[{
"type": "thumb-down",
"id": "hardToUnderstand",
"label":"Hard to understand"
},{
"type": "thumb-down",
"id": "incorrectInformationOrSampleCode",
"label":"Incorrect information or sample code"
},{
"type": "thumb-down",
"id": "missingTheInformationSamplesINeed",
"label":"Missing the information/samples I need"
},{
"type": "thumb-down",
"id": "translationIssue",
"label":"Translation issue"
},{
"type": "thumb-down",
"id": "otherDown",
"label":"Other"
}]
[{
"type": "thumb-up",
"id": "easyToUnderstand",
"label":"Easy to understand"
},{
"type": "thumb-up",
"id": "solvedMyProblem",
"label":"Solved my problem"
},{
"type": "thumb-up",
"id": "otherUp",
"label":"Other"
}]
Modèles de référence compatibles
Cloud TPU fournit les modèles de référence suivants, optimisés pour un entraînement rapide et précis.
Framework |
Version majeure |
Catégorie du modèle |
Modèles de référence |
Versions compatibles |
TensorFlow |
2.x |
Classification d'images |
ResNet-2.x,
MNIST-2.x,
EfficientNet-2.x
|
2.1, 2.2, 2.3, 2.4 |
|
|
Modélisation du langage |
Transformer-2.x,
BERT-2.x |
2.1, 2.2, 2.3, 2.4 |
|
|
Modélisation du langage |
XlNet-2.x
|
2.4 |
|
|
Détection d'objets |
RetinaNet-2.x |
2.1, 2.2, 2.3, 2.4 |
|
|
Segmentation d'image |
Mask-RCNN-2.x |
2.2, 2,3 et 2,4 |
|
|
Segmentation d'image |
ShapeMask-2.x |
2,3 et 2,4 |
|
|
Systèmes de recommandation |
NCF-2.x |
2,3 et 2,4 |
|
1.x |
Classification d'images |
ResNet,
AmoebaNet,
MNasNet, MNIST
EfficientNet
|
1.15 |
|
|
Modélisation du langage |
Transformer,
BERT
|
1.15 |
|
|
Détection d'objets |
RetinaNet |
1.15 |
PyTorch |
1.x |
Classification d'images |
ResNet-PyTorch
|
1,7 |
|
|
Modélisation du langage |
FairSeq Transformer,
FairSeq RoBERTa |
1,7 |
|
|
Systèmes de recommandation |
Modèle de recommandation de deep learning (DLRM) |
1,7 |
[{
"type": "thumb-down",
"id": "hardToUnderstand",
"label":"Hard to understand"
},{
"type": "thumb-down",
"id": "incorrectInformationOrSampleCode",
"label":"Incorrect information or sample code"
},{
"type": "thumb-down",
"id": "missingTheInformationSamplesINeed",
"label":"Missing the information/samples I need"
},{
"type": "thumb-down",
"id": "translationIssue",
"label":"Translation issue"
},{
"type": "thumb-down",
"id": "otherDown",
"label":"Other"
}]
[{
"type": "thumb-up",
"id": "easyToUnderstand",
"label":"Easy to understand"
},{
"type": "thumb-up",
"id": "solvedMyProblem",
"label":"Solved my problem"
},{
"type": "thumb-up",
"id": "otherUp",
"label":"Other"
}]
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2021-02-02 UTC.