From 5b8cb67f7bca3612000863bbaac80f0880789d00 Mon Sep 17 00:00:00 2001 From: binliunls <107988372+binliunls@users.noreply.github.com> Date: Wed, 11 Sep 2024 11:22:54 +0800 Subject: [PATCH] Udpate TensorRT benchmark results (#647) ### Description Update TensorRT benchmark results according to the latest MONAI version (1.4.0). ### Status **Ready** ### Please ensure all the checkboxes: - [x] Codeformat tests passed locally by running `./runtests.sh --codeformat`. - [ ] In-line docstrings updated. - [ ] Update `version` and `changelog` in `metadata.json` if changing an existing bundle. - [ ] Please ensure the naming rules in config files meet our requirements (please refer to: `CONTRIBUTING.md`). - [ ] Ensure versions of packages such as `monai`, `pytorch` and `numpy` are correct in `metadata.json`. - [ ] Descriptions should be consistent with the content, such as `eval_metrics` of the provided weights and TorchScript modules. - [ ] Files larger than 25MB are excluded and replaced by providing download links in `large_file.yml`. - [ ] Avoid using path that contains personal information within config files (such as use `/home/your_name/` for `"bundle_root"`). Signed-off-by: binliu --- models/pathology_nuclei_classification/configs/metadata.json | 3 ++- models/pathology_nuclei_classification/docs/README.md | 2 +- .../configs/metadata.json | 3 ++- .../docs/README.md | 2 +- models/swin_unetr_btcv_segmentation/configs/metadata.json | 3 ++- models/swin_unetr_btcv_segmentation/docs/README.md | 4 ++-- models/vista2d/configs/metadata.json | 3 ++- models/vista2d/docs/README.md | 4 ++-- models/vista3d/configs/metadata.json | 3 ++- models/vista3d/docs/README.md | 4 ++-- 10 files changed, 18 insertions(+), 13 deletions(-) diff --git a/models/pathology_nuclei_classification/configs/metadata.json b/models/pathology_nuclei_classification/configs/metadata.json index 88220ef6..b5d8f780 100644 --- a/models/pathology_nuclei_classification/configs/metadata.json +++ b/models/pathology_nuclei_classification/configs/metadata.json @@ -1,7 +1,8 @@ { "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json", - "version": "0.1.8", + "version": "0.1.9", "changelog": { + "0.1.9": "update tensorrt benchmark results", "0.1.8": "enable tensorrt", "0.1.7": "update to use monai 1.3.1", "0.1.6": "set image_only to False", diff --git a/models/pathology_nuclei_classification/docs/README.md b/models/pathology_nuclei_classification/docs/README.md index 70166910..5fb03872 100644 --- a/models/pathology_nuclei_classification/docs/README.md +++ b/models/pathology_nuclei_classification/docs/README.md @@ -144,7 +144,7 @@ This bundle supports acceleration with TensorRT. The table below displays the sp | method | torch_tf32(ms) | torch_amp(ms) | trt_tf32(ms) | trt_fp16(ms) | speedup amp | speedup tf32 | speedup fp16 | amp vs fp16| | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | -| model computation | 12.06 | 20.57 | 3.23 | 1.48 | 0.59 | 3.73 | 8.15 | 13.90 | +| model computation | 20.47 | 20.57 | 2.49 | 1.48 | 1.00 | 8.22 | 13.83 | 13.90 | | end2end | 45 | 49 | 18 | 18 | 0.92 | 2.50 | 2.50 | 2.72 | Where: diff --git a/models/pathology_nuclei_segmentation_classification/configs/metadata.json b/models/pathology_nuclei_segmentation_classification/configs/metadata.json index 6ed9c461..86e30a6f 100644 --- a/models/pathology_nuclei_segmentation_classification/configs/metadata.json +++ b/models/pathology_nuclei_segmentation_classification/configs/metadata.json @@ -1,7 +1,8 @@ { "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_hovernet_20221124.json", - "version": "0.2.5", + "version": "0.2.6", "changelog": { + "0.2.6": "update tensorrt benchmark results", "0.2.5": "enable tensorrt", "0.2.4": "update to use monai 1.3.1", "0.2.3": "remove meta_dict usage", diff --git a/models/pathology_nuclei_segmentation_classification/docs/README.md b/models/pathology_nuclei_segmentation_classification/docs/README.md index 1169625a..a68cfc1c 100644 --- a/models/pathology_nuclei_segmentation_classification/docs/README.md +++ b/models/pathology_nuclei_segmentation_classification/docs/README.md @@ -98,7 +98,7 @@ This bundle supports acceleration with TensorRT. The table below displays the sp | method | torch_tf32(ms) | torch_amp(ms) | trt_tf32(ms) | trt_fp16(ms) | speedup amp | speedup tf32 | speedup fp16 | amp vs fp16| | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | -| model computation | 27.15 | 20.14 | 19.54 | 5.63 | 1.35 | 1.39 | 4.82 | 3.58 | +| model computation | 24.55 | 20.14 | 10.85 | 