Skip to content
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

docs: add requirements and graph compile time to readmes #813

Merged
merged 4 commits into from
Nov 22, 2024
Merged

docs: add requirements and graph compile time to readmes #813

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
bd1dbd5
docs: add requirements and renew forms for readmes
Sep 3, 2024
65f41ec
fix: change ops.pad input format which changed in mindspore in Nov 2022
Oct 14, 2024
7bc4a44
docs: add requirements and renew forms for example models (ssd and de…
Nov 12, 2024
25f5e62
docs: fix readme bugs
Nov 14, 2024
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
14 changes: 7 additions & 7 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,13 +29,13 @@ MindCV is an open-source toolbox for computer vision research and development ba

The following is the corresponding `mindcv` versions and supported `mindspore` versions.

| mindcv | mindspore |
|:------:|:----------:|
| main | master |
| v0.4.0 | 2.3.0 |
| 0.3.0 | 2.2.10 |
| 0.2 | 2.0 |
| 0.1 | 1.8 |
| mindcv | mindspore |
| :----: | :---------: |
| main | master |
| v0.4.0 | 2.3.0/2.3.1 |
| 0.3.0 | 2.2.10 |
| 0.2 | 2.0 |
| 0.1 | 1.8 |


### Major Features
Expand Down
203 changes: 102 additions & 101 deletions benchmark_results.md
View file
Open in desktop

Large diffs are not rendered by default.

28 changes: 14 additions & 14 deletions configs/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,24 +31,24 @@ Please follow the outline structure and **table format** shown in [densenet/READ

#### Table Format

<div align="center">

| model | top-1 (%) | top-5 (%) | params (M) | batch size | cards | ms/step | jit_level | recipe | download |
| ----------- | --------- | --------- | ---------- | ---------- | ----- | ------- | --------- | --------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- |
| densenet121 | 75.67 | 92.77 | 8.06 | 32 | 8 | 47,34 | O2 | [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) |

</div>
| model name | params(M) | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | acc@top1 | acc@top5 | recipe | weight |
| ----------- | --------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | -------- | -------- | --------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- |
| densenet121 | 8.06 | 8 | 32 | 224x224 | O2 | 300s | 47,34 | 5446.81 | 75.67 | 92.77 | [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) |



Illustration:
- Model: model name in lower case with _ seperator.
- Top-1 and Top-5: Accuracy reported on the validatoin set of ImageNet-1K. Keep 2 digits after the decimal point.
- Params (M): # of model parameters in millions (10^6). Keep **2 digits** after the decimal point
- Batch Size: Training batch size
- Cards: # of cards
- Ms/step: Time used on training per step in ms
- Jit_level: Jit level of mindspore context, which contains 3 levels: O0/O1/O2
- Recipe: Training recipe/configuration linked to a yaml config file.
- Download: url of the pretrained model weights
- model name: model name in lower case with _ seperator.
- top-1 and top-5: Accuracy reported on the validatoin set of ImageNet-1K. Keep 2 digits after the decimal point.
- params(M): # of model parameters in millions (10^6). Keep **2 digits** after the decimal point
- batch size: Training batch size
- cards: # of cards
- ms/step: Time used on training per step in ms
- jit level: Jit level of mindspore context, which contains 3 levels: O0/O1/O2
- recipe: Training recipe/configuration linked to a yaml config file.
- weight: url of the pretrained model weights

### Model Checkpoint Format
The checkpoint (i.e., model weight) name should follow this format: **{model_name}_{specification}-{sha256sum}.ckpt**, e.g., `poolformer_s12-5be5c4e4.ckpt`.
Expand Down
49 changes: 25 additions & 24 deletions configs/bit/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@

> [Big Transfer (BiT): General Visual Representation Learning](https://arxiv.org/abs/1912.11370)


## Introduction

Transfer of pre-trained representations improves sample efficiency and simplifies hyperparameter tuning when training deep neural networks for vision.
Expand All @@ -12,30 +13,10 @@ is required. 3) Long pre-training time: Pretraining on a larger dataset requires
BiT use GroupNorm combined with Weight Standardisation instead of BatchNorm. Since BatchNorm performs worse when the number of images on each accelerator is
too low. 5) With BiT fine-tuning, good performance can be achieved even if there are only a few examples of each type on natural images.[[1, 2](#References)]


## Results

Our reproduced model performance on ImageNet-1K is reported as follows.

