first commit

configs/efficientnet_v2/efficientnetv2-m_8xb32_in1k-480px.py

+_base_ = [
+    'efficientnetv2-s_8xb32_in1k-384px.py',
+]
+
+# model setting
+model = dict(backbone=dict(arch='m'), )
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetRandomCrop', scale=384, crop_padding=0),
+    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
+    dict(type='PackInputs'),
+]
+
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetCenterCrop', crop_size=480, crop_padding=0),
+    dict(type='PackInputs'),
+]
+
+train_dataloader = dict(dataset=dict(pipeline=train_pipeline))
+val_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+test_dataloader = dict(dataset=dict(pipeline=test_pipeline))

configs/efficientnet_v2/efficientnetv2-m_8xb32_in21k.py

+_base_ = ['./efficientnetv2-s_8xb32_in21k.py']
+
+# model setting
+model = dict(backbone=dict(arch='m'), )

configs/efficientnet_v2/efficientnetv2-s_8xb32_in1k-384px.py

+_base_ = [
+    '../_base_/models/efficientnet_v2/efficientnetv2_s.py',
+    '../_base_/datasets/imagenet_bs32.py',
+    '../_base_/schedules/imagenet_bs256.py',
+    '../_base_/default_runtime.py',
+]
+
+# dataset settings
+dataset_type = 'ImageNet'
+data_preprocessor = dict(
+    num_classes=1000,
+    # RGB format normalization parameters
+    mean=[127.5, 127.5, 127.5],
+    std=[127.5, 127.5, 127.5],
+    # convert image from BGR to RGB
+    to_rgb=True,
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetRandomCrop', scale=300, crop_padding=0),
+    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
+    dict(type='PackInputs'),
+]
+
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetCenterCrop', crop_size=384, crop_padding=0),
+    dict(type='PackInputs'),
+]
+
+train_dataloader = dict(dataset=dict(pipeline=train_pipeline))
+val_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+test_dataloader = dict(dataset=dict(pipeline=test_pipeline))

configs/efficientnet_v2/efficientnetv2-s_8xb32_in21k.py

+_base_ = [
+    '../_base_/models/efficientnet_v2/efficientnetv2_s.py',
+    '../_base_/datasets/imagenet_bs32.py',
+    '../_base_/schedules/imagenet_bs256.py',
+    '../_base_/default_runtime.py',
+]
+
+# model setting
+model = dict(head=dict(num_classes=21843))
+
+# dataset settings
+dataset_type = 'ImageNet21k'
+data_preprocessor = dict(
+    num_classes=21843,
+    # RGB format normalization parameters
+    mean=[127.5, 127.5, 127.5],
+    std=[127.5, 127.5, 127.5],
+    # convert image from BGR to RGB
+    to_rgb=True,
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetRandomCrop', scale=224),
+    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
+    dict(type='PackInputs'),
+]
+
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetCenterCrop', crop_size=224, crop_padding=0),
+    dict(type='PackInputs'),
+]
+
+train_dataloader = dict(dataset=dict(pipeline=train_pipeline))
+val_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+test_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+
+# schedule setting
+optim_wrapper = dict(
+    optimizer=dict(lr=4e-3),
+    clip_grad=dict(max_norm=5.0),
+)

configs/efficientnet_v2/efficientnetv2-xl_8xb32_in1k-512px.py

+_base_ = [
+    'efficientnetv2-s_8xb32_in1k-384px.py',
+]
+
+# model setting
+model = dict(backbone=dict(arch='xl'), )
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetRandomCrop', scale=384, crop_padding=0),
+    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
+    dict(type='PackInputs'),
+]
+
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='EfficientNetCenterCrop', crop_size=512, crop_padding=0),
+    dict(type='PackInputs'),
+]
+
+train_dataloader = dict(dataset=dict(pipeline=train_pipeline))
+val_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+test_dataloader = dict(dataset=dict(pipeline=test_pipeline))

configs/efficientnet_v2/efficientnetv2-xl_8xb32_in21k.py

+_base_ = ['./efficientnetv2-s_8xb32_in21k.py']
+
+# model setting
+model = dict(backbone=dict(arch='xl'), )

