[Enhance] Add ConvNeXt pretrain models on ImageNet-1k. (#707)
* [Docs] Add ConvNeXt pretrain models on ImageNet-1k * Update README * Update metafile Co-authored-by: mzr1996 <mzr1996@163.com>pull/713/head
yingfhu
GitHub
parent
9ee41577d0
commit
8374c05b0a
|
|
@ -29,15 +29,18 @@ The "Roaring 20s" of visual recognition began with the introduction of Vision Tr
|
|||
|
||||
*Models with \* are converted from the [official repo](https://github.com/facebookresearch/ConvNeXt). The config files of these models are only for inference. We don't ensure these config files' training accuracy and welcome you to contribute your reproduction results.*
|
||||
|
||||
### ImageNet-21k
|
||||
### Pre-trained Models
|
||||
|
||||
The pre-trained models on ImageNet-21k are used to fine-tune, and therefore don't have evaluation results.
|
||||
The pre-trained models on ImageNet-1k or ImageNet-21k are used to fine-tune on the downstream tasks.
|
||||
|
||||
| Model | Params(M) | Flops(G) | Download |
|
||||
|:--------------------------------:|:---------:|:--------:|:--------:|
|
||||
| convnext-base_3rdparty_in21k\* | 88.59 | 15.36 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-base_3rdparty_in21k_20220124-13b83eec.pth) |
|
||||
| convnext-large_3rdparty_in21k\* | 197.77 | 34.37 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-large_3rdparty_in21k_20220124-41b5a79f.pth) |
|
||||
| convnext-xlarge_3rdparty_in21k\* | 350.20 | 60.93 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-xlarge_3rdparty_in21k_20220124-f909bad7.pth) |
|
||||
| Model | Training Data | Params(M) | Flops(G) | Download |
|
||||
|:--------------:|:-------------:|:---------:|:--------:|:--------:|
|
||||
| ConvNeXt-T\* | ImageNet-1k | 28.59 | 4.46 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-tiny_3rdparty_32xb128-noema_in1k_20220222-2908964a.pth) |
|
||||
| ConvNeXt-S\* | ImageNet-1k | 50.22 | 8.69 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_3rdparty_32xb128-noema_in1k_20220222-fa001ca5.pth) |
|
||||
| ConvNeXt-B\* | ImageNet-1k | 88.59 | 15.36 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-base_3rdparty_32xb128-noema_in1k_20220222-dba4f95f.pth) |
|
||||
| ConvNeXt-B\* | ImageNet-21k | 88.59 | 15.36 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-base_3rdparty_in21k_20220124-13b83eec.pth) |
|
||||
| ConvNeXt-L\* | ImageNet-21k | 197.77 | 34.37 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-large_3rdparty_in21k_20220124-41b5a79f.pth) |
|
||||
| ConvNeXt-XL\* | ImageNet-21k | 350.20 | 60.93 | [model](https://download.openmmlab.com/mmclassification/v0/convnext/convnext-xlarge_3rdparty_in21k_20220124-f909bad7.pth) |
|
||||
|
||||
*Models with \* are converted from the [official repo](https://github.com/facebookresearch/ConvNeXt).*
|
||||
|
||||
|
|
|
|||
|
|
@ -30,6 +30,23 @@ Models:
|
|||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_tiny_1k_224_ema.pth
|
||||
Code: https://github.com/facebookresearch/ConvNeXt
|
||||
- Name: convnext-tiny_3rdparty_32xb128-noema_in1k
|
||||
Metadata:
|
||||
Training Data: ImageNet-1k
|
||||
FLOPs: 4457472768
|
||||
Parameters: 28589128
|
||||
In Collections: ConvNeXt
|
||||
Results:
|
||||
- Dataset: ImageNet-1k
|
||||
Metrics:
|
||||
Top 1 Accuracy: 81.81
|
||||
Top 5 Accuracy: 95.67
|
||||
Task: Image Classification
|
||||