- ascend 910* with graph mode

*coming soon*

- ascend 910 with graph mode


<div align="center">


| model | top-1 (%) | top-5 (%) | params(M) | batch size | cards | ms/step | jit_level | recipe | download |
| ------------ | --------- | --------- | --------- | ---------- | ----- |---------| --------- | ---------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------- |
| bit_resnet50 | 76.81 | 93.17 | 25.55 | 32 | 8 | 74.52 | O2 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/bit/BiT_resnet50-1e4795a4.ckpt) |


</div>

#### Notes
- Top-1 and Top-5: Accuracy reported on the validation set of ImageNet-1K.
## Requirements
| mindspore | ascend driver | firmware | cann toolkit/kernel |
| :-------: | :-----------: | :---------: | :-----------------: |
| 2.3.1 | 24.1.RC2 | 7.3.0.1.231 | 8.0.RC2.beta1 |

## Quick Start

Expand Down Expand Up @@ -82,6 +63,26 @@ To validate the accuracy of the trained model, you can use `validate.py` and par
python validate.py -c configs/bit/bit_resnet50_ascend.yaml --data_dir /path/to/imagenet --ckpt_path /path/to/ckpt
```

## Performance

Our reproduced model performance on ImageNet-1K is reported as follows.

Experiments are tested on ascend 910* with mindspore 2.3.1 graph mode.

*coming soon*

Experiments are tested on ascend 910 with mindspore 2.3.1 graph mode.


| model name | params(M) | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | acc@top1 | acc@top5 | recipe | weight |
| ------------ | --------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | -------- | -------- | ---------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------- |
| bit_resnet50 | 25.55 | 8 | 32 | 224x224 | O2 | 146s | 74.52 | 3413.33 | 76.81 | 93.17 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/bit/BiT_resnet50-1e4795a4.ckpt) |



### Notes
- top-1 and top-5: Accuracy reported on the validation set of ImageNet-1K.

## References

<!--- Guideline: Citation format should follow GB/T 7714. -->
Expand Down
44 changes: 22 additions & 22 deletions configs/cmt/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,37 +2,20 @@

> [CMT: Convolutional Neural Networks Meet Vision Transformers](https://arxiv.org/abs/2107.06263)


## Introduction

CMT is a method to make full use of the advantages of CNN and transformers so that the model could capture long-range
dependencies and extract local information. In addition, to reduce computation cost, this method use lightweight MHSA(multi-head self-attention)
and depthwise convolution and pointwise convolution like MobileNet. By combing these parts, CMT could get a SOTA performance
on ImageNet-1K dataset.


## Results

Our reproduced model performance on ImageNet-1K is reported as follows.

- ascend 910* with graph mode

*coming soon*

- ascend 910 with graph mode

<div align="center">


| model | top-1 (%) | top-5 (%) | params(M) | batch size | cards | ms/step | jit_level | recipe | download |
| --------- | --------- | --------- | --------- | ---------- | ----- |---------| --------- | ------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------ |
| cmt_small | 83.24 | 96.41 | 26.09 | 128 | 8 | 500.64 | O2 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/cmt/cmt_small_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/cmt/cmt_small-6858ee22.ckpt) |
## Requirements
| mindspore | ascend driver | firmware | cann toolkit/kernel |
| :-------: | :-----------: | :---------: | :-----------------: |
| 2.3.1 | 24.1.RC2 | 7.3.0.1.231 | 8.0.RC2.beta1 |


</div>

#### Notes
- Top-1 and Top-5: Accuracy reported on the validation set of ImageNet-1K.

## Quick Start

### Preparation
Expand Down Expand Up @@ -78,6 +61,23 @@ To validate the accuracy of the trained model, you can use `validate.py` and par
python validate.py -c configs/cmt/cmt_small_ascend.yaml --data_dir /path/to/imagenet --ckpt_path /path/to/ckpt
```

## Performance

Our reproduced model performance on ImageNet-1K is reported as follows.

Experiments are tested on ascend 910* with mindspore 2.3.1 graph mode.

*coming soon*

Experiments are tested on ascend 910 with mindspore 2.3.1 graph mode.

| model name | params(M) | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | acc@top1 | acc@top5 | recipe | weight |
| ---------- | --------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | -------- | -------- | ------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------ |
| cmt_small | 26.09 | 8 | 128 | 224x224 | O2 | 1268s | 500.64 | 2048.01 | 83.24 | 96.41 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/cmt/cmt_small_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/cmt/cmt_small-6858ee22.ckpt) |

### Notes
- top-1 and top-5: Accuracy reported on the validation set of ImageNet-1K.

## References

<!--- Guideline: Citation format should follow GB/T 7714. -->
Expand Down
49 changes: 28 additions & 21 deletions configs/coat/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,28 +6,11 @@

Co-Scale Conv-Attentional Image Transformer (CoaT) is a Transformer-based image classifier equipped with co-scale and conv-attentional mechanisms. First, the co-scale mechanism maintains the integrity of Transformers' encoder branches at individual scales, while allowing representations learned at different scales to effectively communicate with each other. Second, the conv-attentional mechanism is designed by realizing a relative position embedding formulation in the factorized attention module with an efficient convolution-like implementation. CoaT empowers image Transformers with enriched multi-scale and contextual modeling capabilities.

## Results
## Requirements
| mindspore | ascend driver | firmware | cann toolkit/kernel |
| :-------: | :-----------: | :---------: | :-----------------: |
| 2.3.1 | 24.1.RC2 | 7.3.0.1.231 | 8.0.RC2.beta1 |

Our reproduced model performance on ImageNet-1K is reported as follows.