configs/efficientnet_v2/metafile.yml

+Collections:
+  - Name: EfficientNetV2
+    Metadata:
+      Training Data: ImageNet-1k
+      Architecture:
+        - 1x1 Convolution
+        - Average Pooling
+        - Convolution
+        - Dense Connections
+        - Dropout
+        - Inverted Residual Block
+        - RMSProp
+        - Squeeze-and-Excitation Block
+        - Swish
+    Paper:
+      URL: https://arxiv.org/abs/2104.00298
+      Title: "EfficientNetV2: Smaller Models and Faster Training"
+    README: configs/efficientnet_v2/README.md
+    Code:
+      URL: https://github.com/open-mmlab/mmpretrain/blob/main/mmpretrain/models/backbones/beit.py
+      Version: v1.0.0rc4
+
+Models:
+  - Name: efficientnetv2-b0_3rdparty_in1k
+    Metadata:
+      FLOPs: 919843360
+      Parameters: 7139704
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 78.52
+          Top 5 Accuracy: 94.44
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-b0_3rdparty_in1k_20221221-9ef6e736.pth
+    Config: configs/efficientnet_v2/efficientnetv2-b0_8xb32_in1k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b0-c7cc451f.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-b1_3rdparty_in1k
+    Metadata:
+      FLOPs: 1438287552
+      Parameters: 8141052
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 79.80
+          Top 5 Accuracy: 94.89
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-b1_3rdparty_in1k_20221221-6955d9ce.pth
+    Config: configs/efficientnet_v2/efficientnetv2-b1_8xb32_in1k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b1-be6e41b0.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-b2_3rdparty_in1k
+    Metadata:
+      FLOPs: 1986433080
+      Parameters: 10096086
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 80.63
+          Top 5 Accuracy: 95.30
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-b2_3rdparty_in1k_20221221-74f7d493.pth
+    Config: configs/efficientnet_v2/efficientnetv2-b2_8xb32_in1k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b2-847de54e.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-b3_3rdparty_in1k
+    Metadata:
+      FLOPs: 3498068400
+      Parameters: 14358406
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 82.03
+          Top 5 Accuracy: 95.88
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-b3_3rdparty_in1k_20221221-b6f07a36.pth
+    Config: configs/efficientnet_v2/efficientnetv2-b3_8xb32_in1k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_b3-57773f13.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-s_3rdparty_in1k
+    Metadata:
+      FLOPs: 9719420928
+      Parameters: 21458488
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 83.82
+          Top 5 Accuracy: 96.67
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-s_3rdparty_in1k_20221220-f0eaff9d.pth
+    Config: configs/efficientnet_v2/efficientnetv2-s_8xb32_in1k-384px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s-eb54923e.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-m_3rdparty_in1k
+    Metadata:
+      FLOPs: 26880363584
+      Parameters: 54139356
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 85.01
+          Top 5 Accuracy: 97.26
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-m_3rdparty_in1k_20221220-9dc0c729.pth
+    Config: configs/efficientnet_v2/efficientnetv2-m_8xb32_in1k-480px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_m-cc09e0cd.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-l_3rdparty_in1k
+    Metadata:
+      FLOPs: 60142387008
+      Parameters: 118515272
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 85.43
+          Top 5 Accuracy: 97.31
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-l_3rdparty_in1k_20221220-5c3bac0f.pth
+    Config: configs/efficientnet_v2/efficientnetv2-l_8xb32_in1k-480px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_l-d664b728.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-s_in21k-pre_3rdparty_in1k
+    Metadata:
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+      FLOPs: 9719420928
+      Parameters: 21458488
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 84.29
+          Top 5 Accuracy: 97.26
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-s_in21k-pre-3rdparty_in1k_20221220-7a7c8475.pth
+    Config: configs/efficientnet_v2/efficientnetv2-s_8xb32_in1k-384px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21ft1k-d7dafa41.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-m_in21k-pre_3rdparty_in1k
+    Metadata:
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+      FLOPs: 26880363584
+      Parameters: 54139356
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 85.47
+          Top 5 Accuracy: 97.76
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-m_in21k-pre-3rdparty_in1k_20221220-a1013a04.pth
+    Config: configs/efficientnet_v2/efficientnetv2-m_8xb32_in1k-480px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_m_21ft1k-bf41664a.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-l_in21k-pre_3rdparty_in1k
+    Metadata:
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+      FLOPs: 60142387008
+      Parameters: 118515272
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 86.31
+          Top 5 Accuracy: 97.99
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-l_in21k-pre-3rdparty_in1k_20221220-63df0efd.pth
+    Config: configs/efficientnet_v2/efficientnetv2-l_8xb32_in1k-480px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_l_21ft1k-60127a9d.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-xl_in21k-pre_3rdparty_in1k
+    Metadata:
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+      FLOPs: 98341230592
+      Parameters: 208119808
+    In Collection: EfficientNetV2
+    Results:
+      - Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 86.39
+          Top 5 Accuracy: 97.83
+        Task: Image Classification
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-xl_in21k-pre-3rdparty_in1k_20221220-583ac18b.pth
+    Config: configs/efficientnet_v2/efficientnetv2-xl_8xb32_in1k-512px.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_xl_in21ft1k-06c35c48.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-s_3rdparty_in21k
+    Metadata:
+      FLOPs: 3309720768
+      Parameters: 48158371
+    In Collection: EfficientNetV2
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-s_3rdparty_in21k_20221220-c0572b56.pth
+    Config: configs/efficientnet_v2/efficientnetv2-s_8xb32_in21k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21k-6337ad01.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-m_3rdparty_in21k
+    Metadata:
+      FLOPs: 5861638208
+      Parameters: 80839239
+    In Collection: EfficientNetV2
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-m_3rdparty_in21k_20221220-073e944c.pth
+    Config: configs/efficientnet_v2/efficientnetv2-m_8xb32_in21k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_m_21k-361418a2.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-l_3rdparty_in21k
+    Metadata:
+      FLOPs: 13114950464
+      Parameters: 145215155
+    In Collection: EfficientNetV2
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-l_3rdparty_in21k_20221220-f28f91e1.pth
+    Config: configs/efficientnet_v2/efficientnetv2-l_8xb32_in21k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_l_21k-91a19ec9.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py
+  - Name: efficientnetv2-xl_3rdparty_in21k
+    Metadata:
+      FLOPs: 18855244288
+      Parameters: 234819691
+    In Collection: EfficientNetV2
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/efficientnetv2/efficientnetv2-xl_3rdparty_in21k_20221220-b2c9329c.pth
+    Config: configs/efficientnet_v2/efficientnetv2-xl_8xb32_in21k.py
+    Converted From:
+      Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_xl_in21k-fd7e8abf.pth
+      Code: https://github.com/rwightman/pytorch-image-models/blob/main/timm/models/efficientnet.py

