Skip to content

update performance info #824

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 3 commits into
base: v0.5.0
Choose a base branch
from
Open

update performance info #824

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
4559528
fix jit bug
Apr 3, 2025
8975035
update performance data 2
Apr 7, 2025
c0d49d2
update ckpt
Apr 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions configs/bit/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,8 +70,8 @@ Our reproduced model performance on ImageNet-1K is reported as follows.
Experiments are tested on ascend 910* with mindspore 2.5.0 graph mode.

| model name | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | recipe | weight | acc@top1 | acc@top5 |
| ----------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------- | -------- | -------- |
| bit | 8 | 32 | 224x224 | O0 | 171s | 60.48 | 4232.80 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/bit/BiT_resnet50-77dcaf0f-910v2.ckpt) | 76.72 | 93.25 |
| ----------- | ----- | ---------- | ---------- |-----------| ------------- | ------- | ------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------- | -------- | -------- |
| bit | 8 | 32 | 224x224 | O2 | 171s | 60.48 | 4232.80 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/bit/BiT_resnet50-77dcaf0f-910v2.ckpt) | 76.72 | 93.25 |

### Notes
- top-1 and top-5: Accuracy reported on the validation set of ImageNet-1K.
Expand Down
4 changes: 2 additions & 2 deletions configs/ghostnet/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,8 +78,8 @@ Our reproduced model performance on ImageNet-1K is reported as follows.
Experiments are tested on ascend 910* with mindspore 2.5.0 graph mode.

| model name | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | recipe | weight | acc@top1 | acc@top5 |
| ----------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------ | -------- | -------- |
| ghostnet | 8 | 128 | 224x224 | O3 | 125s | 201.46 | 5082.89 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/ghostnet/ghostnet_050_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/ghostnet/ghostnet_050-ae7771cb-910v2.ckpt) | 65.93 | 86.65 |
| ----------- | ----- | ---------- | ---------- |-----------| ------------- | ------- | ------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------ | -------- | -------- |
| ghostnet | 8 | 128 | 224x224 | O2 | 125s | 201.46 | 5082.89 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/ghostnet/ghostnet_050_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/ghostnet/ghostnet_050-ae7771cb-910v2.ckpt) | 65.93 | 86.65 |


### Notes
Expand Down
4 changes: 2 additions & 2 deletions configs/halonet/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,8 +83,8 @@ Our reproduced model performance on ImageNet-1K is reported as follows.
Experiments are tested on ascend 910* with mindspore 2.5.0 graph mode.

| model name | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | recipe | weight | acc@top1 | acc@top5 |
| ----------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------- | -------- | -------- |
| halonet | 8 | 64 | 224x224 | O3 | 351s | 134.72 | 3800.48 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/halonet/halonet_50t_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/halonet/halonet_50t-533da6be.ckpt) | 0.10 | 0.10 |
| ----------- | ----- | ---------- | ---------- |-----------| ------------- | ------- | ------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------- | -------- | -------- |
| halonet | 8 | 64 | 224x224 | O2 | 351s | 134.72 | 3800.48 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/halonet/halonet_50t_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/halonet/halonet_50t-533da6be.ckpt) | 0.10 | 0.10 |


### Notes
Expand Down
4 changes: 2 additions & 2 deletions configs/visformer/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,8 +84,8 @@ Our reproduced model performance on ImageNet-1K is reported as follows.
Experiments are tested on ascend 910* with mindspore 2.5.0 graph mode.

| model name | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | recipe | weight | acc@top1 | acc@top5 |
| ----------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | ---------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- | -------- | -------- |
| visformer | 8 | 128 | 224x224 | O3 | 141s | 207.35 | 4938.51 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/visformer/visformer_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/visformer/visformer_tiny-df995ba4-910v2.ckpt) | 74.93 | 92.55 |
| ----------- | ----- | ---------- | ---------- |-----------| ------------- | ------- | ------- | ---------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- | -------- | -------- |
| visformer | 8 | 128 | 224x224 | O2 | 141s | 207.35 | 4938.51 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/visformer/visformer_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/visformer/visformer_tiny-df995ba4-910v2.ckpt) | 74.93 | 92.55 |


### Notes
Expand Down
4 changes: 2 additions & 2 deletions configs/volo/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,8 +85,8 @@ Our reproduced model performance on ImageNet-1K is reported as follows.
performance tested on ascend 910*(8p) with graph mode

| model name | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | recipe | weight | acc@top1 | acc@top5 |
| ----------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- |--------------------------------------------------------------------------------------------| ------------------------------------------------------------------------------------------------------- | -------- | -------- |
| volo | 8 | 128 | 224x224 | O3 | 368s | 230.05 | 4451.21 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/volo/volo_d4_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/volo/volo_d4-6c88cd33.ckpt) | 82.97 | 96.21 |
| ----------- | ----- | ---------- | ---------- |-----------| ------------- | ------- | ------- |--------------------------------------------------------------------------------------------| ------------------------------------------------------------------------------------------------------- | -------- | -------- |
| volo | 8 | 128 | 224x224 | O2 | 368s | 230.05 | 4451.21 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/volo/volo_d4_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/volo/volo_d4-6c88cd33.ckpt) | 82.97 | 96.21 |


performance tested on ascend 910(8p) with graph mode
Expand Down