| 512| 24 |84 |3.50 |1.00 | 1.00\n|8 | 6 and 12 | 576 and 288 | 512| 190 |656 |3.45 |7.92 | 7.81 \n \n###### Fine-tuning NVIDIA DGX-1 (8x V100 16GB)\n \n| GPUs | Batch size / GPU (FP32 : mixed precision) | Sequence length | Throughput - FP32 (sequences/sec) | Throughput - mixed precision (sequences/sec) | Throughput speedup (FP32 - mixed precision) | Weak scaling - FP32 | Weak scaling - mixed precision |\n|------------------|-----------|-----------|-----------------------------------------------|------------------------------------|---------------------------------|----------------------|----------------------------------------------\n|1 | 8 : 16| 384| 35| 154| 4.4 | 1.00| 1.00\n|8 | 8 : 16| 384|268|1051| 3.92| 7.66| 6.82\n \nTo achieve these same results, follow the steps in the [Quick Start Guide](#quick-start-guide).\n \n##### Training performance: NVIDIA DGX-2 (16x V100 32GB)\n \nOur results were obtained by running the `scripts/benchmark_squad.sh` training scripts in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX-2 with (16x V100 32G) GPUs. Performance numbers (in sequences per second) were averaged over an entire training epoch.\n \n###### Pre-training NVIDIA DGX-2 (16x V100 32GB)\n \n| GPUs | Batch size / GPU (FP32 and FP16) | Accumulation steps (FP32 and FP16) | Sequence length | Throughput - FP32(sequences/sec) | Throughput - mixed precision(sequences/sec) | Throughput speedup (FP32 - mixed precision) | Weak scaling - FP32 | Weak scaling - mixed precision\n|------------------|----------------------|----------------------|-------------------|-----------------------------------------------|------------------------------------|---------------------------------|----------------------|----------------------------------------------\n|1 | 88 and 176| 768 and 384 | 128| 128 |500 |3.91| 1.00 | 1.00\n|8 | 88 and 176| 96 and 48 | 128| 1011|3916|3.87| 7.90 | 7.83\n|16| 88 and 176| 48 and 24 | 128| 2018|7773|3.85|15.77 |15.55\n|1 | 12 and 24 | 2304 and 1152| 512| 27 |96 |3.55| 1.00 | 1.00\n|8 | 12 and 24 | 288 and 144 | 512| 213 |754 |3.54| 7.89 | 7.85 \n|16| 12 and 24 | 144 and 72 | 512| 426 |1506|3.54| 15.78|15.69 \n \n###### Fine-tuning NVIDIA DGX-2 (16x V100 32GB)\n \n| GPUs | Batch size / GPU | Sequence length | Throughput - FP32 (sequences/sec) | Throughput - mixed precision (sequences/sec) | Throughput speedup (FP32 - mixed precision) | Weak scaling - FP32 | Weak scaling - mixed precision |\n|------|-----------|-------|----------------------------------|---------------------------------------------|---------------------------------------------|---------------------|--------------------------------|\n| 1 | 16 | 384 | 40 | 184 | 4.6 | 1.00 | 1.00 |\n| 8 | 16 | 384 | 311 | 1289 | 4.14 | 7.77 | 7.00 |\n| 16 | 16 | 384 | 626 | 2594 | 4.14 | 15.65 | 14.09 |\n \nTo achieve these same results, follow the steps in the [Quick Start Guide](#quick-start-guide).\n \n#### Inference performance results\n \n##### Inference performance: NVIDIA DGX A100 (1x A100 40GB)\n \nOur results were obtained by running the `scripts/benchmark_squad.sh` inferencing benchmarking script in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA DGX A100 (1x A100 40GB) GPU.\n \n###### Fine-tuning inference on NVIDIA DGX A100 (1x A100 40GB)\n \nFP16\n \n| Batch size | Sequence length | Throughput Avg (sequences/sec) | Latency Avg (ms) | Latency 90% (ms) | Latency 95% (ms) | Latency 99% (ms) |\n|------------|-----------------|--------------------------------|------------------|------------------|------------------|------------------|\n| 1 | 384 | 166 | 6.035 | 5.995 | 6.013 | 6.029 |\n| 256 | 384 | 886 | 276.26 | 274.53 | 275.276 | 275.946 |\n| 512 | 384 | 886 | 526.5 | 525.014 | 525.788 | 525.788 |\n \nTF32\n \n| Batch size | Sequence length | Throughput Avg (sequences/sec) | Latency Avg (ms) | Latency 90% (ms) | Latency 95% (ms) | Latency 99% (ms) |\n|------------|-----------------|--------------------------------|------------------|------------------|------------------|------------------|\n| 1 | 384 | 122 | 8.228 | 8.171 | 8.198 | 8.221 |\n| 256 | 384 | 342 | 729.293 | 727.990 | 728.505 | 729.027 |\n| 512 | 384 | 350 | 1429.314 | 1427.719 | 1428.550 | 1428.550 |\n \n##### Inference performance: NVIDIA T4\n \nOur results were obtained by running the `scripts/benchmark_squad.sh` script in the tensorflow:20.07-tf2-py3 NGC container on NVIDIA Tesla T4 (1x T4 16GB) GPU. \n \n###### Fine-tuning inference on NVIDIA T4\n \nFP16\n \n| Batch size | Sequence length | Throughput Avg (sequences/sec) | Latency Avg (ms) | Latency 90% (ms) | Latency 95% (ms) | Latency 99% (ms) |\n|------------|-----------------|--------------------------------|------------------|------------------|------------------|------------------|\n| 1 | 384 | 58 | 17.413 | 17.295 | 17.349 | 17.395 |\n| 128 | 384 | 185 | 677.298 | 675.211 | 675.674 | 676.269 |\n| 256 | 384 | 169 | 1451.396 | 1445.070 | 1447.654 | 1450.141 |\n \nTo achieve these same results, follow the steps in the [Quick Start Guide](#quick-start-guide).","hasPerformance","policies",[],{"_76":21,"_78":198,"_169":-5,"_175":31,"_177":31,"_193":197,"_180":86},{"_76":23,"_78":23,"_169":-5,"_175":31,"_177":31,"_193":197,"_180":86},[],"Latest (20.07.0)","nvidia/dle/electra_tf2","key","nvidia/dle/resources/electra_tf2","resourceType","resources","resourceId","electra_tf2","shortDescription","ELECTRA is method of pre-training language representations which outperforms existing techniques on a wide array of NLP tasks.","This resource is using open-source code maintained in github (see the quick-start-guide section) and available for download from [NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/dle/resources/electra-tf2)\n\n \nElectra (Efficiently Learning an Encoder that Classifies Token Replacements Accurately), is a novel pre-training method for language representations which outperforms existing techniques, given the same compute budget on a wide array of Natural Language Processing (NLP) tasks. This model is based on the [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/forum?id=r1xMH1BtvB) paper. NVIDIA's implementation of ELECTRA is an optimized version of the [Hugging Face implementation](https://huggingface.co/transformers/model_doc/electra.html), leveraging mixed precision arithmetic and Tensor Cores on Volta, Turing, and the NVIDIA Ampere GPU architectures for faster training times with state-of-the-art accuracy.\n \nThis repository contains the scripts to interactively launch data download, training, benchmarking and inference routines in a Docker container for pre-training on your own dataset (Wikipedia and BookCorpus shown as an example), and fine-tuning for tasks such as question answering. The major differences between the original implementation as described in the paper and this version of ELECTRA are as follows:\n \n- Scripts to download Wikipedia and BookCorpus datasets\n- Scripts to preprocess downloaded data or a custom corpus into inputs and targets for pre-training in a modular fashion\n- Automatic mixed precision (AMP) support and optimized for performance\n- Multi-GPU and Multi-node training support with push-button scripts to reach state-of-the-art accuracy and performance.\n \nOther publicly available implementations of Electra include:\n1. [Hugging Face](https://huggingface.co/transformers/model_doc/electra.html)\n2. [Google's implementation](https://github.com/google-research/electra)\n \nThis model is trained with mixed precision using Tensor Cores on Volta, Turing, and the NVIDIA Ampere GPU architectures. Additionally, this model provides push-button solutions to pre-training, fine-tuning and inference and on a corpus of choice. As a result, researchers can get results up to 4x faster than training without Tensor Cores. This model is tested against each NGC monthly container release to ensure consistent accuracy and performance over time.\n \n### Model architecture\n \nELECTRA is a combination of two Transformer models: a generator and a discriminator. The generator's role is to replace tokens in a sequence, and is therefore trained as a masked language model. The discriminator, which is the model we are interested in, tries to identify which tokens were replaced by the generator in the sequence. Both generator and discriminator use the same architecture as the encoder of the Transformer. The encoder is simply a stack of Transformer blocks, which consist of a multi-head attention layer followed by successive stages of feed-forward networks and layer normalization. The multi-head attention layer performs self-attention on multiple input representations.\n \n![Figure 1-1](https://1.bp.blogspot.com/-sHybc03nJRo/XmfLongdVYI/AAAAAAAAFbI/a0t5w_zOZ-UtxYaoQlVkmTRsyFJyFddtQCLcBGAsYHQ/s1600/image1.png \"ELECTRA architecture\")\n \n### Default configuration\n \nELECTRA uses a new pre-training task called replaced token detection (RTD), that trains a bidirectional model (like a MLM) while learning from all input positions (like a LM). Inspired by generative adversarial networks (GANs), instead of corrupting the input by replacing tokens with \"[MASK]\" as in BERT, the generator is trained to corrupt the input by replacing some input tokens with incorrect, but somewhat plausible, fakes. On the other hand, the discriminator is trained to distinguish between \"real\" and \"fake\" input data. \n \nThe [Google ELECTRA repository](https://github.com/google-research/electra) reports the results for three configurations of ELECTRA, each corresponding to a unique model size. This implementation provides the same configurations by default, which are described in the table below.