def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
model = PretrainedModelInfo(
pretrained_model_name="Joint_Intent_Slot_Assistant",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemonlpmodels/versions/1.0.0a5/files/Joint_Intent_Slot_Assistant.nemo",
description="This models is trained on this https://github.com/xliuhw/NLU-Evaluation-Data dataset which includes 64 various intents and 55 slots. Final Intent accuracy is about 87%, Slot accuracy is about 89%.",
)
result.append(model)
return result
def list_available_models(cls) -> List[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
model = PretrainedModelInfo(
pretrained_model_name="speakerverification_speakernet",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/speakerverification_speakernet/versions/1.0.0rc1/files/speakerverification_speakernet.nemo",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:speakerverification_speakernet",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="ecapa_tdnn",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/ecapa_tdnn/versions/v1/files/ecapa_tdnn.nemo",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:ecapa_tdnn",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="titanet_large",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/titanet_large/versions/v0/files/titanet-l.nemo",
description=
"For details about this model, please visit https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/titanet_large",
)
result.append(model)
return result
def list_available_models(cls) -> 'List[PretrainedModelInfo]':
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
list_of_models = []
model = PretrainedModelInfo(
pretrained_model_name="UniGlow-22050Hz",
location=
"https://drive.google.com/file/d/18JO5heoz1pBicZnGGqJzAJYMpzxiDQDa/view?usp=sharing",
description=
"The model is trained on LJSpeech sampled at 22050Hz, and can be used as an universal vocoder",
)
list_of_models.append(model)
return list_of_models
def list_available_models(cls) -> 'List[PretrainedModelInfo]':
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
list_of_models = []
model = PretrainedModelInfo(
pretrained_model_name="tts_en_fastpitch",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/tts_en_fastpitch/versions/1.8.1/files/tts_en_fastpitch_align.nemo",
description="This model is trained on LJSpeech sampled at 22050Hz with and can be used to generate female English voices with an American accent.",
class_=cls,
)
list_of_models.append(model)
return list_of_models
def list_available_models(cls) -> 'Optional[Dict[str, str]]':
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
list_of_models = []
model = PretrainedModelInfo(
pretrained_model_name="tts_melgan",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/tts_melgan/versions/1.0.0/files/tts_melgan.nemo",
description=
"This model is trained on LJSpeech sampled at 22050Hz, and has been tested on generating female English voices with an American accent.",
class_=cls,
)
list_of_models.append(model)
return list_of_models
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
model = PretrainedModelInfo(
pretrained_model_name="ContextNet-192-WPE-1024-8x-Stride",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/ContextNet-192-WPE-1024-8x-Stride.nemo",
description=
"ContextNet initial implementation with CTC loss model trained on the Librispeech corpus and achieves a WER of 10.09% on test-other and 10.11% on dev-other.",
)
result.append(model)
return result
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
result.append(
PretrainedModelInfo(
pretrained_model_name="sgdqa_bertbasecased",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/sgdqa_bertbasecased/versions/1.0.0/files/sgdqa_bertbasecased.nemo",
description="Dialogue State Tracking model finetuned from NeMo BERT Base Cased on Google SGD dataset which has a joint goal accuracy of 59.72% on dev set and 45.85% on test set.",
)
)
return result
def list_available_models(cls) -> 'List[PretrainedModelInfo]':
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
list_of_models = []
model = PretrainedModelInfo(
pretrained_model_name="WaveGlow-22050Hz",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemottsmodels/versions/1.0.0a5/files/WaveGlow-22050Hz.nemo",
description=
"This model is trained on LJSpeech sampled at 22050Hz, and can be used as an universal vocoder.",
class_=cls,
)
list_of_models.append(model)
return list_of_models
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
model = PretrainedModelInfo(
pretrained_model_name="NERModel",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemonlpmodels/versions/1.0.0a5/files/NERModel.nemo",
description=
"The model was trained on GMB (Groningen Meaning Bank) corpus for entity recognition and achieves 74.61 F1 Macro score.",
)
result.append(model)
return result
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
results = []
model = PretrainedModelInfo(
pretrained_model_name="stt_en_citrinet_256",
description="For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_citrinet_256",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_citrinet_256/versions/1.0.0rc1/files/stt_en_citrinet_256.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_citrinet_512",
description="For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_citrinet_512",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_citrinet_512/versions/1.0.0rc1/files/stt_en_citrinet_512.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_citrinet_1024",