configs/eva/README.md

+# EVA
+
+> [EVA: Exploring the Limits of Masked Visual Representation Learning at Scale](https://arxiv.org/abs/2211.07636)
+
+<!-- [ALGORITHM] -->
+
+## Abstract
+
+We launch EVA, a vision-centric foundation model to explore the limits of visual representation at scale using only publicly accessible data. EVA is a vanilla ViT pre-trained to reconstruct the masked out image-text aligned vision features conditioned on visible image patches. Via this pretext task, we can efficiently scale up EVA to one billion parameters, and sets new records on a broad range of representative vision downstream tasks, such as image recognition, video action recognition, object detection, instance segmentation and semantic segmentation without heavy supervised training. Moreover, we observe quantitative changes in scaling EVA result in qualitative changes in transfer learning performance that are not present in other models. For instance, EVA takes a great leap in the challenging large vocabulary instance segmentation task: our model achieves almost the same state-of-the-art performance on LVISv1.0 dataset with over a thousand categories and COCO dataset with only eighty categories. Beyond a pure vision encoder, EVA can also serve as a vision-centric, multi-modal pivot to connect images and text. We find initializing the vision tower of a giant CLIP from EVA can greatly stabilize the training and outperform the training from scratch counterpart with much fewer samples and less compute, providing a new direction for scaling up and accelerating the costly training of multi-modal foundation models.
+
+<div align=center>
+<img src="https://user-images.githubusercontent.com/24734142/205410193-f1164e56-c117-4165-86f5-4cbfd797bc87.png" width="70%"/>
+</div>
+
+## How to use it?
+
+<!-- [TABS-BEGIN] -->
+
+**Predict image**
+
+```python
+from mmpretrain import inference_model
+
+predict = inference_model('vit-base-p16_eva-mae-style-pre_8xb128-coslr-100e_in1k', 'demo/bird.JPEG')
+print(predict['pred_class'])
+print(predict['pred_score'])
+```
+
+**Use the model**
+
+```python
+import torch
+from mmpretrain import get_model
+
+model = get_model('eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k', pretrained=True)
+inputs = torch.rand(1, 3, 224, 224)
+out = model(inputs)
+print(type(out))
+# To extract features.
+feats = model.extract_feat(inputs)
+print(type(feats))
+```
+
+**Train/Test Command**
+
+Prepare your dataset according to the [docs](https://mmpretrain.readthedocs.io/en/latest/user_guides/dataset_prepare.html#prepare-dataset).
+
+Train:
+
+```shell
+python tools/train.py configs/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k.py
+```
+
+Test:
+
+```shell
+python tools/test.py configs/eva/benchmarks/vit-base-p16_8xb128-coslr-100e_in1k.py https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k_20221226-f61cf992.pth
+```
+
+<!-- [TABS-END] -->
+
+## Models and results
+
+### Pretrained models
+
+| Model                                                | Params (M) | Flops (G) |                             Config                              |                              Download                              |
+| :--------------------------------------------------- | :--------: | :-------: | :-------------------------------------------------------------: | :----------------------------------------------------------------: |
+| `eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k` |   111.78   |   17.58   | [config](eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k.py) | [model](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k_20221226-26d90f07.pth) \| [log](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k_20221226-26d90f07.json) |
+| `beit-l-p14_3rdparty-eva_in21k`\*                    |   303.18   |   81.08   |                 [config](eva-l-p14_headless.py)                 | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_3rdparty-mim_in21k_20221213-3a5da50b.pth) |
+| `beit-l-p14_eva-pre_3rdparty_in21k`\*                |   303.18   |   81.08   |                 [config](eva-l-p14_headless.py)                 | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-pre_3rdparty_in21k_20221213-8f194fa2.pth) |
+| `beit-g-p16_3rdparty-eva_30m`\*                      |  1011.32   |  203.52   |                 [config](eva-g-p16_headless.py)                 | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p16_3rdparty_30m_20221213-7bed23ee.pth) |
+| `beit-g-p14_3rdparty-eva_30m`\*                      |  1011.60   |  267.17   |                 [config](eva-g-p14_headless.py)                 | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_3rdparty_30m_20221213-3b7aca97.pth) |
+| `beit-g-p14_eva-30m-pre_3rdparty_in21k`\*            |  1011.60   |  267.17   |                 [config](eva-g-p14_headless.py)                 | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-pre_3rdparty_in21k_20221213-d72285b7.pth) |
+
+*Models with * are converted from the [official repo](https://github.com/baaivision/EVA). The config files of these models are only for inference. We haven't reproduce the training results.*
+
+### Image Classification on ImageNet-1k
+
+| Model                                   |                  Pretrain                  | Params (M) | Flops (G) | Top-1 (%) | Top-5 (%) |                  Config                  |                  Download                  |
+| :-------------------------------------- | :----------------------------------------: | :--------: | :-------: | :-------: | :-------: | :--------------------------------------: | :----------------------------------------: |
+| `vit-base-p16_eva-mae-style-pre_8xb128-coslr-100e_in1k` | [EVA MAE STYLE](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k_20221226-26d90f07.pth) |   86.57    |   17.58   |   83.70   |    N/A    | [config](benchmarks/vit-base-p16_8xb128-coslr-100e_in1k.py) | [model](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k_20221226-f61cf992.pth) \| [log](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k_20221226-f61cf992.json) |
+| `vit-base-p16_eva-mae-style-pre_8xb2048-linear-coslr-100e_in1k` | [EVA MAE STYLE](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k_20221226-26d90f07.pth) |   86.57    |   17.58   |   69.00   |    N/A    | [config](benchmarks/vit-base-p16_8xb2048-linear-coslr-100e_in1k.py) | [model](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_linear-8xb2048-coslr-100e_in1k/vit-base-p16_linear-8xb2048-coslr-100e_in1k_20221226-ef51bf09.pth) \| [log](https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_linear-8xb2048-coslr-100e_in1k/vit-base-p16_linear-8xb2048-coslr-100e_in1k_20221226-ef51bf09.json) |
+| `beit-l-p14_eva-pre_3rdparty_in1k-196px`\* | [EVA](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_3rdparty-mim_in21k_20221213-3a5da50b.pth) |   304.14   |   61.57   |   87.94   |   98.5    | [config](eva-l-p14_8xb16_in1k-196px.py)  | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-pre_3rdparty_in1k-196px_20221214-2adf4d28.pth) |
+| `beit-l-p14_eva-in21k-pre_3rdparty_in1k-196px`\* |              EVA ImageNet-21k              |   304.14   |   61.57   |   88.58   |   98.65   | [config](eva-l-p14_8xb16_in1k-196px.py)  | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-in21k-pre_3rdparty_in1k-196px_20221213-b730c7e7.pth) |
+| `beit-l-p14_eva-pre_3rdparty_in1k-336px`\* | [EVA](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_3rdparty-mim_in21k_20221213-3a5da50b.pth) |   304.53   |  191.10   |   88.66   |   98.75   | [config](eva-l-p14_8xb16_in1k-336px.py)  | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-pre_3rdparty_in1k-336px_20221214-07785cfd.pth) |
+| `beit-l-p14_eva-in21k-pre_3rdparty_in1k-336px`\* |              EVA ImageNet-21k              |   304.53   |  191.10   |   89.17   |   98.86   | [config](eva-l-p14_8xb16_in1k-336px.py)  | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-in21k-pre_3rdparty_in1k-336px_20221213-f25b7634.pth) |
+| `beit-g-p14_eva-30m-in21k-pre_3rdparty_in1k-336px`\* | [EVA 30M ImageNet-21k](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-pre_3rdparty_in21k_20221213-d72285b7.pth) |  1013.01   |  620.64   |   89.61   |   98.93   | [config](eva-g-p14_8xb16_in1k-336px.py)  | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-in21k-pre_3rdparty_in1k-336px_20221213-210f9071.pth) |
+| `beit-g-p14_eva-30m-in21k-pre_3rdparty_in1k-560px`\* | [EVA 30M ImageNet-21k](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-pre_3rdparty_in21k_20221213-d72285b7.pth) |  1014.45   |  1906.76  |   89.71   |   98.96   | [config](eva-g-p14_8xb16_in1k-560px.py)  | [model](https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-in21k-pre_3rdparty_in1k-560px_20221213-fa1c3652.pth) |
+
+*Models with * are converted from the [official repo](https://github.com/baaivision/EVA). The config files of these models are only for inference. We haven't reproduce the training results.*
+
+## Citation
+
+```bibtex
+@article{EVA,
+  title={EVA: Exploring the Limits of Masked Visual Representation Learning at Scale},
+  author={Fang, Yuxin and Wang, Wen and Xie, Binhui and Sun, Quan and Wu, Ledell and Wang, Xinggang and Huang, Tiejun and Wang, Xinlong and Cao, Yue},
+  journal={arXiv preprint arXiv:2211.07636},
+  year={2022}
+}
+```