\n \n| **Model** | **Hidden layers** | **Hidden unit size** | **Parameters** |\n|:---------:|:----------:|:---:|:----:|\n|ELECTRA_SMALL|12 encoder| 256 | 14M|\n|ELECTRA_BASE |12 encoder| 768 |110M|\n|ELECTRA_LARGE|24 encoder|1024 |335M|\n \nThe following features were implemented in this model:\n- General:\n - Mixed precision support with TensorFlow Automatic Mixed Precision (TF-AMP)\n - Multi-GPU support using Horovod\n - XLA support\n - Multi-Node support\n \n- Training\n - Pre-training support\n - Fine-tuning example\n \n- Inference:\n - Joint predictions with beam search.\n \n### Feature support matrix\n \nThe following features are supported by this model.\n \n| **Feature** | **ELECTRA** |\n|:---------:|:----------:|\n|LAMB|Yes|\n|Automatic mixed precision (AMP)|Yes|\n|XLA|Yes|\n|Horovod Multi-GPU|Yes|\n|Multi-node|Yes|\n \n#### Features\n \n**Automatic Mixed Precision (AMP)**\n \nThis implementation of ELECTRA uses AMP to implement mixed precision training. It allows us to use FP16 training with FP32 master weights by modifying just a few lines of code.\n \n**Horovod**\n \nHorovod is a distributed training framework for TensorFlow, Keras, PyTorch, and MXNet. The goal of Horovod is to make distributed deep learning fast and easy to use. For more information about how to get started with Horovod, see the [Horovod: Official repository](https://github.com/horovod/horovod).\n \nMulti-GPU training with Horovod\n \nOur model uses Horovod to implement efficient multi-GPU training with NCCL. For details, see example sources in this repository or see the [TensorFlow tutorial](https://github.com/horovod/horovod/#usage).\n \n**XLA support (experimental)**\n \nXLA is a domain-specific compiler for linear algebra that can accelerate TensorFlow models with potentially no source code changes. The results are improvements in speed and memory usage: most internal benchmarks run ~1.1-1.5x faster after XLA is enabled. \n[AMP](https://nvidia.github.io/apex/amp.html) is an abbreviation used for automatic mixed precision training.\n \n**Multi-node Training**\n \nSupported on a Pyxis/Enroot Slurm cluster.\n \n### Mixed precision training\n \nMixed precision is the combined use of different numerical precisions in a computational method. [Mixed precision](https://arxiv.org/abs/1710.03740) training offers significant computational speedup by performing operations in half-precision format, while storing minimal information in single-precision to retain as much information as possible in critical parts of the network. Since the introduction of [Tensor Cores](https://developer.nvidia.com/tensor-cores) in Volta, and following with both the Turing and Ampere architectures, significant training speedups are experienced by switching to mixed precision -- up to 3x overall speedup on the most arithmetically intense model architectures. Using mixed precision training requires two steps:\n \n1. Porting the model to use the FP16 data type where appropriate.\n2. Adding loss scaling to preserve small gradient values.\n \nThis can now be achieved using Automatic Mixed Precision (AMP) for TensorFlow to enable the full [mixed precision methodology](https://docs.nvidia.com/deeplearning/sdk/mixed-precision-training/index.html#tensorflow) in your existing TensorFlow model code. AMP enables mixed precision training on Volta, Turing, and NVIDIA Ampere GPU architectures automatically. The TensorFlow framework code makes all necessary model changes internally.\n \nIn TF-AMP, the computational graph is optimized to use as few casts as necessary and maximize the use of FP16, and the loss scaling is automatically applied inside of supported optimizers. AMP can be configured to work with the existing tf.contrib loss scaling manager by disabling the AMP scaling with a single environment variable to perform only the automatic mixed-precision optimization. It accomplishes this by automatically rewriting all computation graphs with the necessary operations to enable mixed precision training and automatic loss scaling.\n \nFor information about:\n- How to train using mixed precision, see the [Mixed Precision Training](https://arxiv.org/abs/1710.03740) paper and [Training With Mixed Precision](https://docs.nvidia.com/deeplearning/performance/mixed-precision-training/index.html) documentation.\n- Techniques used for mixed precision training, see the [Mixed-Precision Training of Deep Neural Networks](https://devblogs.nvidia.com/mixed-precision-training-deep-neural-networks/) blog.\n- How to access and enable AMP for TensorFlow, see [Using TF-AMP](https://docs.nvidia.com/deeplearning/dgx/tensorflow-user-guide/index.html#tfamp) from the TensorFlow User Guide.\n \n#### Enabling mixed precision\n \nThis implementation exploits the TensorFlow Automatic Mixed Precision feature. To enable AMP, you simply need to supply the `--amp` flag to the `run_pretraining.py` or `run_tf_squad.py` script. For reference, enabling AMP required us to apply the following changes to the code:\n \n1. Set the Keras mixed precision policy:\n ```python\n if config.amp:\n policy = tf.keras.mixed_precision.experimental.Policy(\"mixed_float16\", loss_scale=\"dynamic\")\n tf.keras.mixed_precision.experimental.set_policy(policy)\n ```\n \n2. Use the loss scaling wrapper on the optimizer:\n ```python\n if config.amp:\n optimizer = tf.keras.mixed_precision.experimental.LossScaleOptimizer(optimizer, \"dynamic\")\n ```\n \n3. Use scaled loss to calculate the gradients:\n ```python\n #Scale loss\n if config.amp:\n total_loss = optimizer.get_scaled_loss(total_loss)\n gradients = tape.gradient(total_loss, model.trainable_variables)\n #Get unscaled gradients if AMP\n if config.amp:\n gradients = optimizer.get_unscaled_gradients(gradients)\n ```\n \n#### Enabling TF32\n \nTensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. \n \nTF32 Tensor Cores can speed up networks using FP32, typically with no loss of accuracy. It is more robust than FP16 for models which require high dynamic range for weights or activations.\n \nFor more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.\n \nTF32 is supported in the NVIDIA Ampere GPU architecture and is enabled by default.\n \n### Glossary\n \n**Fine-tuning** \nTraining an already pretrained model further using a task specific dataset for subject-specific refinements, by adding task-specific layers on top if required.\n \n**Language Model** \nAssigns a probability distribution over a sequence of words. Given a sequence of words, it assigns a probability to the whole sequence.\n \n**Pre-training** \nTraining a model on vast amounts of data on the same (or different) task to build general understandings.\n \n**Transformer** \nThe paper [Attention Is All You Need](https://arxiv.org/abs/1706.03762) introduces a novel architecture called Transformer that uses an attention mechanism and transforms one sequence into another.\n \n **Phase 1**\nPretraining on samples of sequence length 128 and at most 15% masked predictions per sequence.\n \n**Phase 2**\nPretraining on samples of sequence length 512 and at most 15% masked predictions per sequence.","displayName","ELECTRA for TensorFlow2",["D",1680631798417],"publisher","NVIDIA Deep Learning Examples","latestVersion","versions","collectionCounts","imageUrl","/api/image/nvidia/dle/resources/electra_tf2","rawImageUrl","https://assets.nvidiagrid.net/ngc/logos/Deep-Learning-Examples.png","canGuestDownload","latestVersionSizeInBytes","licenseTerms","isMultinodeEnabled","isMultiArchitecture","bias","explainability","privacy","safetyAndSecurity","featureLabels",[],"displayLabels",[241],"policyLabels",[],"systemLabels",[],"products",[],"hasNimMetadata",{"_209":242,"_243":244,"_245":246,"_247":248},"Natural Language Understanding","value","uscs_natural_language_understanding","filter",{"_249":250,"_243":244},"category","label","field","application",{"_107":1236,"_109":1237},"ngcApps",[1189],"csaApiBaseUrl","https://www.nvidia.com/ai-agents/csa-agent/dcsm","datadogRum",{"_1170":1171,"_1172":1173,"_1174":1175,"_1176":1177,"_1178":262,"_138":290,"_1179":1180,"_1181":56,"_1182":86,"_1183":86,"_1184":1185,"_1186":86,"_1187":1188},"amplitude",{"_1167":1168,"_803":-7,"_1169":31},"disableJs","environment","prod","featureFlagsBlob","JTdCJTIydGVzdEZsYWclMjIlM0FmYWxzZSUyQyUyMnNob3dTZWFyY2hMYWJlbFNldEZpbHRlcnMlMjIlM0FmYWxzZSUyQyUyMmVuYWJsZUNzYVdpZGdldCUyMiUzQWZhbHNlJTJDJTIyZW5hYmxlUHJvZHVjdEdyb3VwaW5nQnlTdWJzY3JpcHRpb24lMjIlM0FmYWxzZSUyQyUyMmNvbnRhaW5lclNjYW5uaW5nUmVkZXNpZ25NMiUyMiUzQWZhbHNlJTdE","featureFlagsGuarded","isDevLikeEnv","isPlaywright","isMobileDevice","categorizedOrgs",{"_1161":1162,"_1163":1164,"_1165":1166},"currentOrgContext","currentContextRoles",[],"enterpriseOrgs",[],"manageableExternalUserGroups",[],"locale","en","ngcApiOrigin","ngcHelmOrigin","origin","https://catalog.ngc.nvidia.com","usesMocks","theme","year",2026,"now",1786122320816,"1.11.0","defaultSearchSuggestions",{"_1043":370,"_1044":1045,"_1046":1047,"_1048":1049},"announcements",[],"consentState",{"_327":1033,"_1034":86,"_1035":31,"_1036":31,"_1037":1038},"consentConfig",{"_309":310,"_311":312,"_313":314,"_315":316,"_317":318,"_319":320,"_321":322,"_323":324,"_325":326,"_327":328},"showContactAdmin",1786122320817,[],[],[304,305,306,307,308],{"_45":59,"_47":70,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":59,"_47":68,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":59,"_47":60,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":57,"_47":58,"_49":102,"_51":-5,"_53":-5,"_55":56},{"_45":46,"_47":48,"_49":102,"_51":-5,"_53":-5,"_55":56},"mainInfoText","We and our third-party partners (including social media, advertising, and analytics partners) use cookies and other tracking technologies to collect, store, monitor, and process certain information about you when you visit our website. The information collected might relate to you, your preferences, or your device. We use that information to make the site work, analyze performance and traffic on our website, provide a more personalized web experience, and assist in our marketing efforts.