Weights: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-tiny_3rdparty_32xb128-noema_in1k_20220222-2908964a.pth
|
||||
Config: configs/convnext/convnext-tiny_32xb128_in1k.py
|
||||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_tiny_1k_224.pth
|
||||
Code: https://github.com/facebookresearch/ConvNeXt
|
||||
- Name: convnext-small_3rdparty_32xb128_in1k
|
||||
Metadata:
|
||||
FLOPs: 8687008512
|
||||
|
|
@ -46,6 +63,23 @@ Models:
|
|||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_small_1k_224_ema.pth
|
||||
Code: https://github.com/facebookresearch/ConvNeXt
|
||||
- Name: convnext-small_3rdparty_32xb128-noema_in1k
|
||||
Metadata:
|
||||
Training Data: ImageNet-1k
|
||||
FLOPs: 8687008512
|
||||
Parameters: 50223688
|
||||
In Collections: ConvNeXt
|
||||
Results:
|
||||
- Dataset: ImageNet-1k
|
||||
Metrics:
|
||||
Top 1 Accuracy: 83.11
|
||||
Top 5 Accuracy: 96.34
|
||||
Task: Image Classification
|
||||
Weights: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-small_3rdparty_32xb128-noema_in1k_20220222-fa001ca5.pth
|
||||
Config: configs/convnext/convnext-small_32xb128_in1k.py
|
||||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_small_1k_224.pth
|
||||
Code: https://github.com/facebookresearch/ConvNeXt
|
||||
- Name: convnext-base_3rdparty_32xb128_in1k
|
||||
Metadata:
|
||||
FLOPs: 15359124480
|
||||
|
|
@ -62,12 +96,30 @@ Models:
|
|||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_base_1k_224_ema.pth
|
||||
Code: https://github.com/facebookresearch/ConvNeXt
|
||||
- Name: convnext-base_3rdparty_32xb128-noema_in1k
|
||||
Metadata:
|
||||
Training Data: ImageNet-1k
|
||||
FLOPs: 15359124480
|
||||
Parameters: 88591464
|
||||
In Collections: ConvNeXt
|
||||
Results:
|
||||
- Dataset: ImageNet-1k
|
||||
Metrics:
|
||||
Top 1 Accuracy: 83.71
|
||||
Top 5 Accuracy: 96.60
|
||||
Task: Image Classification
|
||||
Weights: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-base_3rdparty_32xb128-noema_in1k_20220222-dba4f95f.pth
|
||||
Config: configs/convnext/convnext-base_32xb128_in1k.py
|
||||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_base_1k_224.pth
|
||||
Code: https://github.com/facebookresearch/ConvNeXt
|
||||
- Name: convnext-base_3rdparty_in21k
|
||||
Metadata:
|
||||
Training Data: ImageNet-21k
|
||||
FLOPs: 15359124480
|
||||
Parameters: 88591464
|
||||
In Collections: ConvNeXt
|
||||
Results: null
|
||||
Weights: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-base_3rdparty_in21k_20220124-13b83eec.pth
|
||||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_base_22k_224.pth
|
||||
|
|
@ -113,6 +165,7 @@ Models:
|
|||
FLOPs: 34368026112
|
||||
Parameters: 197767336
|
||||
In Collections: ConvNeXt
|
||||
Results: null
|
||||
Weights: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-large_3rdparty_in21k_20220124-41b5a79f.pth
|
||||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_224.pth
|
||||
|
|
@ -142,6 +195,7 @@ Models:
|
|||
FLOPs: 60929820672
|
||||
Parameters: 350196968
|
||||
In Collections: ConvNeXt
|
||||
Results: null
|
||||
Weights: https://download.openmmlab.com/mmclassification/v0/convnext/convnext-xlarge_3rdparty_in21k_20220124-f909bad7.pth
|
||||
Converted From:
|
||||
Weights: https://dl.fbaipublicfiles.com/convnext/convnext_xlarge_22k_224.pth
|
||||
|
|
|
|||
Loading…
Reference in New Issue