docs: add performance data to 910* benchmarks (#808)

Co-authored-by: ChongWei905 <weichong4@huawei.com>
mindspore-lab · Sep 6, 2024 · 8b21b8f · 8b21b8f
1 parent 489b57a
commit 8b21b8f
Show file tree

Hide file tree

Showing 40 changed files with 167 additions and 166 deletions.
diff --git a/benchmark_results.md b/benchmark_results.md
diff --git a/configs/convit/README.md b/configs/convit/README.md
@@ -29,9 +29,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model       | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                           | Download    |
-| ----------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------ |-------------|
-| convit_tiny | 73.79     | 91.70     | 5.71       | 256        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convit/convit_tiny-1961717e-910v2.ckpt) |
+| Model       | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                           | Download                                                                                                |
+| ----------- | --------- | --------- | ------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------- |
+| convit_tiny | 73.79     | 91.70     | 342.81  | 5.71       | 256        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convit/convit_tiny-1961717e-910v2.ckpt) |
 
 </div>
 

diff --git a/configs/convnext/README.md b/configs/convnext/README.md
@@ -25,9 +25,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model         | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                               | Download    |
-| ------------- | --------- | --------- | ---------- | ---------- | ---------------------------------------------------------------------------------------------------- |-------------|
-| convnext_tiny | 81.28     | 95.61     | 28.59      | 16         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convnext/convnext_tiny-db11dc82-910v2.ckpt) |
+| Model         | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                               | Download                                                                                                    |
+| ------------- | --------- | --------- | ------- | ---------- | ---------- | ---------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
+| convnext_tiny | 81.28     | 95.61     | 54.08   | 28.59      | 16         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convnext/convnext_tiny-db11dc82-910v2.ckpt) |
 
 
 </div>

diff --git a/configs/convnextv2/README.md b/configs/convnextv2/README.md
@@ -22,11 +22,11 @@ Our reproduced model performance on ImageNet-1K is reported as follows.
 
 performance tested on ascend 910*(8p) with graph mode
 
-<div align="center">
 
-| Model           | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                                   | Download    |
-| --------------- | --------- | --------- | ---------- | ---------- | -------------------------------------------------------------------------------------------------------- |-------------|
-| convnextv2_tiny | 82.39     | 95.95     | 28.64      | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnextv2/convnextv2_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convnextv2/convnextv2_tiny-a35b79ce-910v2.ckpt) |
+
+| Model           | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                                   | Download                                                                                                        |
+| --------------- | --------- | --------- | ------- | ---------- | ---------- | -------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------- |
+| convnextv2_tiny | 82.39     | 95.95     | 360.29  | 28.64      | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnextv2/convnextv2_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convnextv2/convnextv2_tiny-a35b79ce-910v2.ckpt) |
 
 </div>
 

diff --git a/configs/crossvit/README.md b/configs/crossvit/README.md
@@ -23,9 +23,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model      | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                            | Download    |
-| ---------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------- |-------------|
-| crossvit_9 | 73.38     | 91.51     | 8.55       | 256        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_9_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/crossvit/crossvit_9-32c69c96-910v2.ckpt) |
+| Model      | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                            | Download                                                                                                 |
+| ---------- | --------- | --------- | ------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------- |
+| crossvit_9 | 73.38     | 91.51     | 711.19  | 8.55       | 256        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_9_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/crossvit/crossvit_9-32c69c96-910v2.ckpt) |
 
 </div>
 

diff --git a/configs/densenet/README.md b/configs/densenet/README.md
@@ -40,9 +40,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model       | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                              | Download    |
-| ----------- | --------- | --------- | ---------- | ---------- | --------------------------------------------------------------------------------------------------- |-------------|
-| densenet121 | 75.67     | 92.77     | 8.06       | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/densenet/densenet_121_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/densenet/densenet121-bf4ab27f-910v2.ckpt) |
+| Model       | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                              | Download                                                                                                  |
+| ----------- | --------- | --------- | ------- | ---------- | ---------- | --------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- |
+| densenet121 | 75.67     | 92.77     | 50.55   | 8.06       | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/densenet/densenet_121_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/densenet/densenet121-bf4ab27f-910v2.ckpt) |
 
 performance tested on ascend 910(8p) with graph mode
 

diff --git a/configs/edgenext/README.md b/configs/edgenext/README.md
@@ -25,9 +25,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model             | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                                   | Download    |
-| ----------------- | --------- | --------- | ---------- | ---------- | -------------------------------------------------------------------------------------------------------- |-------------|
-| edgenext_xx_small | 70.64     | 89.75     | 1.33       | 256        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/edgenext/edgenext_xx_small_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/edgenext/edgenext_xx_small-cad13d2c-910v2.ckpt) |
+| Model             | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                                   | Download                                                                                                        |
+| ----------------- | --------- | --------- | ------- | ---------- | ---------- | -------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------- |
+| edgenext_xx_small | 70.64     | 89.75     | 295.88  | 1.33       | 256        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/edgenext/edgenext_xx_small_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/edgenext/edgenext_xx_small-cad13d2c-910v2.ckpt) |
 
