| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125 |
- checkpoints:
- - name: electricity_bin
- url: https://api.ngc.nvidia.com/v2/models/nvidia/dle/tft_base_pyt_ckpt_ds-electricity/versions/22.11.0_amp/zip
- - name: traffic_bin
- url: https://api.ngc.nvidia.com/v2/models/nvidia/dle/tft_base_pyt_ckpt_ds-traffic/versions/22.11.0_amp/zip
- configurations:
- - accelerator: none
- batch_size:
- - 1
- - 2
- - 4
- - 8
- - 16
- - 32
- - 64
- - 128
- - 256
- - 512
- - 1024
- batch_sizes: 1 2 4 8 16 32 64 128 256 512 1024
- capture_cuda_graph: 0
- checkpoint_variant: electricity_bin
- dataset: electricity_bin
- device: gpu
- export_format: onnx
- export_precision: fp32
- format: trt
- max_batch_size: 1024
- precision: fp16
- request_count: 500
- triton_gpu_engine_count: 2
- triton_max_queue_delay: 1
- triton_preferred_batch_sizes: 512 1024
- - accelerator: none
- batch_size:
- - 1
- - 2
- - 4
- - 8
- - 16
- - 32
- - 64
- - 128
- - 256
- - 512
- - 1024
- batch_sizes: 1 2 4 8 16 32 64 128 256 512 1024
- capture_cuda_graph: 0
- checkpoint_variant: traffic_bin
- dataset: traffic_bin
- device: gpu
- export_format: onnx
- export_precision: fp32
- format: trt
- max_batch_size: 1024
- precision: fp16
- request_count: 500
- triton_gpu_engine_count: 2
- triton_max_queue_delay: 1
- triton_preferred_batch_sizes: 512 1024
- - accelerator: none
- batch_size:
- - 1
- - 2
- - 4
- - 8
- - 16
- - 32
- - 64
- - 128
- - 256
- - 512
- - 1024
- batch_sizes: 1 2 4 8 16 32 64 128 256 512 1024
- capture_cuda_graph: 0
- checkpoint_variant: electricity_bin
- dataset: electricity_bin
- device: gpu
- export_format: ts-trace
- export_precision: fp32
- format: ts-trace
- max_batch_size: 1024
- precision: fp16
- request_count: 500
- triton_gpu_engine_count: 2
- triton_max_queue_delay: 1
- triton_preferred_batch_sizes: 512 1024
- - accelerator: none
- batch_size:
- - 1
- - 2
- - 4
- - 8
- - 16
- - 32
- - 64
- - 128
- - 256
- - 512
- - 1024
- batch_sizes: 1 2 4 8 16 32 64 128 256 512 1024
- capture_cuda_graph: 0
- checkpoint_variant: traffic_bin
- dataset: traffic_bin
- device: gpu
- export_format: ts-trace
- export_precision: fp32
- format: ts-trace
- max_batch_size: 1024
- precision: fp16
- request_count: 500
- triton_gpu_engine_count: 2
- triton_max_queue_delay: 1
- triton_preferred_batch_sizes: 512 1024
- container_version: '22.11'
- datasets:
- - name: electricity_bin
- - name: traffic_bin
- datasets_dir: datasets
- framework: PyTorch
- model_name: TFT
- triton_container_image: null
- triton_custom_operations: null
- triton_dockerfile: null
- triton_load_model_method: explicit
|