configs/eva/benchmarks/vit-base-p16_8xb128-coslr-100e_in1k.py

+_base_ = [
+    '../../_base_/datasets/imagenet_bs64_swin_224.py',
+    '../../_base_/schedules/imagenet_bs1024_adamw_swin.py',
+    '../../_base_/default_runtime.py'
+]
+
+# dataset settings
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='RandomResizedCrop',
+        scale=224,
+        backend='pillow',
+        interpolation='bicubic'),
+    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
+    dict(
+        type='RandAugment',
+        policies='timm_increasing',
+        num_policies=2,
+        total_level=10,
+        magnitude_level=9,
+        magnitude_std=0.5,
+        hparams=dict(pad_val=[104, 116, 124], interpolation='bicubic')),
+    dict(
+        type='RandomErasing',
+        erase_prob=0.25,
+        mode='rand',
+        min_area_ratio=0.02,
+        max_area_ratio=0.3333333333333333,
+        fill_color=[103.53, 116.28, 123.675],
+        fill_std=[57.375, 57.12, 58.395]),
+    dict(type='PackInputs')
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='ResizeEdge',
+        scale=256,
+        edge='short',
+        backend='pillow',
+        interpolation='bicubic'),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='PackInputs')
+]
+
+train_dataloader = dict(batch_size=128, dataset=dict(pipeline=train_pipeline))
+val_dataloader = dict(batch_size=128, dataset=dict(pipeline=test_pipeline))
+test_dataloader = val_dataloader
+
+# model settings
+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        type='VisionTransformer',
+        arch='base',
+        img_size=224,
+        patch_size=16,
+        drop_path_rate=0.1,
+        out_type='avg_featmap',
+        final_norm=False,
+        init_cfg=dict(type='Pretrained', checkpoint='', prefix='backbone.')),
+    neck=None,
+    head=dict(
+        type='LinearClsHead',
+        num_classes=1000,
+        in_channels=768,
+        loss=dict(
+            type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'),
+        init_cfg=[dict(type='TruncNormal', layer='Linear', std=0.02)]),
+    train_cfg=dict(augments=[
+        dict(type='Mixup', alpha=0.8),
+        dict(type='CutMix', alpha=1.0)
+    ]))
+
+# optimizer wrapper
+optim_wrapper = dict(
+    optimizer=dict(
+        type='AdamW', lr=4e-4, weight_decay=0.05, betas=(0.9, 0.999)),
+    constructor='LearningRateDecayOptimWrapperConstructor',
+    paramwise_cfg=dict(
+        layer_decay_rate=0.65,
+        custom_keys={
+            '.ln': dict(decay_mult=0.0),
+            '.bias': dict(decay_mult=0.0),
+            '.cls_token': dict(decay_mult=0.0),
+            '.pos_embed': dict(decay_mult=0.0)
+        }))
+
+# learning rate scheduler
+param_scheduler = [
+    dict(
+        type='LinearLR',
+        start_factor=1e-4,
+        by_epoch=True,
+        begin=0,
+        end=5,
+        convert_to_iter_based=True),
+    dict(
+        type='CosineAnnealingLR',
+        T_max=95,
+        by_epoch=True,
+        begin=5,
+        end=100,
+        eta_min=1e-6,
+        convert_to_iter_based=True)
+]
+
+# runtime settings
+train_cfg = dict(by_epoch=True, max_epochs=100)
+default_hooks = dict(
+    # save checkpoint per epoch.
+    checkpoint=dict(type='CheckpointHook', interval=1, max_keep_ckpts=3))
+
+randomness = dict(seed=0, diff_rank_seed=True)

configs/eva/benchmarks/vit-base-p16_8xb2048-linear-coslr-100e_in1k.py

+_base_ = [
+    '../../_base_/datasets/imagenet_bs32_pil_resize.py',
+    '../../_base_/schedules/imagenet_bs1024_adamw_swin.py',
+    '../../_base_/default_runtime.py'
+]
+
+train_dataloader = dict(batch_size=2048, drop_last=True)
+val_dataloader = dict(drop_last=False)
+test_dataloader = dict(drop_last=False)
+
+# model settings
+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        type='VisionTransformer',
+        arch='base',
+        img_size=224,
+        patch_size=16,
+        frozen_stages=12,
+        out_type='cls_token',
+        final_norm=True,
+        init_cfg=dict(type='Pretrained', checkpoint='', prefix='backbone.')),
+    neck=dict(type='ClsBatchNormNeck', input_features=768),
+    head=dict(
+        type='VisionTransformerClsHead',
+        num_classes=1000,
+        in_channels=768,
+        loss=dict(type='CrossEntropyLoss'),
+        init_cfg=[dict(type='TruncNormal', layer='Linear', std=0.01)]),
+    data_preprocessor=dict(
+        num_classes=1000,
+        mean=[123.675, 116.28, 103.53],
+        std=[58.395, 57.12, 57.375],
+        to_rgb=True,
+    ))
+
+# optimizer
+optim_wrapper = dict(
+    _delete_=True,
+    type='AmpOptimWrapper',
+    optimizer=dict(type='LARS', lr=3.2, weight_decay=0.0, momentum=0.9),
+)
+
+# learning rate scheduler
+param_scheduler = [
+    dict(
+        type='LinearLR',
+        start_factor=1e-4,
+        by_epoch=True,
+        begin=0,
+        end=10,
+        convert_to_iter_based=True),
+    dict(
+        type='CosineAnnealingLR',
+        T_max=90,
+        by_epoch=True,
+        begin=10,
+        end=100,
+        eta_min=0.0,
+        convert_to_iter_based=True)
+]
+
+# runtime settings
+train_cfg = dict(by_epoch=True, max_epochs=100)
+
+default_hooks = dict(
+    checkpoint=dict(type='CheckpointHook', interval=1, max_keep_ckpts=3),
+    logger=dict(type='LoggerHook', interval=10))
+
+randomness = dict(seed=0, diff_rank_seed=True)