description="For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_citrinet_1024",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_citrinet_1024/versions/1.0.0rc1/files/stt_en_citrinet_1024.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_conformer_ctc_small",
description="For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_conformer_ctc_small",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_conformer_ctc_small/versions/1.0.0rc1/files/stt_en_conformer_ctc_small.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_conformer_ctc_medium",
description="For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_conformer_ctc_medium",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_conformer_ctc_medium/versions/1.0.0rc1/files/stt_en_conformer_ctc_medium.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_conformer_ctc_large",
description="For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_conformer_ctc_large",
location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_conformer_ctc_large/versions/1.0.0rc1/files/stt_en_conformer_ctc_large.nemo",
)
results.append(model)
return results
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
results = []
model = PretrainedModelInfo(
pretrained_model_name="stt_zh_conformer_transducer_large",
description=
"For details about this model, please visit https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/stt_zh_conformer_transducer_large",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_zh_conformer_transducer_large/versions/1.8.0/files/stt_zh_conformer_transducer_large.nemo",
)
results.append(model)
return results
def list_available_models(cls) -> Optional[List[PretrainedModelInfo]]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-3x1x64-v1",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet-3x1x64-v1.nemo",
description=
"MatchboxNet model trained on Google Speech Commands dataset (v1, 30 classes) which obtains 97.32% accuracy on test set.",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-3x2x64-v1",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet-3x2x64-v1.nemo",
description=
"MatchboxNet model trained on Google Speech Commands dataset (v1, 30 classes) which obtains 97.68% accuracy on test set.",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-3x1x64-v2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet-3x1x64-v2.nemo",
description=
"MatchboxNet model trained on Google Speech Commands dataset (v2, 35 classes) which obtains 97.12% accuracy on test set.",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-3x1x64-v2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet-3x1x64-v2.nemo",
description=
"MatchboxNet model trained on Google Speech Commands dataset (v2, 30 classes) which obtains 97.29% accuracy on test set.",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-3x1x64-v2-subset-task",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet-3x1x64-v2-subset-task.nemo",
description=
"MatchboxNet model trained on Google Speech Commands dataset (v2, 10+2 classes) which obtains 98.2% accuracy on test set.",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-3x2x64-v2-subset-task",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet-3x2x64-v2-subset-task.nemo",
description=
"MatchboxNet model trained on Google Speech Commands dataset (v2, 10+2 classes) which obtains 98.4% accuracy on test set.",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="MatchboxNet-VAD-3x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/MatchboxNet_VAD_3x2.nemo",
description=
"Voice Activity Detection MatchboxNet model trained on google speech command (v2) and freesound background data, which obtains 0.992 accuracy on testset from same source and 0.852 TPR for FPR=0.315 on testset (ALL) of AVA movie data",
)
result.append(model)
return result
def list_available_models(cls) -> Optional[List[PretrainedModelInfo]]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
results = []
model = PretrainedModelInfo(
pretrained_model_name="vad_telephony_marblenet",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:vad_telephony_marblenet",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/vad_telephony_marblenet/versions/1.0.0rc1/files/vad_telephony_marblenet.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="vad_marblenet",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:vad_marblenet",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/vad_marblenet/versions/1.0.0rc1/files/vad_marblenet.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="commandrecognition_en_matchboxnet3x1x64_v1",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:commandrecognition_en_matchboxnet3x1x64_v1",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/commandrecognition_en_matchboxnet3x1x64_v1/versions/1.0.0rc1/files/commandrecognition_en_matchboxnet3x1x64_v1.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="commandrecognition_en_matchboxnet3x2x64_v1",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:commandrecognition_en_matchboxnet3x2x64_v1",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/commandrecognition_en_matchboxnet3x2x64_v1/versions/1.0.0rc1/files/commandrecognition_en_matchboxnet3x2x64_v1.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="commandrecognition_en_matchboxnet3x1x64_v2",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:commandrecognition_en_matchboxnet3x1x64_v2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/commandrecognition_en_matchboxnet3x1x64_v2/versions/1.0.0rc1/files/commandrecognition_en_matchboxnet3x1x64_v2.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="commandrecognition_en_matchboxnet3x2x64_v2",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:commandrecognition_en_matchboxnet3x2x64_v2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/commandrecognition_en_matchboxnet3x2x64_v2/versions/1.0.0rc1/files/commandrecognition_en_matchboxnet3x2x64_v2.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name=
