Skip to content

Commit

Permalink
Update model download links
Browse files Browse the repository at this point in the history
  • Loading branch information
HYPJUDY authored Apr 24, 2024
1 parent 6d50d49 commit d65767a
Showing 1 changed file with 22 additions and 22 deletions.
44 changes: 22 additions & 22 deletions dit/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,19 +8,19 @@ DiT (Document Image Transformer) is a self-supervised pre-trained Document Image

## What's New
- Demos on HuggingFace: [Document Layout Analysis](https://huggingface.co/spaces/nielsr/dit-document-layout-analysis), [Document Image Classification](https://huggingface.co/spaces/microsoft/document-image-transformer)
- March 2022: release pre-trained checkpoints and fine-tuning codes (DiT-base and DiT-large)
- March 2022: release pre-trained [checkpoints]((https://mail2sysueducn-my.sharepoint.com/:f:/g/personal/huangyp28_mail2_sysu_edu_cn/Eh7si2lpRqZLsmHASlnsw00BewkWLtqtQ3YGw5b0oi4YzQ?e=qKdZ4A)) and fine-tuning [checkpoints](https://mail2sysueducn-my.sharepoint.com/:f:/g/personal/huangyp28_mail2_sysu_edu_cn/EsK1oCJRNnZBi_VfswfD2j8BKCMad94BDmJWlX5rKx_L0Q?e=NVDb1k) & codes (DiT-base and DiT-large)
- March 2022: release preprint in [arXiv](https://arxiv.org/abs/2203.02378)

## Pretrained models
## [Pretrained models](https://mail2sysueducn-my.sharepoint.com/:f:/g/personal/huangyp28_mail2_sysu_edu_cn/Eh7si2lpRqZLsmHASlnsw00BewkWLtqtQ3YGw5b0oi4YzQ?e=qKdZ4A)

We provide two DiT weights pretrained on [IIT-CDIP Test Collection 1.0](https://dl.acm.org/doi/10.1145/1148170.1148307). The models were pretrained with 224x224 resolution.

- `DiT-base`: #layer=12; hidden=768; FFN factor=4x; #head=12; patch=16x16 (#parameters: 86M)
- `DiT-large`: #layer=24; hidden=1024; FFN factor=4x; #head=16; patch=16x16 (#parameters: 304M)

Download checkpoints that are **self-supervised pretrained** on IIT-CDIP Test Collection 1.0:
- DiT-base: [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D)
- DiT-large: [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D)
- DiT-base: [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx)
- DiT-large: [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc)


## Setup
Expand Down Expand Up @@ -57,8 +57,8 @@ We summarize the validation results as follows. We also provide the fine-tuned w

| name | initialized checkpoint | resolution | accuracy | weight |
|------------|:----------------------------------------|:----------:|:-------:|-----|
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | 224x224 | 92.11 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/rvlcdip_dit-b.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | 224x224 | 92.69 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/rvlcdip_dit-l.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | 224x224 | 92.11 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EdZaGg6kCnxAtGyWUFfVQTABbMsp6zcW78JKVpk0XsnEmw) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | 224x224 | 92.69 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EWHOJYI1bZVKr-saboJTWt8B-cS2RRuCZ-TNEfvcvriG3g?e=ykRYDm) |


## Fine-tuning on PubLayNet (Document Layout Analysis)
Expand All @@ -67,10 +67,10 @@ We summarize the validation results as follows. We also provide the fine-tuned w

| name | initialized checkpoint | detection algorithm | mAP| weight |
|------------|:----------------------------------------|:----------:|-------------------|-----|
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 0.935 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/publaynet_dit-b_mrcnn.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 0.941 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/publaynet_dit-l_mrcnn.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 0.945 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/publaynet_dit-b_cascade.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 0.949 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/publaynet_dit-l_cascade.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Mask R-CNN | 0.935 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EdRCgV60BxNJkMW-DUwg5toBYrfZ13Taiuwidx24LL9uCQ?e=zXH3et) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Mask R-CNN | 0.941 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EbiF6b0pushDuMGeL13uEFMBI-tO8UdGuOaKF1jAJhqjWg?e=hKhaL7) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Cascade R-CNN | 0.945 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ebu8BdNRx6ZBp4t1KHWyy8cBeytG3evrVm8ae7XVf8Ox0A?e=hlfaML) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Cascade R-CNN | 0.949 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ESKnk2I_O09Em52V1xb2Ux0BrO_Z-7cuzL3H1KQRxipb7Q?e=iqTfGc) |

