Datasets:
GPUMODE
/
backendbench_tests

Dataset card Data Studio Files
xet
Community
6
operator name
stringclasses
180 values
used in model
stringclasses
155 values
args
stringlengths
19
5.24k
aten.sum.default
TorchBench/timm_nfnet
((T([128, 256, 48, 48], f16),), {})
aten.sum.default
TIMM/dm_nfnet_f0
((T([128, 512, 24, 24], f16),), {})
aten.sum.default
TorchBench/timm_nfnet
((T([128, 512, 24, 24], f16),), {})
aten.sum.default
TorchBench/resnet18
((T([16, 1000], f16),), {})
aten.sum.default
TorchBench/BERT_pytorch
((T([16, 128, 768], f16),), {})
aten.sum.default
TorchBench/pytorch_stargan
((T([16, 3, 128, 128], f16),), {})
aten.sum.default
TorchBench/hf_Longformer
((T([2, 1024, 50265], f16),), {})
aten.sum.default
TorchBench/hf_BigBird
((T([2, 1024, 50358], f16),), {})
aten.sum.default
TorchBench/nvidia_deeprecommender
((T([256, 197951], f16),), {})
aten.sum.default
TorchBench/timm_efficientdet
((T([2], f16),), {})
aten.sum.default
TorchBench/Background_Matting
((T([3, 1, 512, 512], f16),), {})
aten.sum.default
TorchBench/Background_Matting
((T([3, 3, 512, 512], f16),), {})
aten.sum.default
TorchBench/dcgan
((T([32, 1, 1, 1], f16),), {})
aten.sum.default
TorchBench/mnasnet1_0
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/mobilenet_v3_large
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/resnet50
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/squeezenet1_1
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/timm_efficientnet
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/timm_regnet
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/timm_resnest
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/timm_vovnet
((T([32, 1000], f16),), {})
aten.sum.default
TorchBench/timm_efficientdet
((T([3], f16),), {})
aten.sum.default
TorchBench/densenet121
((T([4, 1000], f16),), {})
aten.sum.default
TorchBench/hf_Bert
((T([4, 512, 30522], f16),), {})
aten.sum.default
TorchBench/hf_GPT2
((T([4, 512, 50257], f16),), {})
aten.sum.default
TorchBench/hf_Bart
((T([4, 512, 50265], f16),), {})
aten.sum.default
TorchBench/pytorch_struct
((T([40, 29, 30], f16),), {})
aten.sum.default
TorchBench/pytorch_struct
((T([40, 30, 60, 60], f16, stride=(0, 3600, 60, 1)),), {})
aten.sum.default
TorchBench/pytorch_struct
((T([40, 30], f16, stride=(0, 1)),), {})
aten.sum.default
TorchBench/maml_omniglot
((T([5, 5], f16),), {})
aten.sum.default
TorchBench/Super_SloMo
((T([6, 3, 352, 352], f16),), {})
aten.sum.default
TorchBench/fastNLP_Bert
((T([6, 474], f16, stride=(948, 2)),), {})
aten.sum.default
TorchBench/vgg16
((T([64, 1000], f16),), {})
aten.sum.default
TorchBench/tts_angular
((T([64, 256], f16),), {})
aten.sum.default
TorchBench/attention_is_all_you_need_pytorch
((T([7936, 9521], f16),), {})
aten.sum.default
TorchBench/resnext50_32x4d
((T([8, 1000], f16),), {})
aten.sum.default
TorchBench/timm_vision_transformer
((T([8, 1000], f16),), {})
aten.sum.default
TorchBench/yolov3
((T([8, 12096, 85], f16),), {})
aten.sum.default
TorchBench/yolov3
((T([8, 3, 12, 16, 85], f16),), {})
aten.sum.default
TorchBench/yolov3
((T([8, 3, 24, 32, 85], f16),), {})
aten.sum.default
TorchBench/yolov3
((T([8, 3, 48, 64, 85], f16),), {})
aten.sum.default
TorchBench/hf_Albert
((T([8, 512, 30000], f16),), {})
aten.sum.default
TorchBench/hf_DistilBert
((T([8, 512, 30522], f16),), {})
aten.sum.default
TorchBench/mobilenet_v2
((T([96, 1000], f16),), {})
aten.sum.default
TorchBench/LearningToPaint
((T([96, 65], f16),), {})
aten.sum.default
TorchBench/Super_SloMo
((T([], f16),), {})
aten.sum.dim_IntList
HuggingFace/GPTNeoForSequenceClassification
((T([1, 128], b8), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 10, 10, 2], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 10, 10, 3], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 20, 20, 2], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 20, 20, 3], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 40, 40, 2], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 40, 40, 3], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 5, 5, 2], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/timm_efficientdet