configs/eva/eva-g-p14_8xb16_in1k-336px.py

+_base_ = [
+    '../_base_/models/eva/eva-g.py',
+    '../_base_/datasets/imagenet_bs16_eva_336.py',
+    '../_base_/schedules/imagenet_bs1024_adamw_swin.py',
+    '../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(backbone=dict(img_size=336))

configs/eva/eva-g-p14_8xb16_in1k-560px.py

+_base_ = [
+    '../_base_/models/eva/eva-g.py',
+    '../_base_/datasets/imagenet_bs16_eva_560.py',
+    '../_base_/schedules/imagenet_bs1024_adamw_swin.py',
+    '../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(backbone=dict(img_size=560))

configs/eva/eva-g-p14_headless.py

+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        type='BEiTViT',
+        arch='eva-g',
+        img_size=224,
+        patch_size=14,
+        layer_scale_init_value=0.0,
+        out_type='avg_featmap',
+        use_abs_pos_emb=True,
+        use_rel_pos_bias=False,
+        use_shared_rel_pos_bias=False,
+    ),
+    neck=None,
+    head=None,
+)
+
+data_preprocessor = dict(
+    # RGB format normalization parameters
+    mean=[0.48145466 * 255, 0.4578275 * 255, 0.40821073 * 255],
+    std=[0.26862954 * 255, 0.26130258 * 255, 0.27577711 * 255],
+    # convert image from BGR to RGB
+    to_rgb=True,
+)

configs/eva/eva-g-p16_headless.py

+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        type='BEiTViT',
+        arch='eva-g',
+        img_size=224,
+        patch_size=16,
+        layer_scale_init_value=0.0,
+        out_type='avg_featmap',
+        use_abs_pos_emb=True,
+        use_rel_pos_bias=False,
+        use_shared_rel_pos_bias=False,
+    ),
+    neck=None,
+    head=None,
+)
+
+data_preprocessor = dict(
+    # RGB format normalization parameters
+    mean=[0.48145466 * 255, 0.4578275 * 255, 0.40821073 * 255],
+    std=[0.26862954 * 255, 0.26130258 * 255, 0.27577711 * 255],
+    # convert image from BGR to RGB
+    to_rgb=True,
+)

configs/eva/eva-l-p14_8xb16_in1k-196px.py

+_base_ = [
+    '../_base_/models/eva/eva-l.py',
+    '../_base_/datasets/imagenet_bs16_eva_196.py',
+    '../_base_/schedules/imagenet_bs1024_adamw_swin.py',
+    '../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(backbone=dict(img_size=196))

configs/eva/eva-l-p14_8xb16_in1k-336px.py

+_base_ = [
+    '../_base_/models/eva/eva-l.py',
+    '../_base_/datasets/imagenet_bs16_eva_336.py',
+    '../_base_/schedules/imagenet_bs1024_adamw_swin.py',
+    '../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(backbone=dict(img_size=336))

configs/eva/eva-l-p14_headless.py

+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        type='BEiTViT',
+        arch='l',
+        img_size=224,
+        patch_size=14,
+        layer_scale_init_value=0.0,
+        out_type='avg_featmap',
+        use_abs_pos_emb=True,
+        use_rel_pos_bias=False,
+        use_shared_rel_pos_bias=False,
+        layer_cfgs=dict(bias=True),
+    ),
+    neck=None,
+    head=None,
+)
+
+data_preprocessor = dict(
+    # RGB format normalization parameters
+    mean=[0.48145466 * 255, 0.4578275 * 255, 0.40821073 * 255],
+    std=[0.26862954 * 255, 0.26130258 * 255, 0.27577711 * 255],
+    # convert image from BGR to RGB
+    to_rgb=True,
+)

configs/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k.py

+_base_ = [
+    '../_base_/models/mae_vit-base-p16.py',
+    '../_base_/datasets/imagenet_bs512_mae.py',
+    '../_base_/default_runtime.py',
+]
+
+# dataset settings
+train_dataloader = dict(batch_size=256)
+
+# model settings
+model = dict(
+    type='EVA',
+    backbone=dict(init_cfg=[
+        dict(type='Xavier', distribution='uniform', layer='Linear'),
+        dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0)
+    ]),
+    neck=dict(
+        type='MAEPretrainDecoder',
+        predict_feature_dim=512,
+        init_cfg=[
+            dict(type='Xavier', distribution='uniform', layer='Linear'),
+            dict(type='Constant', layer='LayerNorm', val=1.0, bias=0.0)
+        ]),
+    head=dict(
+        _delete_=True,
+        type='MIMHead',
+        loss=dict(
+            type='CosineSimilarityLoss', shift_factor=2.0, scale_factor=2.0),
+    ),
+    target_generator=dict(
+        type='CLIPGenerator',
+        tokenizer_path=  # noqa
+        'https://download.openmmlab.com/mmselfsup/1.x/target_generator_ckpt/clip_vit_base_16.pth.tar'  # noqa
+    ),
+    init_cfg=None)
+
+# optimizer wrapper
+optim_wrapper = dict(
+    type='OptimWrapper',
+    optimizer=dict(
+        type='AdamW',
+        lr=1.5e-4 * 4096 / 256,
+        betas=(0.9, 0.95),
+        weight_decay=0.05),
+    paramwise_cfg=dict(
+        custom_keys={
+            'ln': dict(decay_mult=0.0),
+            'bias': dict(decay_mult=0.0),
+            'pos_embed': dict(decay_mult=0.),
+            'mask_token': dict(decay_mult=0.),
+            'cls_token': dict(decay_mult=0.)
+        }))
+find_unused_parameters = True
+
+# learning rate scheduler
+param_scheduler = [
+    dict(
+        type='LinearLR',
+        start_factor=1e-4,
+        by_epoch=True,
+        begin=0,
+        end=40,
+        convert_to_iter_based=True),
+    dict(
+        type='CosineAnnealingLR',
+        T_max=360,
+        by_epoch=True,
+        begin=40,
+        end=400,
+        convert_to_iter_based=True)
+]
+
+# runtime settings
+train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=400)
+default_hooks = dict(
+    # only keeps the latest 3 checkpoints
+    checkpoint=dict(type='CheckpointHook', interval=1, max_keep_ckpts=3))
+
+randomness = dict(seed=0, diff_rank_seed=True)
+
+# auto resume
+resume = True
+
+# NOTE: `auto_scale_lr` is for automatically scaling LR
+# based on the actual training batch size.
+auto_scale_lr = dict(base_batch_size=4096)