## Fine-tuning on ICDAR 2019 cTDaR (Table Detection)

Expand All @@ -80,28 +80,28 @@ We summarize the validation results as follows. We also provide the fine-tuned w

| name | initialized checkpoint | detection algorithm | Weighted Average F1 | weight |
|------------|:----------------------------------------|:----------:|-------------------|-----|
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 94.74 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19modern_dit-b_mrcnn.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 95.50 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19modern_dit-l_mrcnn.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 95.85 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19modern_dit-b_cascade.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 96.29 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19modern_dit-l_cascade.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Mask R-CNN | 94.74 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EfJ-YwqPB99FjDL4M-Vc8SgBDLlU3A-YOVlmGy6Cpkwc4A?e=2hr83V) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Mask R-CNN | 95.50 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EfzBky0q6ulGr5AWFDMJOAUBkQu1HWCfeL8Cuuqvdcb5ow?e=tZysTu) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Cascade R-CNN | 95.85 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EeK-N5V6FFhKkReVk6VPWg8Be_PRySGwSBSUyNVjz8Djew?e=fnVxUg) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Cascade R-CNN | 96.29 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EWZYlqMX8HpMoSTkdBOBCY0BWtYlXPfXfWavDsXUjG1xWg?e=dR4YUc) |

**Archival**

| name | initialized checkpoint | detection algorithm | Weighted Average F1 | weight |
|------------|:----------------------------------------|:----------:|-------------------|-----|
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 96.24 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19archival_dit-b_mrcnn.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 96.46 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19archival_dit-l_mrcnn.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 96.63 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19archival_dit-b_cascade.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 97.00 | [link](https://layoutlm.blob.core.windows.net/dit/dit-fts/icdar19archival_dit-l_cascade.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Mask R-CNN | 96.24 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EWAzBbNFunZBvPXVo7oeLK0B52fqOofE60JhzYVmEo4SSg?e=IOQgbl) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Mask R-CNN | 96.46 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ETI0TwK9MsdBlK5k2KDPRzcBDxpQ1tlS6H2qS94SI0s2rQ?e=fx3htJ) |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Cascade R-CNN | 96.63 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EWmpA4gS58tOj8kPmIl72ZQBWh11eqsjRKR30ZGTRLYbOg?e=yXdfp8) |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Cascade R-CNN | 97.00 | [link](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/EQY8PZMPCIVKrtJuwJOdSRABOwwsP937dr0EHKCILhCOgQ?e=zeCuNa) |

**Combined (Combine the inference results of Modern and Archival)**

| name | initialized checkpoint | detection algorithm | Weighted Average F1 | weight |
|------------|:----------------------------------------|:----------:|-------------------|-----|
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 95.30 | - |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Mask R-CNN | 95.85 | - |
| DiT-base | [dit_base_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-base-224-p16-500k-62d53a.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 96.14 | - |
| DiT-large | [dit_large_patch16_224](https://layoutlm.blob.core.windows.net/dit/dit-pts/dit-large-224-p16-500k-d7a2fb.pth?sv=2022-11-02&ss=b&srt=o&sp=r&se=2033-06-08T16:48:15Z&st=2023-06-08T08:48:15Z&spr=https&sig=a9VXrihTzbWyVfaIDlIT1Z0FoR1073VB0RLQUMuudD4%3D) | Cascade R-CNN | 96.55 | - |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Mask R-CNN | 95.30 | - |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Mask R-CNN | 95.85 | - |
| DiT-base | [dit_base_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/ERfMnkSl2mZFlv68E6xxhH4BvJVQOYcwpna3QMbbbLDQCA?e=tOPZRx) | Cascade R-CNN | 96.14 | - |
| DiT-large | [dit_large_patch16_224](https://mail2sysueducn-my.sharepoint.com/:u:/g/personal/huangyp28_mail2_sysu_edu_cn/Ec3vqfjSDLBMoVI9WIK6vZUBEtEFlOD163ChojDQzPoUvQ?e=ccEeNc) | Cascade R-CNN | 96.55 | - |


## Citation
Expand Down

0 comments on commit d65767a

Please sign in to comment.