5.63 | 1.22 | 2.26 | 4.36 | 3.58 | | end2end | 3451 | 3312 | 1318 | 878 | 1.04 | 2.62 | 3.93 | 3.77 | Where: diff --git a/models/swin_unetr_btcv_segmentation/configs/metadata.json b/models/swin_unetr_btcv_segmentation/configs/metadata.json index 4ef7b552..fbd6309b 100644 --- a/models/swin_unetr_btcv_segmentation/configs/metadata.json +++ b/models/swin_unetr_btcv_segmentation/configs/metadata.json @@ -1,7 +1,8 @@ { "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json", - "version": "0.5.5", + "version": "0.5.6", "changelog": { + "0.5.6": "update tensorrt benchmark results", "0.5.5": "enable tensorrt", "0.5.4": "update to use monai 1.3.1", "0.5.3": "add load_pretrain flag for infer", diff --git a/models/swin_unetr_btcv_segmentation/docs/README.md b/models/swin_unetr_btcv_segmentation/docs/README.md index 7317a1aa..7b61daa1 100644 --- a/models/swin_unetr_btcv_segmentation/docs/README.md +++ b/models/swin_unetr_btcv_segmentation/docs/README.md @@ -76,8 +76,8 @@ The `swin_unetr` bundle supports acceleration with TensorRT. The table below dis | method | torch_tf32(ms) | torch_amp(ms) | trt_tf32(ms) | trt_fp16(ms) | speedup amp | speedup tf32 | speedup fp16 | amp vs fp16| | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | -| model computation | 503.1 | 123.77 | 229.85 | 42.87 | 4.06 | 2.19 | 11.74 | 2.89 | -| end2end |5102 | 4895 | 4264 | 2835 | 1.04 | 1.20 | 1.80 | 1.73 | +| model computation | 123.64 | 123.77 | 93.22 | 42.87 | 1.00 | 1.33 | 2.88 | 2.89 | +| end2end | 5102 | 4895 | 2863 | 2835 | 1.04 | 1.78 | 1.80 | 1.73 | Where: - `model computation` means the speedup ratio of model's inference with a random input without preprocessing and postprocessing diff --git a/models/vista2d/configs/metadata.json b/models/vista2d/configs/metadata.json index e9b54193..9a870f64 100644 --- a/models/vista2d/configs/metadata.json +++ b/models/vista2d/configs/metadata.json @@ -1,7 +1,8 @@ { "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json", - "version": "0.2.5", + "version": "0.2.6", "changelog": { + "0.2.6": "update tensorrt benchmark results", "0.2.5": "add tensorrt benchmark results", "0.2.4": "enable tensorrt inference", "0.2.3": "update weights link", diff --git a/models/vista2d/docs/README.md b/models/vista2d/docs/README.md index c78b5513..38b1a26f 100644 --- a/models/vista2d/docs/README.md +++ b/models/vista2d/docs/README.md @@ -27,8 +27,8 @@ The `vista2d` bundle supports acceleration with TensorRT. The table below displa | method | torch_tf32(ms) | torch_amp(ms) | trt_tf32(ms) | trt_fp16(ms) | speedup amp | speedup tf32 | speedup fp16 | amp vs fp16| | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | -| model computation | 90.11 | 39.68 | 71.7 | 17.32 | 2.27 | 1.26 | 5.20 | 2.29 | -| end2end |1562 | 1903 | 1494 | 1440 | 0.82 | 1.05 | 1.08 | 1.32| +| model computation | 39.72 | 39.68 | 26.13 | 17.32 | 1.00 | 1.52 | 2.29 | 2.29 | +| end2end | 1562 | 1903 | 1494 | 1440 | 0.82 | 1.05 | 1.08 | 1.32| Where: - `model computation` means the speedup ratio of model's inference with a random input without preprocessing and postprocessing diff --git a/models/vista3d/configs/metadata.json b/models/vista3d/configs/metadata.json index 73de2a76..1255e7ca 100644 --- a/models/vista3d/configs/metadata.json +++ b/models/vista3d/configs/metadata.json @@ -1,7 +1,8 @@ { "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json", - "version": "0.4.6", + "version": "0.4.7", "changelog": { + "0.4.7": "update tensorrt benchmark results", "0.4.6": "add tensorrt benchmark result and remove the metric part", "0.4.5": "remove wrong path", "0.4.4": "enable tensorrt inference", diff --git a/models/vista3d/docs/README.md b/models/vista3d/docs/README.md index f7005600..0f1481ec 100644 --- a/models/vista3d/docs/README.md +++ b/models/vista3d/docs/README.md @@ -47,8 +47,8 @@ The `vista3d` bundle supports acceleration with TensorRT. The table below displa | method | torch_tf32(ms) | torch_amp(ms) | trt_tf32(ms) | trt_fp16(ms) | speedup amp | speedup tf32 | speedup fp16 | amp vs fp16| | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | -| model computation | 577.00 | 91.90 | 353.69 | 60.02 | 6.28 | 1.63 | 9.58 | 1.53 | -| end2end |6740 | 5166 | 5242 | 3386 | 1.30 | 1.29 | 1.99 | 1.53 | +| model computation | 108.53| 91.9 | 106.84 | 60.02 | 1.18 | 1.02 | 1.81 | 1.53 | +| end2end | 6740 | 5166 | 5242 | 3386 | 1.30 | 1.29 | 1.99 | 1.53 | Where: - `model computation` means the speedup ratio of model's inference with a random input without preprocessing and postprocessing