configs/eva/metafile.yml

+Collections:
+  - Name: EVA
+    Metadata:
+      Architecture:
+        - Attention Dropout
+        - Convolution
+        - Dense Connections
+        - Dropout
+        - GELU
+        - Layer Normalization
+        - Multi-Head Attention
+        - Scaled Dot-Product Attention
+        - Tanh Activation
+    Paper:
+      Title: 'EVA: Exploring the Limits of Masked Visual Representation Learning at
+        Scale'
+      URL: https://arxiv.org/abs/2211.07636
+    README: configs/eva/README.md
+    Code:
+      URL: null
+      Version: null
+
+Models:
+  - Name: eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k
+    Metadata:
+      Epochs: 400
+      Batch Size: 4096
+      FLOPs: 17581972224
+      Parameters: 111776512
+      Training Data: ImageNet-1k
+    In Collection: EVA
+    Results: null
+    Weights: https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k_20221226-26d90f07.pth
+    Config: configs/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k.py
+    Downstream:
+      - vit-base-p16_eva-mae-style-pre_8xb128-coslr-100e_in1k
+      - vit-base-p16_eva-mae-style-pre_8xb2048-linear-coslr-100e_in1k
+  - Name: vit-base-p16_eva-mae-style-pre_8xb128-coslr-100e_in1k
+    Metadata:
+      Epochs: 100
+      Batch Size: 1024
+      FLOPs: 17581215744
+      Parameters: 86566120
+      Training Data: ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Task: Image Classification
+        Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 83.7
+    Weights: https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k/vit-base-p16_ft-8xb128-coslr-100e_in1k_20221226-f61cf992.pth
+    Config: configs/eva/benchmarks/vit-base-p16_8xb128-coslr-100e_in1k.py
+  - Name: vit-base-p16_eva-mae-style-pre_8xb2048-linear-coslr-100e_in1k
+    Metadata:
+      Epochs: 100
+      Batch Size: 16384
+      FLOPs: 17581972992
+      Parameters: 86567656
+      Training Data: ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Task: Image Classification
+        Dataset: ImageNet-1k
+        Metrics:
+          Top 1 Accuracy: 69.0
+    Weights: https://download.openmmlab.com/mmselfsup/1.x/eva/eva-mae-style_vit-base-p16_16xb256-coslr-400e_in1k/vit-base-p16_linear-8xb2048-coslr-100e_in1k/vit-base-p16_linear-8xb2048-coslr-100e_in1k_20221226-ef51bf09.pth
+    Config: configs/eva/benchmarks/vit-base-p16_8xb2048-linear-coslr-100e_in1k.py
+  - Name: beit-l-p14_eva-pre_3rdparty_in1k-196px
+    Metadata:
+      FLOPs: 61565981696
+      Parameters: 304142312
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Dataset: ImageNet-1k
+        Task: Image Classification
+        Metrics:
+          Top 1 Accuracy: 87.94
+          Top 5 Accuracy: 98.5
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-pre_3rdparty_in1k-196px_20221214-2adf4d28.pth
+    Config: configs/eva/eva-l-p14_8xb16_in1k-196px.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_l_psz14_196px_1k_ft_88p0.pt
+      Code: https://github.com/baaivision/EVA
+  - Name: beit-l-p14_eva-in21k-pre_3rdparty_in1k-196px
+    Metadata:
+      FLOPs: 61565981696
+      Parameters: 304142312
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Dataset: ImageNet-1k
+        Task: Image Classification
+        Metrics:
+          Top 1 Accuracy: 88.58
+          Top 5 Accuracy: 98.65
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-in21k-pre_3rdparty_in1k-196px_20221213-b730c7e7.pth
+    Config: configs/eva/eva-l-p14_8xb16_in1k-196px.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_l_psz14_196px_21k_to_1k_ft_88p6.pt
+      Code: https://github.com/baaivision/EVA
+  - Name: beit-l-p14_3rdparty-eva_in21k
+    Metadata:
+      FLOPs: 81075147776
+      Parameters: 303178752
+      Training Data:
+        - ImageNet-21k
+    In Collection: EVA
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_3rdparty-mim_in21k_20221213-3a5da50b.pth
+    Config: configs/eva/eva-l-p14_headless.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_l_psz14.pt
+      Code: https://github.com/baaivision/EVA
+    Downstream:
+      - beit-l-p14_eva-pre_3rdparty_in21k
+      - beit-l-p14_eva-pre_3rdparty_in1k-336px
+      - beit-l-p14_eva-pre_3rdparty_in1k-196px
+  - Name: beit-l-p14_eva-pre_3rdparty_in21k
+    Metadata:
+      FLOPs: 81075147776
+      Parameters: 303178752
+      Training Data:
+        - ImageNet-21k
+    In Collection: EVA
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-pre_3rdparty_in21k_20221213-8f194fa2.pth