Under certain privacy laws, you have the right to direct us not to \"sell\" or \"share\" your personal information for targeted advertising. To opt-out of the \"sale\" and \"sharing\" of personal information through cookies, you must opt-out of optional cookies using the toggles below. To opt out of the \"sale\" and \"sharing\" of data collected by other means (e.g., online forms) you must also update your data sharing preferences through the NVIDIA Preference Center.

Click on the different category headings below to find out more and change the settings according to your preference. You cannot opt out of Required Cookies as they are deployed to ensure the proper functioning of our website (such as prompting the cookie banner and remembering your settings, etc.). By clicking \"Save and Accept\" or \"Decline All\" at the bottom, you consent to the use of cookies and other tools as described in our Cookie Policy in accordance with your settings and accept our Terms of Service (which contains important waivers). For more information about our privacy practices, please see our Privacy Policy.","mainText","Cookie Settings","alertNoticeText",{"_1025":1026,"_1027":1028,"_1029":1030,"_1031":1032},"bannerAcceptLabel","Accept All","bannerRejectLabel","Reject Optional","bannerSettingsLabel","Manage Settings","preferenceSaveLabel","Save Choices","preferenceRejectLabel","Decline All","alwaysActiveText","Always Active","groups",[329,330,331,332,333],{"_334":715,"_78":716,"_167":717,"_338":86,"_339":16,"_341":718,"_343":719},{"_334":618,"_78":619,"_167":620,"_338":31,"_339":54,"_341":621,"_343":622},{"_334":525,"_78":526,"_167":527,"_338":31,"_339":528,"_341":529,"_343":530},{"_334":345,"_78":346,"_167":347,"_338":31,"_339":348,"_341":349,"_343":350},{"_334":335,"_78":336,"_167":337,"_338":31,"_339":340,"_341":342,"_343":344},"optanonGroupId","C0005","Social Media Cookies","These cookies are set by a range of social media services that we have added to the site to enable you to share our content with your friends and networks. They are capable of tracking your browser across other sites and building up a profile of your interests. This may impact the content and messages you see on other websites you visit. If you do not allow these cookies you may not be able to use or see these sharing tools.","alwaysActive","order",11,"cookies",[],"hosts",[],"C0004","Advertising Cookies","These cookies record your visit to our websites, the pages you have visited and the links you have followed to influence the advertisements that you see on other websites. These cookies and the information they collect may be managed by other companies, including our advertising partners, and may be used to build a profile of your interests and show you relevant advertising on other sites. We and our advertising partners will use this information to make our websites and the advertising displayed on it, more relevant to your interests.",5,[467,468,469,470,471,472,473,474,475,476,477,478,479,480,481,482,483,484,485,486,487,488],[351,352,353,354,355,356,357,358,359,360,361,362,363,364,365,366,367],{"_368":462,"_209":462,"_167":370,"_341":463},{"_368":458,"_209":458,"_167":370,"_341":459},{"_368":453,"_209":453,"_167":370,"_341":454},{"_368":449,"_209":449,"_167":370,"_341":450},{"_368":444,"_209":444,"_167":370,"_341":445},{"_368":439,"_209":439,"_167":370,"_341":440},{"_368":431,"_209":431,"_167":370,"_341":432},{"_368":423,"_209":423,"_167":370,"_341":424},{"_368":418,"_209":418,"_167":370,"_341":419},{"_368":413,"_209":413,"_167":370,"_341":414},{"_368":408,"_209":408,"_167":370,"_341":409},{"_368":403,"_209":403,"_167":370,"_341":404},{"_368":398,"_209":398,"_167":370,"_341":399},{"_368":391,"_209":391,"_167":370,"_341":392},{"_368":386,"_209":386,"_167":370,"_341":387},{"_368":379,"_209":379,"_167":370,"_341":380},{"_368":369,"_209":369,"_167":370,"_341":371},"hostName","amazon-adsystem.com","",[372,373],{"_78":378,"_375":369,"_376":31,"_167":377},{"_78":374,"_375":369,"_376":31,"_167":377},"ad-privacy","host","isSession","This domain is owned by online retailer Amazon and is used as part of its affiliate marketing programme.","ad-id","doubleclick.net",[381,382],{"_78":385,"_375":379,"_376":31,"_167":384},{"_78":383,"_375":379,"_376":31,"_167":384},"IDE","This domain is owned by Doubleclick (Google). The main business activity is: Doubleclick is Googles real time bidding advertising exchange","test_cookie","bing.com",[388],{"_78":389,"_375":386,"_376":31,"_167":390},"MUID","This domain is owned by Mircosoft - it is the site for the search engine Bing.","assets.adobedtm.com",[393,394],{"_78":397,"_375":391,"_376":31,"_167":396},{"_78":395,"_375":391,"_376":31,"_167":396},"kndctr_F207D74D549850760A4C98C6_AdobeOrg_*","If you know anything about this host and its uses please add your contributions here. ","kndctr_F207D74D549850760A4C98C6_AdobeOrg_consent","tiktok.com",[400],{"_78":401,"_375":398,"_376":31,"_167":402},"_ttp","TikTok is a social media platform owned by ByteDance, known for short-form video sharing. The service is used to create and share videos, and it utilizes cookies for various purposes including advertising and analytics.","t.co",[405],{"_78":406,"_375":403,"_376":31,"_167":407},"muc_ads","This domain is owned by Twitter. It is used as a URL shortening service, which allows Twitter to collect data on how links are shared.","snowplow.com",[410],{"_78":411,"_375":408,"_376":31,"_167":412},"sp","Testing","google.com",[415],{"_78":416,"_375":413,"_376":31,"_167":417},"NID","This domain is owned by Google Inc. Although Google is primarily known as a search engine, the company provides a diverse range of products and services. Its main source of revenue however is advertising. Google tracks users extensively both through its own products and sites, and the numerous technologies embedded into many millions of websites around the world. It uses the data gathered from most of these services to profile the interests of web users and sell advertising space to organisations based on such interest profiles as well as aligning adverts to the content on the pages where its customer's adverts appear.","linkedin.com",[420],{"_78":421,"_375":418,"_376":31,"_167":422},"bcookie","This domain is owned by LinkedIn, the business networking platform. It typically acts as a third party host where website owners have placed one of its content sharing buttons in their pages, although its content and services can be embedded in other ways. Although such buttons add functionality to the website they are on, cookies are set regardless of whether or not the visitor has an active Linkedin profile, or agreed to their terms and conditions. For this reason it is classified as a primarily tracking/targeting domain.","twitter.com",[425,426],{"_78":429,"_375":423,"_376":31,"_167":430},{"_78":427,"_375":423,"_376":31,"_167":428},"personalization_id","This domain is owned by Twitter. The main business activity is: Social Networking Services. Where twitter acts as a third party host, it collects data through a range of plug-ins and integrations, that is primarily used for tracking and targeting.","guest_*","This domain is owned by X (formerly Twitter). The main business activity is: Social Networking Services. Where X (formerly Twitter) acts as a third party host, it collects data through a range of plug-ins and integrations, that is primarily used for tracking and targeting.","youtube.com",[433,434],{"_78":437,"_375":431,"_376":31,"_167":438},{"_78":435,"_375":431,"_376":31,"_167":436},"VISITOR_INFO1_LIVE","This cookie is used as a unique identifier to track viewing of videos","__Secure-*","YouTube is a Google owned platform for hosting and sharing videos. YouTube collects user data through videos embedded in websites, which is aggregated with profile data from other Google services in order to display targeted advertising to web visitors across a broad range of their own and other websites.","demdex.net",[441],{"_78":442,"_375":439,"_376":31,"_167":443},"demdex","This cookie helps Adobe Audience Manger perform basic functions such as visitor identification, ID synchronization, segmentation, modeling, reporting, etc.","im-apps.net",[446],{"_78":447,"_375":444,"_376":31,"_167":448},"receive-cookie-deprecation","This domain is owned by im-apps.net which provides information and resources about various instant messaging (IM) applications to help users make informed decisions based on their communication needs.","influ2.com",[451],{"_78":452,"_375":449,"_376":31,"_167":412},"R","6sc.co",[455],{"_78":456,"_375":453,"_376":31,"_167":457},"6suuid","This