- ascend 910* with graph mode

*coming soon*


- ascend 910 with graph mode

<div align="center">


| model | top-1 (%) | top-5 (%) | params (M) | batch size | cards | ms/step | jit_level | recipe | Weight |
| --------- | --------- | --------- | ---------- | ---------- | ----- |---------| --------- | -------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
| coat_tiny | 79.67 | 94.88 | 5.50 | 32 | 8 | 254.95 | O2 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_tiny-071cb792.ckpt) |

</div>

#### Notes
- Top-1 and Top-5: Accuracy reported on the validation set of ImageNet-1K.


## Quick Start
Expand Down Expand Up @@ -74,6 +57,30 @@ To validate the accuracy of the trained model, you can use `validate.py` and par
python validate.py -c configs/coat/coat_lite_tiny_ascend.yaml --data_dir /path/to/imagenet --ckpt_path /path/to/ckpt
```

## Performance

Our reproduced model performance on ImageNet-1K is reported as follows.

Experiments are tested on ascend 910* with mindspore 2.3.1 graph mode.

*coming soon*


Experiments are tested on ascend 910 with mindspore 2.3.1 graph mode.




| model name | params(M) | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | acc@top1 | acc@top5 | recipe | weight |
| ---------- | --------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | -------- | -------- | -------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
| coat_tiny | 5.50 | 8 | 32 | 224x224 | O2 | 543s | 254.95 | 1003.92 | 79.67 | 94.88 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_tiny-071cb792.ckpt) |



### Notes
- top-1 and top-5: Accuracy reported on the validation set of ImageNet-1K.


## References

[1] Han D, Yun S, Heo B, et al. Rethinking channel dimensions for efficient model design[C]//Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition. 2021: 732-741.
53 changes: 25 additions & 28 deletions configs/convit/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
# ConViT
> [ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases](https://arxiv.org/abs/2103.10697)


## Introduction

ConViT combines the strengths of convolutional architectures and Vision Transformers (ViTs).
Expand All @@ -19,36 +20,12 @@ while offering a much improved sample efficiency.[[1](#references)]
<em>Figure 1. Architecture of ConViT [<a href="#references">1</a>] </em>
</p>


## Results

Our reproduced model performance on ImageNet-1K is reported as follows.

- ascend 910* with graph mode


<div align="center">


| model | top-1 (%) | top-5 (%) | params (M) | batch size | cards | ms/step | jit_level | recipe | download |
| ----------- | --------- | --------- | ---------- | ---------- | ----- | ------- | --------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------- |
| convit_tiny | 73.79 | 91.70 | 5.71 | 256 | 8 | 226.51 | O2 | [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>

- ascend 910 with graph mode

<div align="center">
## Requirements
| mindspore | ascend driver | firmware | cann toolkit/kernel |
| :-------: | :-----------: | :---------: | :-----------------: |
| 2.3.1 | 24.1.RC2 | 7.3.0.1.231 | 8.0.RC2.beta1 |


| model | top-1 (%) | top-5 (%) | params (M) | batch size | cards | ms/step | jit_level | recipe | download |
| ----------- | --------- | --------- | ---------- | ---------- | ----- | ------- | --------- | ------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------- |
| convit_tiny | 73.66 | 91.72 | 5.71 | 256 | 8 | 231.62 | O2 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_tiny-e31023f2.ckpt) |

</div>

#### Notes
- Top-1 and Top-5: Accuracy reported on the validation set of ImageNet-1K.

## Quick Start

Expand Down Expand Up @@ -93,6 +70,26 @@ To validate the accuracy of the trained model, you can use `validate.py` and par
python validate.py -c configs/convit/convit_tiny_ascend.yaml --data_dir /path/to/imagenet --ckpt_path /path/to/ckpt
```

## Performance

Our reproduced model performance on ImageNet-1K is reported as follows.

Experiments are tested on ascend 910* with mindspore 2.3.1 graph mode.

| model name | params(M) | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | acc@top1 | acc@top5 | recipe | weight |
| ----------- | --------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | -------- | -------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------- |
| convit_tiny | 5.71 | 8 | 256 | 224x224 | O2 | 153s | 226.51 | 9022.03 | 73.79 | 91.70 | [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) |

Experiments are tested on ascend 910 with mindspore 2.3.1 graph mode.

| model name | params(M) | cards | batch size | resolution | jit level | graph compile | ms/step | img/s | acc@top1 | acc@top5 | recipe | weight |
| ----------- | --------- | ----- | ---------- | ---------- | --------- | ------------- | ------- | ------- | -------- | -------- | ------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------- |
| convit_tiny | 5.71 | 8 | 256 | 224x224 | O2 | 133s | 231.62 | 8827.59 | 73.66 | 91.72 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_tiny-e31023f2.ckpt) |


### Notes
- top-1 and top-5: Accuracy reported on the validation set of ImageNet-1K.

## References

<!--- Guideline: Citation format should follow GB/T 7714. -->
Expand Down
Loading
Loading