+    Config: configs/eva/eva-l-p14_headless.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_l_psz14_21k_ft.pt
+      Code: https://github.com/baaivision/EVA
+  - Name: beit-l-p14_eva-pre_3rdparty_in1k-336px
+    Metadata:
+      FLOPs: 191100916736
+      Parameters: 304531432
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Dataset: ImageNet-1k
+        Task: Image Classification
+        Metrics:
+          Top 1 Accuracy: 88.66
+          Top 5 Accuracy: 98.75
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-pre_3rdparty_in1k-336px_20221214-07785cfd.pth
+    Config: configs/eva/eva-l-p14_8xb16_in1k-336px.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_l_psz14_336px_1k_ft_88p65.pt
+      Code: https://github.com/baaivision/EVA
+    Downstream:
+      - beit-l-p14_eva-in21k-pre_3rdparty_in1k-336px
+      - beit-l-p14_eva-in21k-pre_3rdparty_in1k-196px
+  - Name: beit-l-p14_eva-in21k-pre_3rdparty_in1k-336px
+    Metadata:
+      FLOPs: 191100916736
+      Parameters: 304531432
+      Training Data:
+        - ImageNet-21k
+        - ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Dataset: ImageNet-1k
+        Task: Image Classification
+        Metrics:
+          Top 1 Accuracy: 89.17
+          Top 5 Accuracy: 98.86
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-l-p14_mim-in21k-pre_3rdparty_in1k-336px_20221213-f25b7634.pth
+    Config: configs/eva/eva-l-p14_8xb16_in1k-336px.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_l_psz14_336px_21k_to_1k_ft_89p2.pt
+      Code: https://github.com/baaivision/EVA
+  - Name: beit-g-p16_3rdparty-eva_30m
+    Metadata:
+      FLOPs: 203517463424
+      Parameters: 1011315072
+      Training Data:
+        - merged-30M
+    In Collection: EVA
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p16_3rdparty_30m_20221213-7bed23ee.pth
+    Config: configs/eva/eva-g-p16_headless.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_psz14to16.pt
+      Code: https://github.com/baaivision/EVA
+  - Name: beit-g-p14_3rdparty-eva_30m
+    Metadata:
+      FLOPs: 267174833024
+      Parameters: 1011596672
+      Training Data:
+        - merged-30M
+    In Collection: EVA
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_3rdparty_30m_20221213-3b7aca97.pth
+    Config: configs/eva/eva-g-p14_headless.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_psz14.pt
+      Code: https://github.com/baaivision/EVA
+    Downstream:
+      - beit-g-p14_eva-30m-pre_3rdparty_in21k
+  - Name: beit-g-p14_eva-30m-pre_3rdparty_in21k
+    Metadata:
+      FLOPs: 267174833024
+      Parameters: 1011596672
+      Training Data:
+        - merged-30M
+        - ImageNet-21k
+    In Collection: EVA
+    Results: null
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-pre_3rdparty_in21k_20221213-d72285b7.pth
+    Config: configs/eva/eva-g-p14_headless.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_21k_224px_psz14.pt
+      Code: https://github.com/baaivision/EVA
+    Downstream:
+      - beit-g-p14_eva-30m-in21k-pre_3rdparty_in1k-336px
+      - beit-g-p14_eva-30m-in21k-pre_3rdparty_in1k-560px
+  - Name: beit-g-p14_eva-30m-in21k-pre_3rdparty_in1k-336px
+    Metadata:
+      FLOPs: 620642757504
+      Parameters: 1013005672
+      Training Data:
+        - merged-30M
+        - ImageNet-21k
+        - ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Dataset: ImageNet-1k
+        Task: Image Classification
+        Metrics:
+          Top 1 Accuracy: 89.61
+          Top 5 Accuracy: 98.93
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-in21k-pre_3rdparty_in1k-336px_20221213-210f9071.pth
+    Config: configs/eva/eva-g-p14_8xb16_in1k-336px.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_21k_1k_336px_psz14_ema_89p6.pt
+      Code: https://github.com/baaivision/EVA
+  - Name: beit-g-p14_eva-30m-in21k-pre_3rdparty_in1k-560px
+    Metadata:
+      FLOPs: 1906761591680
+      Parameters: 1014447464
+      Training Data:
+        - merged-30M
+        - ImageNet-21k
+        - ImageNet-1k
+    In Collection: EVA
+    Results:
+      - Dataset: ImageNet-1k
+        Task: Image Classification
+        Metrics:
+          Top 1 Accuracy: 89.71
+          Top 5 Accuracy: 98.96
+    Weights: https://download.openmmlab.com/mmclassification/v0/eva/eva-g-p14_30m-in21k-pre_3rdparty_in1k-560px_20221213-fa1c3652.pth
+    Config: configs/eva/eva-g-p14_8xb16_in1k-560px.py
+    Converted From:
+      Weights: https://huggingface.co/BAAI/EVA/blob/main/eva_21k_1k_560px_psz14_ema_89p7.pt
+      Code: https://github.com/baaivision/EVA