domain is associated with the company 6Sense Insights, Inc., which provides predictive analytics and account engagement services. It is primarily used for tracking and targeting in marketing automation and advertising.","hushly.com",[460],{"_78":461,"_375":458,"_376":31,"_167":412},"_hly_gvid","t.contentsquare.net",[464],{"_78":465,"_375":462,"_376":86,"_167":466},"_cs_*","This is a session cookie. This is a pattern type cookie where a random 13-digit number is appended to the prefix _cs_.",{"_78":523,"_375":490,"_376":31,"_167":524},{"_78":521,"_375":490,"_376":31,"_167":522},{"_78":519,"_375":490,"_376":31,"_167":520},{"_78":401,"_375":490,"_376":31,"_167":518},{"_78":495,"_375":490,"_376":86,"_167":497},{"_78":517,"_375":490,"_376":31,"_167":370},{"_78":516,"_375":493,"_376":31,"_167":370},{"_78":514,"_375":490,"_376":31,"_167":515},{"_78":512,"_375":490,"_376":31,"_167":513},{"_78":511,"_375":490,"_376":31,"_167":370},{"_78":509,"_375":490,"_376":31,"_167":510},{"_78":495,"_375":508,"_376":86,"_167":497},{"_78":506,"_375":493,"_376":31,"_167":507},{"_78":506,"_375":496,"_376":31,"_167":507},{"_78":465,"_375":490,"_376":31,"_167":466},{"_78":504,"_375":490,"_376":31,"_167":505},{"_78":411,"_375":502,"_376":31,"_167":503},{"_78":500,"_375":490,"_376":31,"_167":501},{"_78":498,"_375":490,"_376":31,"_167":499},{"_78":495,"_375":496,"_376":86,"_167":497},{"_78":492,"_375":493,"_376":31,"_167":494},{"_78":489,"_375":490,"_376":86,"_167":491},"AMCVS_xxxxxAdobeOrg","nvidia.com","This is a pattern type cookie name associated with Adobe Marketing Cloud. It stores a unique visitor identifier, and uses an organisation identifier to allow a company to track users across their domains and services.This is a pattern type cookie name associated with Adobe Marketing Cloud. It stores a unique visitor identifier, and uses an organisation identifier.","_rdt_pn","www.nvidia.com","This cookie name is associated with www.mind.org.uk which is used for marketing purposes.","AMP_*","build.nvidia.com","Supports Amplitude product analytics.","_gcl_au","Used by Google AdSense for experimenting with advertisement efficiency across websites using their services","_mkto_trk","This cookie is associated with an email marketing service provided by Marketo. This tracking cookie allows a website to link visitor behaviour to the recipient of an email marketing campaign, to measure campaign effectiveness.","investor.nvidia.com","This cookie is associated with Eventbrite and is used to deliver content tailored to the end user's interests and improve content creation. This cookie is also used for event-booking purposes.","_gat_gtag_UA_140825383_2","Google Analytics","_rdt_uuid","This cookie is set by Reddit and is used for tracking the views of embedded videos and performance of the advertisements.","catalog.ngc.nvidia.com","_uetsid","This cookie is used by Bing to determine what ads should be shown that may be relevant to the end user perusing the site. ","kndctr_F207D74D549850760A4C98C6_AdobeOrg_bc_session_id","ttcsid_XXXXXXXXXXXXXXXXXXXXCategory","This cookie name is associated with TikTok which is used to track your activity on the platform, including how you interact with their embedded services on other websites.This cookie is associated with TikTok. It helps track your activity on the platform, including how you interact with their embedded services on other websites.","_fbp","Used by Facebook to deliver a series of advertisement products such as real time bidding from third party advertisers","_zitok","_twpid","This cookie is typically associated with the use of Taptap Digital, a company specializing in digital advertising. It is used for tracking and ad personalization purposes.","_tt_enable_cookie","This cookie name is associated with TikTok which is used to collect data about behavior and purchases on the website and to measure the effect of advertising. It is also used to track website activity to help optimize advertising efforts.","AMCV_xxxxxAdobeOrg","This cookie is used to identify a unique visitor for tracking purposes by Adobe Experience Cloud.This is a pattern type cookie name associated with Adobe Marketing Cloud. It stores a unique visitor identifier, and uses an organisation identifier to allow a company to track users across their domains and services.","ttcsid","This cookie name is associated with TikTok which is used to enable TikTok Ads session tracking by assigning a unique session ID to users who click on TikTok ads, allowing for the attribution of user behavior and conversions within the same session.This cookie is associated with TikTok. It helps track your activity on the platform, including how you interact with their embedded services on other websites.","C0003","Personalization Cookies","These cookies collect data about how you have interacted with our website to help us improve your web experience, such as which pages you have visited. These cookies may store a unique ID so that our system will remember you when you return. They may be set by us or by third party providers whose services we have added to our pages. These cookies enable us to provide enhanced website functionality and personalization as well as make the marketing messages we send to you more relevant to your interests. If you do not allow these cookies, then some or all of these services may not function properly.",4,[560,561,562,563,564,565,566,567,568,569,570,571,572,573,574,575,576,577,578,579,580,581,582,583,584],[531,532,533,534,535],{"_368":444,"_209":444,"_167":370,"_341":555},{"_368":418,"_209":418,"_167":370,"_341":550},{"_368":545,"_209":545,"_167":370,"_341":546},{"_368":540,"_209":540,"_167":370,"_341":541},{"_368":536,"_209":536,"_167":370,"_341":537},"nvidia.partners",[538],{"_78":539,"_375":536,"_376":31,"_167":412},"edge","app.hushly.com",[542],{"_78":543,"_375":540,"_376":31,"_167":544},"*CORS","This domain is associated with Hushly, a platform that provides lead conversion solutions for content marketing by using personalization and content engagement strategies.","github.com",[547],{"_78":548,"_375":545,"_376":31,"_167":549},"_octo","This domain is owned by Github. The main business activity is: Source code repository and collaboration platform",[551,552],{"_78":554,"_375":418,"_376":31,"_167":422},{"_78":553,"_375":418,"_376":31,"_167":422},"lidc","li_gc",[556,557],{"_78":559,"_375":444,"_376":31,"_167":448},{"_78":558,"_375":444,"_376":31,"_167":448},"imid_*","imid",{"_78":585,"_375":493,"_376":31,"_167":587},{"_78":616,"_375":490,"_376":31,"_167":617},{"_78":585,"_375":615,"_376":31,"_167":587},{"_78":613,"_375":490,"_376":86,"_167":614},{"_78":611,"_375":490,"_376":31,"_167":612},{"_78":610,"_375":593,"_376":86,"_167":370},{"_78":585,"_375":589,"_376":31,"_167":587},{"_78":608,"_375":490,"_376":31,"_167":609},{"_78":585,"_375":496,"_376":31,"_167":587},{"_78":596,"_375":493,"_376":31,"_167":370},{"_78":585,"_375":607,"_376":31,"_167":587},{"_78":605,"_375":490,"_376":31,"_167":606},{"_78":602,"_375":603,"_376":31,"_167":604},{"_78":600,"_375":601,"_376":31,"_167":370},{"_78":585,"_375":597,"_376":31,"_167":587},{"_78":599,"_375":490,"_376":31,"_167":370},{"_78":585,"_375":598,"_376":31,"_167":587},{"_78":596,"_375":597,"_376":31,"_167":370},{"_78":595,"_375":593,"_376":86,"_167":370},{"_78":585,"_375":594,"_376":31,"_167":587},{"_78":585,"_375":593,"_376":31,"_167":587},{"_78":585,"_375":592,"_376":31,"_167":587},{"_78":591,"_375":490,"_376":86,"_167":587},{"_78":588,"_375":589,"_376":86,"_167":590},{"_78":585,"_375":586,"_376":31,"_167":587},"_gd_*","info.nvidia.com","This cookie contains no identifying information and is used for diagnostic purposes by Impact Radius, a service we use to track signup commissions for our affiliate program. We have categorized it under Functionality because it is necessary for the operation of our affiliate program.","nv_country_code","developer.nvidia.com","This cookie is likely used to store the country code of the visitor to provide localized content and possibly for legal compliance with regional data privacy laws.","_gdxxxxxxxxxxxxx","nvidianews.nvidia.com","docs.api.nvidia.com","research.nvidia.com","readme_library","wcmmode","ai-podcast.nvidia.com","resources.nvidia.com","geo_locator","openedx-language-preference","learn.nvidia.com","doxygen_width","raytracing-docs.nvidia.com","This cookie is likely used to remember the width preferences of a sidebar or interface element within a documentation platform like Doxygen.","_sfid_eb78","This cookie is categorized as