((T([1, 88, 80, 80, 2], f16), [-1]), {})
aten.sum.dim_IntList
TorchBench/pytorch_stargan
((T([16, 128], f16), [0]), {})
aten.sum.dim_IntList
TorchBench/pytorch_stargan
((T([16, 256], f16), [0]), {})
aten.sum.dim_IntList
TorchBench/pytorch_stargan
((T([16, 64], f16), [0]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 128, 32, 32], f16), [1]), {})
aten.sum.dim_IntList
TorchBench/timm_resnest
((T([32, 2, 128, 56, 56], f16), [1]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 128, 64, 64], f16), [1]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 256, 16, 16], f16), [1]), {})
aten.sum.dim_IntList
TorchBench/timm_resnest
((T([32, 2, 256, 28, 28], f16), [1]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 256, 32, 32], f16), [1]), {})
aten.sum.dim_IntList
TorchBench/timm_resnest
((T([32, 2, 512, 14, 14], f16), [1]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 512, 16, 16], f16), [1]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 512, 8, 8], f16), [1]), {})
aten.sum.dim_IntList
TorchBench/timm_resnest
((T([32, 2, 64, 56, 56], f16), [1]), {})
aten.sum.dim_IntList
TIMM/resnest101e
((T([32, 2, 64, 64, 64], f16), [1]), {})
aten.sum.dim_IntList
HuggingFace/GPT2ForSequenceClassification
((T([4, 1024], b8), [-1]), {})
aten.sum.dim_IntList
TorchBench/fastNLP_Bert
((T([6, 474], b8), [-1]), {})
aten.sum.dim_IntList
TorchBench/fastNLP_Bert
((T([6, 474], i64), [-1]), {})
aten.sum.dim_IntList
HuggingFace/MBartForConditionalGeneration
((T([8, 128], b8), [1]), {})
aten.sum.dim_IntList
HuggingFace/PLBartForConditionalGeneration
((T([8, 128], b8), [1]), {})
aten.tanh.default
HuggingFace/BigBird
((T([1, 1024, 3072], f16),), {})
aten.tanh.default
HuggingFace/BigBird
((T([1, 1024, 768], f16),), {})
aten.tanh.default
HuggingFace/GPTNeoForCausalLM
((T([1, 128, 8192], f16),), {})
aten.tanh.default
HuggingFace/GPTNeoForSequenceClassification
((T([1, 128, 8192], f16),), {})
aten.tanh.default
TorchBench/pytorch_CycleGAN_and_pix2pix
((T([1, 3, 256, 256], f16),), {})
aten.tanh.default
HuggingFace/DistillGPT2
((T([1, 512, 3072], f16),), {})
aten.tanh.default
HuggingFace/GoogleFnet
((T([1, 512, 3072], f16),), {})
aten.tanh.default
HuggingFace/GoogleFnet
((T([1, 512, 768], f16),), {})
aten.tanh.default
HuggingFace/BigBird
((T([1, 768], f16),), {})
aten.tanh.default
HuggingFace/GoogleFnet
((T([1, 768], f16),), {})
aten.tanh.default
TorchBench/pytorch_stargan
((T([16, 3, 128, 128], f16),), {})
aten.tanh.default
HuggingFace/LayoutLMForMaskedLM
((T([16, 768], f16),), {})
aten.tanh.default
HuggingFace/LayoutLMForSequenceClassification
((T([16, 768], f16),), {})
aten.tanh.default
TorchBench/hf_BigBird
((T([2, 1024, 3072], f16),), {})
aten.tanh.default
TorchBench/hf_BigBird
((T([2, 1024, 768], f16),), {})
aten.tanh.default
HuggingFace/AlbertForMaskedLM
((T([2, 512, 128], f16),), {})
aten.tanh.default
HuggingFace/AlbertForMaskedLM
((T([2, 512, 16384], f16),), {})
aten.tanh.default
HuggingFace/AlbertForQuestionAnswering
((T([2, 512, 16384], f16),), {})
aten.tanh.default
TorchBench/hf_BigBird
((T([2, 768], f16),), {})
aten.tanh.default
TorchBench/Background_Matting
((T([3, 1, 512, 512], f16),), {})
aten.tanh.default
HuggingFace/GPT2ForSequenceClassification
((T([4, 1024, 3072], f16),), {})
aten.tanh.default
TorchBench/hf_GPT2
((T([4, 512, 3072], f16),), {})
aten.tanh.default
TorchBench/fastNLP_Bert
((T([6, 768], f16),), {})
aten.tanh.default
TorchBench/hf_Albert
((T([8, 512, 128], f16),), {})
aten.tanh.default
TorchBench/hf_Albert
((T([8, 512, 3072], f16),), {})
aten.tanh_backward.default
HuggingFace/BigBird
((T([1, 1024, 3072], f16), T([1, 1024, 3072], f16)), {})