configs/eva02/README.md

+# EVA-02
+
+> [EVA-02: A Visual Representation for Neon Genesis](https://arxiv.org/abs/2303.11331)
+
+<!-- [ALGORITHM] -->
+
+## Abstract
+
+We launch EVA-02, a next-generation Transformer-based visual representation pre-trained to reconstruct strong and robust language-aligned vision features via masked image modeling. With an updated plain Transformer architecture as well as extensive pre-training from an open & accessible giant CLIP vision encoder, EVA-02 demonstrates superior performance compared to prior state-of-the-art approaches across various representative vision tasks, while utilizing significantly fewer parameters and compute budgets. Notably, using exclusively publicly accessible training data, EVA-02 with only 304M parameters achieves a phenomenal 90.0 fine-tuning top-1 accuracy on ImageNet-1K val set.  Additionally, our EVA-02-CLIP can reach up to 80.4 zero-shot top-1 on ImageNet-1K, outperforming the previous largest & best open-sourced CLIP with only ~1/6 parameters and ~1/6 image-text training data. We offer four EVA-02 variants in various model sizes, ranging from 6M to 304M parameters, all with impressive performance. To facilitate open accessand open research, we release the complete suite of EVA-02 to the community.
+
+<div align=center>
+<img src="https://user-images.githubusercontent.com/40905160/229037980-b83dceb5-41d6-406c-a20b-63b83c80136d.png" width="70%" alt="TrV builds upon the original plain ViT architecture and includes several enhancements: SwinGLU FFN, sub-LN, 2D RoPE, and JAX weight initialization. To keep the parameter & FLOPs consistent with the baseline, the FFN hidden dim of SwiGLU is 2/3× of the typical MLP counterpart."/>
+</div>
+
+## How to use it?
+
+<!-- [TABS-BEGIN] -->
+
+**Predict image**
+
+```python
+from mmpretrain import inference_model
+
+predict = inference_model('vit-tiny-p14_eva02-in21k-pre_3rdparty_in1k-336px', 'demo/bird.JPEG')
+print(predict['pred_class'])
+print(predict['pred_score'])
+```
+
+**Use the model**
+
+```python
+import torch
+from mmpretrain import get_model
+
+model = get_model('vit-tiny-p14_eva02-in21k-pre_3rdparty_in1k-336px', pretrained=True)
+inputs = torch.rand(1, 3, 336, 336)
+out = model(inputs)
+print(type(out))
+# To extract features.
+feats = model.extract_feat(inputs)
+print(type(feats))
+```
+
+**Train/Test Command**
+
+Prepare your dataset according to the [docs](https://mmpretrain.readthedocs.io/en/latest/user_guides/dataset_prepare.html#prepare-dataset).
+
+Train:
+
+```shell
+python tools/train.py configs/eva02/eva02-tiny-p14_in1k.py
+```
+
+Test:
+
+```shell
+python tools/test.py configs/eva02/eva02-tiny-p14_in1k.py /path/to/eva02-tiny-p14_in1k.pth
+```
+
+<!-- [TABS-END] -->
+
+## Models and results
+
+### Pretrained models
+
+| Model                             | Params (M) | Flops (G) |                Config                 |                                                   Download                                                    |
+| :-------------------------------- | :--------: | :-------: | :-----------------------------------: | :-----------------------------------------------------------------------------------------------------------: |
+| `vit-tiny-p14_eva02-pre_in21k`\*  |    5.50    |   1.70    | [config](eva02-tiny-p14_headless.py)  | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-tiny-p14_pre_in21k_20230505-d703e7b1.pth)  |
+| `vit-small-p14_eva02-pre_in21k`\* |   21.62    |   6.14    | [config](eva02-small-p14_headless.py) | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-small-p14_pre_in21k_20230505-3175f463.pth) |
+| `vit-base-p14_eva02-pre_in21k`\*  |   85.77    |   23.22   | [config](eva02-base-p14_headless.py)  | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-base-p14_pre_in21k_20230505-2f2d4d3c.pth)  |
+| `vit-large-p14_eva02-pre_in21k`\* |   303.29   |   81.15   | [config](eva02-large-p14_headless.py) | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-large-p14_pre_in21k_20230505-9072de5d.pth) |
+| `vit-large-p14_eva02-pre_m38m`\*  |   303.29   |   81.15   | [config](eva02-large-p14_headless.py) | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-large-p14_pre_m38m_20230505-b8a1a261.pth)  |
+
+- The input size / patch size of MIM pre-trained EVA-02 is `224x224` / `14x14`.
+
+*Models with * are converted from the [official repo](https://github.com/baaivision/EVA).*
+
+### Image Classification on ImageNet-1k
+
+#### (*w/o* IN-21K intermediate fine-tuning)
+
+| Model                                                 |      Pretrain      | Params (M) | Flops (G) | Top-1 (%) | Top-5 (%) |               Config                |                         Download                          |
+| :---------------------------------------------------- | :----------------: | :--------: | :-------: | :-------: | :-------: | :---------------------------------: | :-------------------------------------------------------: |
+| `vit-tiny-p14_eva02-in21k-pre_3rdparty_in1k-336px`\*  | EVA02 ImageNet-21k |    5.76    |   4.68    |   80.69   |   95.54   | [config](./eva02-tiny-p14_in1k.py)  | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-tiny-p14_in21k-pre_3rdparty_in1k-336px_20230505-a4e8708a.pth) |
+| `vit-small-p14_eva02-in21k-pre_3rdparty_in1k-336px`\* | EVA02 ImageNet-21k |   22.13    |   15.48   |   85.78   |   97.60   | [config](./eva02-small-p14_in1k.py) | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-small-p14_in21k-pre_3rdparty_in1k-336px_20230505-9c5b0e85.pth) |
+| `vit-base-p14_eva02-in21k-pre_3rdparty_in1k-448px`\*  | EVA02 ImageNet-21k |   87.13    |  107.11   |   88.29   |   98.53   | [config](./eva02-base-p14_in1k.py)  | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-base-p14_in21k-pre_3rdparty_in1k-448px_20230505-8ad211c5.pth) |
+
+*Models with * are converted from the  [official repo](https://github.com/baaivision/EVA/tree/master/EVA-02). The config files of these models are only for inference. We haven't reproduce the training results.*
+
+#### (*w* IN-21K intermediate fine-tuning)
+
+| Model                                                 |      Pretrain      | Params (M) | Flops (G) | Top-1 (%) | Top-5 (%) |               Config                |                         Download                          |
+| :---------------------------------------------------- | :----------------: | :--------: | :-------: | :-------: | :-------: | :---------------------------------: | :-------------------------------------------------------: |
+| `vit-base-p14_eva02-in21k-pre_in21k-medft_3rdparty_in1k-448px`\* | EVA02 ImageNet-21k |   87.13    |  107.11   |   88.47   |   98.62   | [config](./eva02-base-p14_in1k.py)  | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-base-p14_in21k-pre_in21k-medft_3rdparty_in1k-448px_20230505-5cd4d87f.pth) |
+| `vit-large-p14_eva02-in21k-pre_in21k-medft_3rdparty_in1k-448px`\* | EVA02 ImageNet-21k |   305.08   |  362.33   |   89.65   |   98.95   | [config](./eva02-large-p14_in1k.py) | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-large-p14_in21k-pre_in21k-medft_3rdparty_in1k-448px_20230505-926d1599.pth) |
+| `vit-large-p14_eva02_m38m-pre_in21k-medft_3rdparty_in1k-448px`\* |  EVA02 Merged-38M  |   305.10   |  362.33   |   89.83   |   99.00   | [config](./eva02-large-p14_in1k.py) | [model](https://download.openmmlab.com/mmpretrain/v1.0/eva02/eva02-large-p14_m38m-pre_in21k-medft_3rdparty_in1k-448px_20230505-150dc5ed.pth) |
+
+*Models with * are converted from the  [official repo](https://github.com/baaivision/EVA/tree/master/EVA-02). The config files of these models are only for inference. We haven't reproduce the training results.*
+
+## Citation
+
+```bibtex
+@article{EVA-02,
+  title={EVA-02: A Visual Representation for Neon Genesis},
+  author={Yuxin Fang and Quan Sun and Xinggang Wang and Tiejun Huang and Xinlong Wang and Yue Cao},
+  journal={arXiv preprint arXiv:2303.11331},
+  year={2023}
+}
+```