Functionality Cookie.","crowdsource.nvidia.com","mbox","Cookie set by Adobe SiteCatalyst analytics software - used for measuring the performance of page content using A/B split testing","readme_language","vid","This cookie name 'vid' is used for tracking the search request made by the user on the 'Find a practitioner' page.","at_check","Used by Adobe Target to check if cookies are enabled/supported on the browser","blogs.nvidia.com","nvHeadless","This cookie is likely used to track if a user is using the site in a 'headless' non-GUI mode, which could be for accessibility reasons or for automated processes.","C0002","Performance Cookies","These cookies are used to provide quantitative measures of our website visitors, such as the number of times you visit, time on page, your mouse movements, scrolling, clicks and keystroke activity on the websites; other browsing, search, or product research behavior; and what brought you to our site. These cookies may store a unique ID so that our system will remember you when you return. Information collected with these cookies is used to measure and find ways to improve website performance.",[643,644,645,646,647,648,649,650,651,652,653,654,655,656,657,658,659,660,661,662,663,664,665,666,667,668,669,670,671],[623,624,625,626],{"_368":638,"_209":638,"_167":370,"_341":639},{"_368":633,"_209":633,"_167":370,"_341":634},{"_368":431,"_209":431,"_167":370,"_341":630},{"_368":545,"_209":545,"_167":370,"_341":627},[628],{"_78":629,"_375":545,"_376":86,"_167":549},"_gh_sess",[631],{"_78":632,"_375":431,"_376":86,"_167":438},"YSC","bizible.com",[635],{"_78":636,"_375":633,"_376":31,"_167":637},"_BUID","Bizible is a marketing attribution software that tracks the performance of marketing efforts and connects them to revenue, primarily used for B2B environments.","a.usbrowserspeed.com",[640],{"_78":641,"_375":638,"_376":31,"_167":642},"tuid","This domain is owned by Ookla which provides internet speed testing services through its Speedtest application.",{"_78":713,"_375":490,"_376":31,"_167":714},{"_78":698,"_375":712,"_376":86,"_167":699},{"_78":681,"_375":493,"_376":31,"_167":606},{"_78":681,"_375":594,"_376":31,"_167":606},{"_78":710,"_375":490,"_376":31,"_167":711},{"_78":698,"_375":496,"_376":86,"_167":699},{"_78":708,"_375":490,"_376":31,"_167":709},{"_78":698,"_375":493,"_376":86,"_167":699},{"_78":706,"_375":593,"_376":31,"_167":707},{"_78":698,"_375":615,"_376":86,"_167":699},{"_78":678,"_375":493,"_376":31,"_167":680},{"_78":704,"_375":490,"_376":86,"_167":705},{"_78":702,"_375":490,"_376":31,"_167":703},{"_78":700,"_375":490,"_376":86,"_167":701},{"_78":698,"_375":589,"_376":86,"_167":699},{"_78":696,"_375":490,"_376":31,"_167":697},{"_78":694,"_375":490,"_376":31,"_167":695},{"_78":692,"_375":490,"_376":86,"_167":693},{"_78":690,"_375":598,"_376":31,"_167":691},{"_78":678,"_375":589,"_376":31,"_167":680},{"_78":688,"_375":490,"_376":31,"_167":689},{"_78":686,"_375":490,"_376":86,"_167":687},{"_78":684,"_375":490,"_376":31,"_167":685},{"_78":682,"_375":490,"_376":31,"_167":683},{"_78":681,"_375":490,"_376":31,"_167":606},{"_78":678,"_375":679,"_376":31,"_167":680},{"_78":676,"_375":490,"_376":31,"_167":677},{"_78":674,"_375":490,"_376":86,"_167":675},{"_78":672,"_375":490,"_376":31,"_167":673},"_ga_xxxxxxxxxx","Used by Google Analytics to identify and track an individual session with your device.","s_ppvl","This cookie is set by Adobe Analytics to store information on the percentage of the page displayed to the end user.","coveo_visitorId","This cookie name is associated with Coveo which is used to store a unique user ID for marketing purposes and to track the number of distinct users who submitted a query or clicked an item, counting them only once even if they did the same query or clicked the same item over the course of many visits.","_dd_s","marketplace.nvidia.com","Datadog Cookie used to group all events generated from a unique user session across multiple pages. It contains the current session ID, whether the session is excluded due to sampling, and the expiration date of the session. The cookie is extended for an extra 15 minutes every time the user interacts with the website, up to the maximum user session duration (4 hours).","nvweb_A","_ga","This cookie name is associated with Google Universal Analytics - which is a significant update to Google's more commonly used analytics service. This cookie is used to distinguish unique users by assigning a randomly generated number as a client identifier. It is included in each page request in a site and used to calculate visitor, session and campaign data for the sites analytics reports. By default it is set to expire after 2 years, although this is customisable by website owners.","_uetvid","This is a cookie utilised by Microsoft Bing Ads and is a tracking cookie. It allows us to engage with a user that has previously visited our website.","tp","This cookie is associated with Adobe Analytics site tracking. It registers if the website's scroll-depth detection is active. This function remembers how far the user has scrolled on the current session across the website's sub-pages.","_hly_vid","This cookie _hly_vid is used to record website and server performance with each visit to the site. The data is aggregated and analyzed; therefore, it remains anonymous upon use.","_session_id","This cookie creates an interim session ID used as an in-session user ID.\n","s_cc","This cookie is associated with the Adobe Site Catalyst. It determines whether cookies are enabled in the web browser.","s_ecid","Adobe Analytics cookie containing the the Adobe Experience Cloud 1st party visitor id if the Analytics AMCV visitor id is not found.","_hjSessionUser_xxxxxx","Hotjar cookie that is set when a user first lands on a page with the Hotjar script. It is used to persist the Hotjar User ID, unique to that site on the browser. This ensures that behavior in subsequent visits to the same site will be attributed to the same user ID.","_hjHasCachedUserAttributes","This cookie name is associated with Hotjar which is used to know whether the data set in _hjUserAttributes Local Storage item is up to date or not.","s_ppv","This cookie is associated with Adobe Analytics site tracking and is used to measure a user's scroll activity to see how much of a page they view before moving to another page.","_gid","This cookie name is associated with Google Universal Analytics. This appears to be a new cookie and as of Spring 2017 no information is available from Google. It appears to store and update a unique value for each page visited.","_hjTLDTest","When the Hotjar script executes we try to determine the most generic cookie path we should use, instead of the page hostname. This is done so that cookies can be shared across subdomains (where applicable). To determine this, we try to store the _hjTLDTest cookie for different URL substring alternatives until it fails. After this check, the cookie is removed.","ekfls-2025-03-27","This cookie name is associated with readme.io which is used to help maintain data for the API reference support page.","s_getNewRepeat","This cookie is used to determine whether a visitor is a new visitor or a repeat visitor to the site, and is used for analytical purposes.","_hly_sid","This cookie is likely associated with website analytics, as the prefix '_hly_' suggests a connection to Holistics, a business intelligence and analytics platform.","docs.nvidia.com","_hjSession_xxxxxx","A cookie that holds the current session data. This ensues that subsequent requests within the session window will be attributed to the same Hotjar session.","C0001","Required Cookies","These cookies enable core functionality such as security, network management, and accessibility. These cookies are required for the site to function and cannot be turned