 
 </div>

diff --git a/configs/efficientnet/README.md b/configs/efficientnet/README.md
@@ -35,9 +35,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model           | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                                     | Download    |
-| --------------- | --------- | --------- | ---------- | ---------- | ---------------------------------------------------------------------------------------------------------- |-------------|
-| efficientnet_b0 | 76.88     | 93.28     | 5.33       | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/efficientnet/efficientnet_b0_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/efficientnet/efficientnet_b0-f8d7aa2a-910v2.ckpt) |
+| Model           | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                                     | Download                                                                                                          |
+| --------------- | --------- | --------- | ------- | ---------- | ---------- | ---------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
+| efficientnet_b0 | 76.88     | 93.28     | 168.78  | 5.33       | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/efficientnet/efficientnet_b0_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/efficientnet/efficientnet_b0-f8d7aa2a-910v2.ckpt) |
 
 
 </div>

diff --git a/configs/googlenet/README.md b/configs/googlenet/README.md
@@ -26,9 +26,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model     | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                            | Download    |
-| --------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------- |-------------|
-| googlenet | 72.89     | 90.89     | 6.99       | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/googlenet/googlenet_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/googlenet/googlenet-de74c31d-910v2.ckpt) |
+| Model     | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                            | Download                                                                                                 |
+| --------- | --------- | --------- | ------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------- |
+| googlenet | 72.89     | 90.89     | 24.29   | 6.99       | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/googlenet/googlenet_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/googlenet/googlenet-de74c31d-910v2.ckpt) |
 
 </div>
 

diff --git a/configs/hrnet/README.md b/configs/hrnet/README.md
@@ -34,9 +34,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model     | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                        | Download    |
-| --------- | --------- | --------- | ---------- | ---------- | --------------------------------------------------------------------------------------------- |-------------|
-| hrnet_w32 | 80.66     | 95.30     | 41.30      | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/hrnet/hrnet_w32_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/hrnet/hrnet_w32-e616cdcb-910v2.ckpt) |
+| Model     | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                        | Download                                                                                             |
+| --------- | --------- | --------- | ------- | ---------- | ---------- | --------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------- |
+| hrnet_w32 | 80.66     | 95.30     | 303.01  | 41.30      | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/hrnet/hrnet_w32_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/hrnet/hrnet_w32-e616cdcb-910v2.ckpt) |
 
 
 

diff --git a/configs/inceptionv3/README.md b/configs/inceptionv3/README.md
@@ -27,9 +27,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model        | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                                 | Download    |
-| ------------ | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------------ |-------------|
-| inception_v3 | 79.25     | 94.47     | 27.20      | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/inceptionv3/inception_v3_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/inception_v3/inception_v3-61a8e9ed-910v2.ckpt) |
+| Model        | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                                 | Download                                                                                                       |
+| ------------ | --------- | --------- | ------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------- |
+| inception_v3 | 79.25     | 94.47     | 79.87   | 27.20      | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/inceptionv3/inception_v3_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/inception_v3/inception_v3-61a8e9ed-910v2.ckpt) |
 
 </div>
 

diff --git a/configs/inceptionv4/README.md b/configs/inceptionv4/README.md
@@ -23,9 +23,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model        | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                                 | Download    |
-| ------------ | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------------ |-------------|
-| inception_v4 | 80.98     | 95.25     | 42.74      | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/inceptionv4/inception_v4_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/inception_v4/inception_v4-56e798fc-910v2.ckpt) |
+| Model        | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                                 | Download                                                                                                       |
+| ------------ | --------- | --------- | ------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------- |
+| inception_v4 | 80.98     | 95.25     | 84.59   | 42.74      | 32         | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/inceptionv4/inception_v4_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/inception_v4/inception_v4-56e798fc-910v2.ckpt) |
 
 </div>
 

diff --git a/configs/mixnet/README.md b/configs/mixnet/README.md
@@ -25,9 +25,9 @@ performance tested on ascend 910*(8p) with graph mode
 
 <div align="center">
 
-| Model    | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe                                                                                        | Download    |
-| -------- | --------- | --------- | ---------- | ---------- | --------------------------------------------------------------------------------------------- |-------------|
-| mixnet_s | 75.58     | 95.54     | 4.17       | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/mixnet/mixnet_s_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/mixnet/mixnet_s-fe4fcc63-910v2.ckpt) |
+| Model    | Top-1 (%) | Top-5 (%) | ms/step | Params (M) | Batch Size | Recipe                                                                                        | Download                                                                                             |
+| -------- | --------- | --------- | ------- | ---------- | ---------- | --------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------- |
+| mixnet_s | 75.58     | 95.54     | 306.16  | 4.17       | 128        | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/mixnet/mixnet_s_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/mixnet/mixnet_s-fe4fcc63-910v2.ckpt) |
 
 
 </div>