"commandrecognition_en_matchboxnet3x1x64_v2_subset_task",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:commandrecognition_en_matchboxnet3x1x64_v2_subset_task",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/commandrecognition_en_matchboxnet3x1x64_v2_subset_task/versions/1.0.0rc1/files/commandrecognition_en_matchboxnet3x1x64_v2_subset_task.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name=
"commandrecognition_en_matchboxnet3x2x64_v2_subset_task",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:commandrecognition_en_matchboxnet3x2x64_v2_subset_task",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/commandrecognition_en_matchboxnet3x2x64_v2_subset_task/versions/1.0.0rc1/files/commandrecognition_en_matchboxnet3x2x64_v2_subset_task.nemo",
)
results.append(model)
return results
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv1.1_bertbase",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv1_1_bertbase/versions/1.0.0rc1/files/qa_squadv1.1_bertbase.nemo",
description=
"Question answering model finetuned from NeMo BERT Base Uncased on SQuAD v1.1 dataset which obtains an exact match (EM) score of 82.78% and an F1 score of 89.97%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv2.0_bertbase",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv2_0_bertbase/versions/1.0.0rc1/files/qa_squadv2.0_bertbase.nemo",
description=
"Question answering model finetuned from NeMo BERT Base Uncased on SQuAD v2.0 dataset which obtains an exact match (EM) score of 75.04% and an F1 score of 78.08%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv1_1_bertlarge",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv1_1_bertlarge/versions/1.0.0rc1/files/qa_squadv1.1_bertlarge.nemo",
description=
"Question answering model finetuned from NeMo BERT Large Uncased on SQuAD v1.1 dataset which obtains an exact match (EM) score of 85.44% and an F1 score of 92.06%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv2.0_bertlarge",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv2_0_bertlarge/versions/1.0.0rc1/files/qa_squadv2.0_bertlarge.nemo",
description=
"Question answering model finetuned from NeMo BERT Large Uncased on SQuAD v2.0 dataset which obtains an exact match (EM) score of 80.22% and an F1 score of 83.05%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv1_1_megatron_cased",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv1_1_megatron_cased/versions/1.0.0rc1/files/qa_squadv1.1_megatron_cased.nemo",
description=
"Question answering model finetuned from Megatron Cased on SQuAD v1.1 dataset which obtains an exact match (EM) score of 88.18% and an F1 score of 94.07%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv2.0_megatron_cased",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv2_0_megatron_cased/versions/1.0.0rc1/files/qa_squadv2.0_megatron_cased.nemo",
description=
"Question answering model finetuned from Megatron Cased on SQuAD v2.0 dataset which obtains an exact match (EM) score of 84.73% and an F1 score of 87.89%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv1.1_megatron_uncased",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv1_1_megatron_uncased/versions/1.0.0rc1/files/qa_squadv1.1_megatron_uncased.nemo",
description=
"Question answering model finetuned from Megatron Unased on SQuAD v1.1 dataset which obtains an exact match (EM) score of 87.61% and an F1 score of 94.00%.",
))
result.append(
PretrainedModelInfo(
pretrained_model_name="qa_squadv2.0_megatron_uncased",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/qa_squadv2_0_megatron_uncased/versions/1.0.0rc1/files/qa_squadv2.0_megatron_uncased.nemo",
description=
"Question answering model finetuned from Megatron Uncased on SQuAD v2.0 dataset which obtains an exact match (EM) score of 84.48% and an F1 score of 87.65%.",
))
return result
def list_available_models(cls) -> Optional[PretrainedModelInfo]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
results = []
model = PretrainedModelInfo(
pretrained_model_name="QuartzNet15x5Base-En",
description=
"QuartzNet15x5 model trained on six datasets: LibriSpeech, Mozilla Common Voice (validated clips from en_1488h_2019-12-10), WSJ, Fisher, Switchboard, and NSC Singapore English. It was trained with Apex/Amp optimization level O1 for 600 epochs. The model achieves a WER of 3.79% on LibriSpeech dev-clean, and a WER of 10.05% on dev-other. Please visit https://ngc.nvidia.com/catalog/models/nvidia:nemospeechmodels for further details.",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemospeechmodels/versions/1.0.0a5/files/QuartzNet15x5Base-En.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_quartznet15x5/versions/1.0.0rc1/files/stt_en_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_zh_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_zh_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_zh_quartznet15x5/versions/1.0.0rc1/files/stt_zh_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_en_jasper10x5dr",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_jasper10x5dr",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_en_jasper10x5dr/versions/1.0.0rc1/files/stt_en_jasper10x5dr.