off.",[860,861,862,863,864,865,866,867,868,869,870,871,872,873,874,875,876,877,878,879,880,881,882,883,884,885,886,887,888,889,890,891,892,893,894,895,896,897,898,899,900,901,902,903,904,905,906,907,908,909,910,911,912,913,914,915,916,917,918,919,920,921,922,923,924,925,926,927,928,929,930,931,932],[720,721,722,723,724,725,726,727,728,729,730,731,732,733,734,735,736,737,738,739,740,741,742,743,744],{"_368":857,"_209":857,"_167":370,"_341":858},{"_368":851,"_209":851,"_167":370,"_341":852},{"_368":848,"_209":848,"_167":370,"_341":849},{"_368":836,"_209":836,"_167":370,"_341":837},{"_368":833,"_209":833,"_167":370,"_341":834},{"_368":830,"_209":830,"_167":370,"_341":831},{"_368":825,"_209":825,"_167":370,"_341":826},{"_368":431,"_209":431,"_167":370,"_341":822},{"_368":819,"_209":819,"_167":370,"_341":820},{"_368":423,"_209":423,"_167":370,"_341":817},{"_368":418,"_209":418,"_167":370,"_341":815},{"_368":810,"_209":810,"_167":370,"_341":811},{"_368":805,"_209":805,"_167":370,"_341":806},{"_368":800,"_209":800,"_167":370,"_341":801},{"_368":797,"_209":797,"_167":370,"_341":798},{"_368":545,"_209":545,"_167":370,"_341":794},{"_368":403,"_209":403,"_167":370,"_341":791},{"_368":540,"_209":540,"_167":370,"_341":784},{"_368":536,"_209":536,"_167":370,"_341":778},{"_368":772,"_209":772,"_167":370,"_341":773},{"_368":767,"_209":767,"_167":370,"_341":768},{"_368":762,"_209":762,"_167":370,"_341":763},{"_368":758,"_209":758,"_167":370,"_341":759},{"_368":753,"_209":753,"_167":370,"_341":754},{"_368":745,"_209":745,"_167":370,"_341":746},"polyfill.io",[747,748],{"_78":751,"_375":745,"_376":31,"_167":752},{"_78":749,"_375":745,"_376":31,"_167":750},"cf_ob_info","This cookie is associated with sites using CloudFlare. It is used to improve page load times and to override any security restrictions based on the visitors IP address. It does not contain any user identification information.","cf_use_ob","This cookie is associated with sites using CloudFlare. It is used to improve page load times and to override any security restrictions based on the visitors IP address. It does not contain any user identification information. ","www.recaptcha.net",[755],{"_78":756,"_375":753,"_376":31,"_167":757},"_GRECAPTCHA","This domain is used by Google reCAPTCHA, which provides spam protection services to websites by detecting if a user is a human or a bot.","api2.hcaptcha.com",[760],{"_78":761,"_375":758,"_376":31,"_167":412},"__cflb","instagram.com",[764],{"_78":765,"_375":762,"_376":31,"_167":766},"csrftoken","Instagram is a social networking service owned by Facebook, Inc. It's designed for sharing photos and videos from a smartphone. The service also uses cookies for various purposes, including analytics, personalization, and advertising.","readme.io",[769],{"_78":770,"_375":767,"_376":31,"_167":771},"__cf_bm","The __cf_bm cookie is a cookie necessary to support Cloudflare Bot Management, currently in private beta. As part of our bot management service, this cookie helps manage incoming traffic that matches criteria associated with bots.","nvidiacorp.us-5.evergage.com",[774,775],{"_78":777,"_375":772,"_376":31,"_167":412},{"_78":776,"_375":772,"_376":31,"_167":412},"AWSALBTG","AWSALBTGCORS",[779,780],{"_78":782,"_375":536,"_376":31,"_167":783},{"_78":781,"_375":536,"_376":31,"_167":412},"ak_bmsc","bm_mi","This domain is likely associated with NVIDIA's partner program, which could involve marketing, sales, and support services for partners.",[785,786,787],{"_78":790,"_375":540,"_376":31,"_167":412},{"_78":789,"_375":540,"_376":86,"_167":412},{"_78":788,"_375":540,"_376":31,"_167":412},"AWSALBCORS","JSESSIONID","AWSALB",[792],{"_78":770,"_375":403,"_376":31,"_167":793},"This is a CloudFoundry cookie",[795],{"_78":796,"_375":545,"_376":31,"_167":549},"logged_in","hcaptcha.com",[799],{"_78":770,"_375":797,"_376":31,"_167":793},"tracking.intentsify.io",[802],{"_78":803,"_375":800,"_376":31,"_167":804},"userId","Used to track usage of the embedded video player.","static.reo.dev",[807],{"_78":808,"_375":805,"_376":31,"_167":809},"cookietest","Common cookie name could have a number of different origins. Where this is first party and a session cookie, its most likely to do with checking to see if the browser is set to block or allow cookies.","www.datadoghq-browser-agent.com",[812],{"_78":813,"_375":810,"_376":31,"_167":814},"dd_cookie_test_","dd_cookie_test",[816],{"_78":770,"_375":418,"_376":31,"_167":793},[818],{"_78":770,"_375":423,"_376":31,"_167":793},"supabase.co",[821],{"_78":770,"_375":819,"_376":31,"_167":793},[823],{"_78":824,"_375":431,"_376":31,"_167":438},"VISITOR_PRIVACY_METADATA","dosa-prd.azurewebsites.net",[827],{"_78":828,"_375":825,"_376":31,"_167":829},"access_token","This cookie is used to obtain the permissions for users of the portal to allow for API access in order to retrieve information required to perform portal functions.","w.hcaptcha.com",[832],{"_78":770,"_375":830,"_376":31,"_167":793},"cdn.readme.io",[835],{"_78":770,"_375":833,"_376":31,"_167":793},"queue-it.net",[838,839,840,841,842],{"_78":847,"_375":836,"_376":31,"_167":370},{"_78":846,"_375":836,"_376":31,"_167":370},{"_78":845,"_375":836,"_376":31,"_167":370},{"_78":844,"_375":836,"_376":31,"_167":370},{"_78":843,"_375":836,"_376":31,"_167":370},"Queue-it-{queueId}","Queue-it-{dmonvidia }________{eventId}","Queue-it-token","queueitsoftblock_{queueId}","Queue-it-{nvidia}________{eventId}","x.com",[850],{"_78":770,"_375":848,"_376":31,"_167":771},"api.hcaptcha.com",[853,854],{"_78":856,"_375":851,"_376":31,"_167":412},{"_78":761,"_375":851,"_376":31,"_167":855},"This domain is associated with hCaptcha, a service that provides CAPTCHA and reCAPTCHA alternatives for user verification and bot protection.","hmt_id","vimeo.com",[859],{"_78":770,"_375":857,"_376":31,"_167":793},{"_78":993,"_375":490,"_376":31,"_167":1024},{"_78":1022,"_375":586,"_376":86,"_167":1023},{"_78":1015,"_375":984,"_376":86,"_167":1016},{"_78":1021,"_375":679,"_376":86,"_167":370},{"_78":1019,"_375":490,"_376":31,"_167":1020},{"_78":777,"_375":1005,"_376":31,"_167":944},{"_78":952,"_375":1014,"_376":86,"_167":954},{"_78":776,"_375":943,"_376":31,"_167":944},{"_78":955,"_375":1014,"_376":31,"_167":956},{"_78":1017,"_375":490,"_376":86,"_167":1018},{"_78":1015,"_375":991,"_376":86,"_167":1016},{"_78":949,"_375":1014,"_376":31,"_167":951},{"_78":1012,"_375":490,"_376":31,"_167":1013},{"_78":776,"_375":935,"_376":31,"_167":944},{"_78":940,"_375":1011,"_376":31,"_167":942},{"_78":788,"_375":941,"_376":31,"_167":957},{"_78":790,"_375":941,"_376":31,"_167":1008},{"_78":955,"_375":950,"_376":31,"_167":956},{"_78":1009,"_375":589,"_376":31,"_167":1010},{"_78":790,"_375":615,"_376":31,"_167":1008},{"_78":1006,"_375":490,"_376":31,"_167":1007},{"_78":782,"_375":615,"_376":31,"_167":1001},{"_78":978,"_375":594,"_376":86,"_167":370},{"_78":776,"_375":1005,"_376":31,"_167":944},{"_78":1003,"_375":943,"_376":86,"_167":1004},{"_78":967,"_375":490,"_376":31,"_167":958},{"_78":940,"_375":490,"_376":31,"_167":942},{"_78":1002,"_375":502,"_376":31,"_167":370},{"_78":782,"_375":490,"_376":31,"_167":1001},{"_78":790,"_375":935,"_376":31,"_167":1000},{"_78":998,"_375":598,"_376":86,"_167":999},{"_78":996,"_375":490,"_376":86,"_167":997},{"_78":995,"_375":490,"_376":31,"_167":370},{"_78":993,"_375":941,"_376":31,"_167":994},{"_78":992,"_375":502,"_376":31,"_167":370},{"_78":983,"_375":991,"_376":86,"_167":985},{"_78":989,"_375":594,"_376":31,"_167":990},{"_78":987,"_375":679,"_376":31,"_167":988},{"_78":986,"_375":496,"_376":31,"_167":370},{"_78":983,"_375":984,"_376":86,"_167":985},{"_78":981,"_375":950,"_376":86,"_167":982},{"_78":979,"_375":490,"_376":31,"_167":980},{"_78":978,"_375":493,"_376":86,"_167":370},{"_78":949,"_375":953,"_376":31,"_167":951},{"_78":781,"_375":968,"_376":31,"_167":958},{"_78":976,"_375":493,"_376":31,"_167":977},{"_78":974,"_375":490,"_376":31,"_167":975},{"_78":972,"_375":490,"_376":31,"_167":973},{"_78":971,"_375":490,"_376":31,"_167":370},{"_78":813,"_375":493,"_376":31,"_167":970},{"_78":967,"_375":615,"_376":31,"_167":958},{"_78":770,"_375":586,"_376":31,"_167":969},{"_78":967,"_375":968,"_376":31,"_167":958},{"_78":966,"_375":502,"_376":31,"_167":370},{"_78":965,"_375":490,"_376":31,"_167":938},{"_78":963,"_375":950,"_376":31,"_167":964},{"_78":776,"_375":961,"_376":31,"_167":962},{"_78":960,"_375":502,"_376":31,"_167":370},{"_78":959,"_375":490,"_376":31,"_167":370},{"_78":952,"_375":950,"_376":86,"_167":954},{"_78":781,"_375":615,"_376":31,"_167":958},{"_78":788,"_375":615,"_376":31,"_167":957},{"_78":955,"_375":953,"_376":31,"_167":956},{"_78":952,"_375":953,"_376":86,"_167":954},{"_78":949,"_375":950,"_376":31,"_167":951},{"_78":947,"_375":943,"_376":86,"_167":948},{"_78":945,"_375":601,"_376":31,"_167":946},{"_78":777,"_375":943,"_376":31,"_167":944},{"_78":940,"_375":941,"_376":31,"_167":942},{"_78":770,"_375":593,"_376":31,"_167":939},{"_78":937,"_375":490,"_376":31,"_167":938},{"_78":788,"_375":935,"_376":31,"_167":936},{"_78":933,"_375":490,"_376":31,"_167":934},"LoginHints","Keeps users logged in across NVIDIA domains.","api.ngc.nvidia.com","This cookie is managed by AWS and is used for load balancing.","bm_so","This cookie name is used to load the site as quickly as possible and to recognize bots using Akamai.","The __cf_bm cookie is a cookie necessary to support Cloudflare Bot Management, currently in private beta. As part of our bot management service, this cookie helps manage incoming traffic that matches criteria associated with bots.