cooper_robot commited on
Commit ·
a7d7363
1
Parent(s): 786d22d
Add release note for v1.2.0
Browse files- .gitattributes +0 -3
- .gitignore +0 -6
- LICENSE +1 -1
- README.md +27 -26
- cv75_convmixer_768_32.bin +2 -2
- cv7_convmixer_768_32.bin +3 -0
.gitattributes
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 2 |
-
*.png filter=lfs diff=lfs merge=lfs -text
|
| 3 |
-
*.tar filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
.gitignore
DELETED
|
@@ -1,6 +0,0 @@
|
|
| 1 |
-
n1_*.bin
|
| 2 |
-
|
| 3 |
-
# 确保所有 Git 相关文件不会被提交
|
| 4 |
-
.git*
|
| 5 |
-
!.gitattributes
|
| 6 |
-
!.gitignore
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
LICENSE
CHANGED
|
@@ -186,7 +186,7 @@
|
|
| 186 |
same "printed page" as the copyright notice for easier
|
| 187 |
identification within third-party archives.
|
| 188 |
|
| 189 |
-
Copyright
|
| 190 |
|
| 191 |
Licensed under the Apache License, Version 2.0 (the "License");
|
| 192 |
you may not use this file except in compliance with the License.
|
|
|
|
| 186 |
same "printed page" as the copyright notice for easier
|
| 187 |
identification within third-party archives.
|
| 188 |
|
| 189 |
+
Copyright 2019 Ross Wightman
|
| 190 |
|
| 191 |
Licensed under the Apache License, Version 2.0 (the "License");
|
| 192 |
you may not use this file except in compliance with the License.
|
README.md
CHANGED
|
@@ -1,26 +1,27 @@
|
|
| 1 |
-
---
|
| 2 |
-
library_name: pytorch
|
| 3 |
-
---
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
ConvMixer is a simple yet effective vision architecture that combines large-kernel depthwise convolutions for spatial mixing with pointwise convolutions for channel mixing, achieving transformer-like performance with minimal complexity.
|
| 7 |
-
|
| 8 |
-
Original paper: [Patches Are All You Need? ConvMixer](https://arxiv.org/abs/2201.09792)
|
| 9 |
-
|
| 10 |
-
# ConvMixer-768/32
|
| 11 |
-
|
| 12 |
-
This model uses the ConvMixer-768/32 variant, which processes 32×32 patches with 768 feature channels, providing strong accuracy while remaining computationally efficient. It is well suited for image classification tasks where simplicity, speed, and high accuracy are desired, and can serve as a lightweight backbone for research or prototyping.
|
| 13 |
-
|
| 14 |
-
Model Configuration:
|
| 15 |
-
- Reference implementation: [Official ConvMixer source code](https://github.com/locuslab/convmixer)
|
| 16 |
-
- Original Weight: [Convmixer_768_32_ks7_p7_relu](https://github.com/locuslab/convmixer/releases/download/v1.0/convmixer_768_32_ks7_p7_relu.pth.tar)
|
| 17 |
-
- Resolution: 3x224x224
|
| 18 |
-
- Support Cooper version:
|
| 19 |
-
- Cooper SDK: [2.5.
|
| 20 |
-
- Cooper Foundry: [2.2]
|
| 21 |
-
|
| 22 |
-
| Model | Device | Model Link |
|
| 23 |
-
| :-----: | :-----: | :-----: |
|
| 24 |
-
| ConvMixer-768/32 | N1-655 | [Model_Link](https://huggingface.co/Ambarella/ConvMixer/blob/main/n1-655_convmixer_768_32.bin) |
|
| 25 |
-
| ConvMixer-768/32 |
|
| 26 |
-
| ConvMixer-768/32 |
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: pytorch
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
ConvMixer is a simple yet effective vision architecture that combines large-kernel depthwise convolutions for spatial mixing with pointwise convolutions for channel mixing, achieving transformer-like performance with minimal complexity.
|
| 7 |
+
|
| 8 |
+
Original paper: [Patches Are All You Need? ConvMixer](https://arxiv.org/abs/2201.09792)
|
| 9 |
+
|
| 10 |
+
# ConvMixer-768/32
|
| 11 |
+
|
| 12 |
+
This model uses the ConvMixer-768/32 variant, which processes 32×32 patches with 768 feature channels, providing strong accuracy while remaining computationally efficient. It is well suited for image classification tasks where simplicity, speed, and high accuracy are desired, and can serve as a lightweight backbone for research or prototyping.
|
| 13 |
+
|
| 14 |
+
Model Configuration:
|
| 15 |
+
- Reference implementation: [Official ConvMixer source code](https://github.com/locuslab/convmixer)
|
| 16 |
+
- Original Weight: [Convmixer_768_32_ks7_p7_relu](https://github.com/locuslab/convmixer/releases/download/v1.0/convmixer_768_32_ks7_p7_relu.pth.tar)
|
| 17 |
+
- Resolution: 3x224x224
|
| 18 |
+
- Support Cooper version:
|
| 19 |
+
- Cooper SDK: [2.5.3]
|
| 20 |
+
- Cooper Foundry: [2.2]
|
| 21 |
+
|
| 22 |
+
| Model | Device | Model Link |
|
| 23 |
+
| :-----: | :-----: | :-----: |
|
| 24 |
+
| ConvMixer-768/32 | N1-655 | [Model_Link](https://huggingface.co/Ambarella/ConvMixer/blob/main/n1-655_convmixer_768_32.bin) |
|
| 25 |
+
| ConvMixer-768/32 | CV7 | [Model_Link](https://huggingface.co/Ambarella/ConvMixer/blob/main/cv7_convmixer_768_32.bin) |
|
| 26 |
+
| ConvMixer-768/32 | CV72 | [Model_Link](https://huggingface.co/Ambarella/ConvMixer/blob/main/cv72_convmixer_768_32.bin) |
|
| 27 |
+
| ConvMixer-768/32 | CV75 | [Model_Link](https://huggingface.co/Ambarella/ConvMixer/blob/main/cv75_convmixer_768_32.bin) |
|
cv75_convmixer_768_32.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38e4bfc5443949777bfd9480c627d5b12e66b619b181b6333ea524fea22f352a
|
| 3 |
+
size 17670492
|
cv7_convmixer_768_32.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55eee0d5862e0b903ef751d820e8c22dab5d5b222fe0cfb5bbb30998a75b162f
|
| 3 |
+
size 17196712
|