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_ca_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_ca_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_ca_quartznet15x5/versions/1.0.0rc1/files/stt_ca_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_it_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_it_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_it_quartznet15x5/versions/1.0.0rc1/files/stt_it_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_fr_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_fr_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_fr_quartznet15x5/versions/1.0.0rc1/files/stt_fr_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_es_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_es_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_es_quartznet15x5/versions/1.0.0rc1/files/stt_es_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_de_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_de_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_de_quartznet15x5/versions/1.0.0rc1/files/stt_de_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_pl_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_pl_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_pl_quartznet15x5/versions/1.0.0rc1/files/stt_pl_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_ru_quartznet15x5",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_ru_quartznet15x5",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_ru_quartznet15x5/versions/1.0.0rc1/files/stt_ru_quartznet15x5.nemo",
)
results.append(model)
model = PretrainedModelInfo(
pretrained_model_name="stt_zh_citrinet_512",
description=
"For details about this model, please visit https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_zh_citrinet_512",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/stt_zh_citrinet_512/versions/1.0.0rc1/files/stt_zh_citrinet_512.nemo",
)
results.append(model)
return results
def list_available_models(cls) -> Optional[Dict[str, str]]:
"""
This method returns a list of pre-trained model which can be instantiated directly from NVIDIA's NGC cloud.
Returns:
List of available pre-trained models.
"""
result = []
model = PretrainedModelInfo(
pretrained_model_name="nmt_en_de_transformer12x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_en_de_transformer12x2/versions/1.0.0rc1/files/nmt_en_de_transformer12x2.nemo",
description=
"En->De translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_en_de_transformer12x2",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_de_en_transformer12x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_de_en_transformer12x2/versions/1.0.0rc1/files/nmt_de_en_transformer12x2.nemo",
description=
"De->En translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_de_en_transformer12x2",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_en_es_transformer12x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_en_es_transformer12x2/versions/1.0.0rc1/files/nmt_en_es_transformer12x2.nemo",
description=
"En->Es translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_en_es_transformer12x2",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_es_en_transformer12x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_es_en_transformer12x2/versions/1.0.0rc1/files/nmt_es_en_transformer12x2.nemo",
description=
"Es->En translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_es_en_transformer12x2",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_en_fr_transformer12x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_en_fr_transformer12x2/versions/1.0.0rc1/files/nmt_en_fr_transformer12x2.nemo",
description=
"En->Fr translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_en_fr_transformer12x2",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_fr_en_transformer12x2",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_fr_en_transformer12x2/versions/1.0.0rc1/files/nmt_fr_en_transformer12x2.nemo",
description=
"Fr->En translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_fr_en_transformer12x2",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_en_ru_transformer6x6",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_en_ru_transformer6x6/versions/1.0.0rc1/files/nmt_en_ru_transformer6x6.nemo",
description=
"En->Ru translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_en_ru_transformer6x6",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_ru_en_transformer6x6",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_ru_en_transformer6x6/versions/1.0.0rc1/files/nmt_ru_en_transformer6x6.nemo",
description=
"Ru->En translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_ru_en_transformer6x6",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_zh_en_transformer6x6",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_zh_en_transformer6x6/versions/1.0.0rc1/files/nmt_zh_en_transformer6x6.nemo",
description=
"Zh->En translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_zh_en_transformer6x6",
)
result.append(model)
model = PretrainedModelInfo(
pretrained_model_name="nmt_en_zh_transformer6x6",
location=
"https://api.ngc.nvidia.com/v2/models/nvidia/nemo/nmt_en_zh_transformer6x6/versions/1.0.0rc1/files/nmt_en_zh_transformer6x6.nemo",
description=
"En->Zh translation model. See details here: https://ngc.nvidia.com/catalog/models/nvidia:nemo:nmt_en_zh_transformer6x6",
)
result.append(model)
return result