\nThis is a CloudFoundry cookie","OptanonConsent","nvid.nvidia.com","This cookie is set by the cookie compliance solution from OneTrust. It stores information about the categories of cookies the site uses and whether visitors have given or withdrawn consent for the use of each category. This enables site owners to prevent cookies in each category from being set in the users browser, when consent is not given. The cookie has a normal lifespan of one year, so that returning visitors to the site will have their preferences remembered. It contains no information that can identify the site visitor.","events.nvidia.com","This cookie is associated with the Amazon Web Services Elastic Load Balancing functionality for routing client request on the server.Used to attribute commission to affiliates when you arrive at the website from an affiliate referral link. It is set when you click on one of our links and used to let the advertiser and us know the website from which you came.","sessionid","This is a very generic cookie name that may have different purposes on different sites, but generally it will be some kind of anonymous session identifier.","_csrf","This cookie is used to prevent Cross-Site Request Forgery (CSRF) attacks, ensuring that each request made to the server is from the user's browser and is intentional.","LSKey-c$CookieConsentPolicy","enterprise-support.nvidia.com","Used to apply end-user cookie consent preferences set by our client-side utility.Custom- Short Duration. Unique ID of an Enquiry- used to validate access to case, and load exisiting data.\n","renderCtx","academicgrants.nvidia.com","Used to store site parameters in the session for reuse across requests by a single client for functionality and performance reasons.","CookieConsentPolicy","Used to apply end-user cookie consent preferences set by our client-side utility.","This cookie is managed by AWS and is used for load balancing.Used to attribute commission to affiliates when you arrive at the website from an affiliate referral link. It is set when you click on one of our links and used to let the advertiser and us know the website from which you came.","This cookie is associated with Akamai and is used to differentiate between traffic from humans and bots.","nv_co","__uxq412__ses.88c1","learn.learn.nvidia.com","This cookie is associated with the Amazon Web Services Elastic Load Balancing functionality for routing client request on the server.","pctrk","Used to count page views by unauthenticated users against license usage.","bm_s","bpazaws52gukakzc__ctrl0_ctl36_uccaptcha","bm_sv","store.nvidia.com","The __cf_bm cookie is a cookie necessary to support Cloudflare Bot Management, currently in private beta. As part of our bot management service, this cookie helps manage incoming traffic that matches criteria associated with bots.\n","This cookie name is used by DataDog to determine if cookies are supported on the website.dd_cookie_test","AMP_TLDTEST_xxxxxxxx","SESSlsd","This cookie is likely to be a session identifier used to maintain user session state across page requests.","mboxEdgeCluster","This cookie name is associated with Adobe Target which is used to communicate the correct edge cluster to at.js so that Target profiles can stay in sync as users navigate across a site.","_rdt_em","This cookie name is associated with Reddit which is used to manage session state and user interactions.","megaNavCollapsed","_abck","The _abck cookie is associated with Akamai's Bot Manager, which is used to prevent fraud and abuse by safeguarding online interactions and detecting irregularities that could suggest malicious activity, such as bot-driven attacks or hacking attempts.","idccsrf","This cookie name is associated with Salesforce which is used to track CrossSiteRequestForgery validation for certain single sign-on (SSO) flows.","ApplicationGatewayAffinity","partners.nvidia.com","This cookie is used to maintain a user session by the server in a load balanced environment, to ensure the user has a consistent experience.","AMP_TEST_xxxxxxxx","bm_lso","This cookie name is associated with Akamai which is used to load the site as quickly as possible and to recognize bots.","SSESS###########","This cookie is likely to be a session cookie used to maintain an anonymized user session by the server in a secure environment, often seen in sites built with the Drupal content management system.","nvdeveloper.nvidia.com","__uxq412__id.88c1","OptanonAlertBoxClosed","This cookie is set by websites using certain versions of the cookie law compliance solution from OneTrust. It is set after visitors have seen a cookie information notice and in some cases only when they actively close the notice down. It enables the website not to show the message more than once to a user. The cookie has a normal lifespan of one year and contains no personal information.","QueueITAccepted-SDFrts345E-V3_{eventId}","TEST_AMCV_COOKIE_WRITE","This cookie is used to test whether cookies are enabled in the browser. It is a temporary cookie used to determine if the browser accepts cookies.","XSRF-TOKEN","This cookie is written to help with site security in preventing Cross-Site Request Forgery attacks.This cookie helps with site security in preventing Cross-Site Request Forgery attacks.","AWS ELB application load balancer","This cookie is associated with Akamai. It is used to analyse traffic to determine if it is automated traffic generated by bots or a human user.","bpazaws52gukakzc__ctrl0_ctl78_list_ctl09_systemcaptcha","PHPSESSID","PHP session cookie associated with embedded content from this domain.\n","base.manager.learn.nvidia.com","bm_ss","This cookie is used to help protect websites from automated bot traffic.","AWS ELB application load balancerUsed to attribute commission to affiliates when you arrive at the website from an affiliate referral link. It is set when you click on one of our links and used to let the advertiser and us know the website from which you came.","_devzone_session","This cookie is likely used to maintain a user session on the website by the server. It is common for session identifiers to be used to keep users logged in and track their interaction with the site during a session.","docs.ngc.nvidia.com","bm_sz","This cookie is used by Akamai to optimize site performance and security. It is typically associated with the delivery of web content and services.","vcalliance.nvidia.com","ASP.NET_SessionId","General purpose platform session cookie, used by sites written with Microsoft .NET based technologies. Usually used to maintain an anonymised user session by the server.","com.adobe.alloy.getTld","This cookie name is associated with the Adobe WebSDK which is used to determine the top-most domain that can accept cookies.","SESSls","This cookie is likely to be a session identifier, used to maintain a user's session state across page requests.","c_code","BIGipServerx","This cookie name is associated with the BIG-IP product suite from company F5. Usually associated with managing sessions on load balanced servers, to ensure user requests are routed consistently to the correct server. The common root is BIGipServer most commonly followed by a domain name, usually the one that it is hosted on, but not always.","This cookie is set by websites using certain versions of the cookie law compliance solution from OneTrust. It is set after visitors have seen a cookie information notice and in some cases only when they actively close the notice down. It enables the website not to show the message more than once to a user. The cookie has a one year lifespan and contains no personal information.","default","

NVIDIA uses cookies to improve your experience on our web site. We and our third-party partners also use cookies and other tools to collect and record information you provide as well as information about your interactions with our websites for performance improvement, analytics, and to assist in marketing efforts. By clicking \"Accept All\", you consent to our use of cookies and other tools as described in our Cookie Policy. You can manage your cookie settings by clicking on \"Manage Settings.\" By continuing to use this site or by clicking one of the buttons below, you agree to our Terms of Service (which contains important waivers). Please see our Privacy Policy for more information on our privacy practices.

","gpc","

We have detected the Global Privacy Control (GPC) signal and recorded your rejection of all optional cookies on this site for this browser. You can manage your cookie settings by clicking on \"Manage Settings\". Please see our Cookie Policy for more information. To opt out of non-cookie personal information \"sales\" / \"sharing\" for targeted advertising purposes, please visit the NVIDIA Preference Center. Please see our Privacy Policy for more information on our privacy practices.

","gpcLoggedIn","

We have detected the Global Privacy Control Signal (GPC) and have opted you out of all optional cookies on this browser. You can manage your cookie settings by clicking on \"Manage Settings\". Please see our Cookie Policy for more information. We have also opted you out of \"sharing\"/\"sales\" of personal information outside of cookies. You can manage these settings in the NVIDIA NVIDIA Preference Center. Please see our Privacy Policy for more information.

","gpcLoggedInOverridden","

We have detected the Global Privacy Control Signal (GPC) and have opted you out of all optional cookies on this browser. You can manage your cookie settings by clicking on \"Manage Settings\". Please see our Cookie Policy for more information. We have also opted you out of \"sharing\"/\"sales\" of personal information outside of cookies which overrides at least one of your previous settings. You can manage them in the NVIDIA Preference Center. Please see our Privacy Policy for more information.

",{"_715":16,"_618":56,"_525":56,"_345":56,"_335":56},"hasConsented","gpcApplied","showBanner","geolocation",{"_1039":1040,"_1041":1042},"country","US","state","OR","query","categories",[1062,1063,1064,1065,1066],"optionUrls",[1050,1051,1052,1053,1054,1055,1056,1057,1058,1059,1060,1061],"total",12,"/orgs/nim/deepseek-ai/containers/deepseek-r1/-","/orgs/nim/nvidia/containers/llama-3.1-nemotron-70b-instruct/-","/orgs/nvidia/-/containers/pytorch/-","/orgs/nvidia/omniverse/collections/kit/-","/orgs/nvidia/-/collections/deepstream_sdk/-","/orgs/nvidia/omniverse/collections/kit-appstreaming-collection/-","/orgs/nvidia/research/models/stylegan3/-","/orgs/nvidia/tao/models/peoplenet/-","/orgs/nvidia/tao/models/trafficcamnet/-","/orgs/nvidia/riva/resources/riva_quickstart/-","/orgs/nvidia/-/helm-charts/gpu-operator/-","/orgs/nvidia/-/helm-charts/k8s-nim-operator/-",{"_202":1141,"_1068":1142,"_1070":56},{"_202":1106,"_1068":1107,"_1070":54},{"_202":1089,"_1068":1090,"_1070":1091},{"_202":203,"_1068":1082,"_1070":1083},{"_202":1067,"_1068":1069,"_1070":1071},"helm-charts","items",[1072,1073],"startIndex",10,{"_202":1067,"_209":1079,"_161":162,"_163":-7,"_204":1080,"_212":1076,"_238":1081,"_1078":1060},{"_202":1067,"_209":1074,"_161":162,"_163":-7,"_204":1075,"_212":1076,"_238":1077,"_1078":1061},"NVIDIA NIM Operator","k8s-nim-operator","NVIDIA",[],"url","GPU Operator","gpu-operator",[],[1084],9,{"_202":203,"_209":1085,"_161":162,"_163":1086,"_204":1087,"_212":1076,"_238":1088,"_1078":1059},"Riva Skills Quick Start","riva","riva_quickstart",[],"models",[1092,1093,1094],6,{"_202":1089,"_209":1102,"_161":162,"_163":1103,"_204":1104,"_212":1076,"_238":1105,"_1078":1056},{"_202":1089,"_209":1099,"_161":162,"_163":1096,"_204":1100,"_212":1076,"_238":1101,"_1078":1057},{"_202":1089,"_209":1095,"_161":162,"_163":1096,"_204":1097,"_212":1076,"_238":1098,"_1078":1058},"TrafficCamNet","tao","trafficcamnet",[],"PeopleNet","peoplenet",[],"StyleGAN3 pretrained models","research","stylegan3",[],"collections",[1108,1109,1110],{"_202":1106,"_209":1138,"_161":162,"_163":1112,"_204":1139,"_212":1076,"_238":1140,"_1078":1053},{"_202":1106,"_209":1135,"_161":162,"_163":-7,"_204":1136,"_212":1076,"_238":1137,"_1078":1054},{"_202":1106,"_209":1111,"_161":162,"_163":1112,"_204":1113,"_212":1076,"_238":1114,"_1078":1055},"Omniverse Kit App Streaming","omniverse","kit-appstreaming-collection",[1115,1116,1117],{"_78":1129,"_209":1130,"_206":1131,"_217":1132,"_1122":1133},{"_78":1112,"_209":1125,"_206":1126,"_217":1121,"_1122":1127},{"_78":1118,"_209":1119,"_206":1120,"_217":1121,"_1122":1123},"omniverse-dev","NVIDIA Developer Program","Join the Developer Program for access to free tools, support, and tech resources.","https://store-images.s-microsoft.com/image/apps.33936.6f1bcf42-55d0-45d1-b4b6-f08c68a50bbe.e91ca707-db64-49d6-9761-4781590a9685.7ba64fc5-7c44-4864-a5e6-7bde05126935","supportedFeatures",[1124],"UNIFIED_CATALOG","NVIDIA Omniverse","NVIDIA Omniverse is a platform of APIs, services, and software development kits (SDKs) that enable developers to build generative AI-enabled tools and applications.",[1124,1128],"SHAREABLE_TO_USER_GROUPS","nv-ai-enterprise","NVIDIA AI Enterprise","Accelerate your AI agent development","https://assets.ngc.nvidia.com/products/catalog/images/product-NVAIE.jpg",[1124,1128,1134],"NGC_CONSOLE","DeepStream SDK","deepstream_sdk",[],"Omniverse Kit (FB)","kit",[1115,1116,1117],"containers",[1143,1144,1145],{"_202":1141,"_209":1157,"_161":1150,"_163":1158,"_204":1159,"_212":1076,"_238":1160,"_1078":1050},{"_202":1141,"_209":1149,"_161":1150,"_163":162,"_204":1151,"_212":1076,"_238":1152,"_1078":1051},{"_202":1141,"_209":1146,"_161":162,"_163":-7,"_204":1147,"_212":1076,"_238":1148,"_1078":1052},"PyTorch","pytorch",[],"Llama-3.1-Nemotron-70B-Instruct","nim","llama-3.1-nemotron-70b-instruct",[1153,1115],{"_78":1154,"_209":1119,"_206":1120,"_217":1155,"_1122":1156},"nim-dev","https://assets.ngc.nvidia.com/products/catalog/images/product-nim-dev.jpg",[1124],"DeepSeek-R1","deepseek-ai","deepseek-r1",[1153,1115],"ownerOrgs",[],"memberOrgs",[],"externalOrgs",[],"apiKey","2b44f86fea50e90ffd2db29ee2028419","consentGranted","applicationId","4c857e4f-5ffa-45fa-b75e-af773c22f0ad","clientToken","pubb6927db33dae60ba5031a186cde0242a","site","datadoghq.com","service","catalog-ui","env","sessionSampleRate",100,"sessionReplaySampleRate","compressIntakeRequests","storeContextsAcrossPages","defaultPrivacyLevel","mask-user-input","trackViewsManually","trackingConsent","not-granted",{"_1190":1191,"_1192":283,"_1193":31,"_1194":1195,"_1068":1196,"_1197":1198,"_1199":1191,"_1200":31,"_1201":31,"_1202":31,"_1203":1204},"appName","ngc-catalog","appOrigin","hideOnPrivateOrg","icon","ObjectsModules2",[1205,1206,1207,1208,1209,1210,1211],"menuIcon","editor-swatches","productName","requiresAllScopes","requiresLogin","requiresOrgLevelScopesOnly","title","Catalog",{"_1203":1233,"_47":1234,"_1197":1235},{"_1226":1227,"_1203":1228,"_47":1229,"_1199":1129},{"_1203":1224,"_47":1225,"_1197":1198},{"_1203":1221,"_47":1222,"_1197":1223},{"_1203":1218,"_47":1219,"_1197":1220},{"_1203":1215,"_47":1216,"_1197":1217},{"_1203":1212,"_47":1213,"_1197":1214},"Resources","/resources","misc-library","Models","/models","misc-neural-network","Helm Charts","/helm-charts","misc-ship-wheel","Containers","/containers","3d-cube","Collections","/collections","access",[1230],"Enterprise Catalog","/enterprise",[1231,1232],"NVAIE","READ","Explore Catalog","/","common-home",{"_112":113,"_114":115,"_116":117,"_118":119,"_120":121,"_122":123},{"_111":31}]