Hugging Face's logo

wgcban
/
mix-bt

English
self-supervised learning
barlow-twins
Model card Files Community
wgcban commited on
Commit
bde0498
·
1 Parent(s): 1fd6fe7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -13
README.md CHANGED
@@ -134,31 +134,31 @@ sh scripts-transfer-resnet18/[dataset]-to-x.sh
134
  ```
135
 
136
  ## 3 Pre-Trained Checkpoints
137
- Download the pre-trained models from [GitHub (Releases v1.0.0)](https://github.com/wgcban/mix-bt/releases/tag/v1.0.0) and store them in `checkpoints/`. This repository provides pre-trained checkpoints for both [`ResNet-18`](https://arxiv.org/abs/1512.03385) and [`ResNet-50`](https://arxiv.org/abs/1512.03385) architectures.
138
 
139
  #### 3.1 ResNet-18
140
  | Dataset | $d$ | $\lambda_{BT}$ | $\lambda_{reg}$ | Download Link to Pretrained Model | KNN Acc. | Linear Acc. |
141
  | ---------- | --- | ---------- | ---------- | ------------------------ | -------- | ----------- |
142
- | `CIFAR-10` | 1024 | 0.0078125 | 4.0 | [4wdhbpcf_0.0078125_1024_256_cifar10_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/4wdhbpcf_0.0078125_1024_256_cifar10_model.pth) | 90.52 | 92.58 |
143
- | `CIFAR-100` | 1024 | 0.0078125 | 4.0 | [76kk7scz_0.0078125_1024_256_cifar100_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/76kk7scz_0.0078125_1024_256_cifar100_model.pth) | 61.25 | 69.31 |
144
- | `TinyImageNet` | 1024 | 0.0009765 | 4.0 | [02azq6fs_0.0009765_1024_256_tiny_imagenet_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/02azq6fs_0.0009765_1024_256_tiny_imagenet_model.pth) | 38.11 | 51.67 |
145
- | `STL-10` | 1024 | 0.0078125 | 2.0 | [i7det4xq_0.0078125_1024_256_stl10_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/i7det4xq_0.0078125_1024_256_stl10_model.pth) | 88.94 | 91.02 |
146
 
147
  #### 3.2 ResNet-50
148
  | Dataset | $d$ | $\lambda_{BT}$ | $\lambda_{reg}$ | Download Link to Pretrained Model | KNN Acc. | Linear Acc. |
149
  | ---------- | --- | ---------- | ---------- | ------------------------ | -------- | ----------- |
150
- | `CIFAR-10` | 1024 | 0.0078125 | 4.0 | [v3gwgusq_0.0078125_1024_256_cifar10_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/v3gwgusq_0.0078125_1024_256_cifar10_model.pth) | 91.39 | 93.89 |
151
- | `CIFAR-100` | 1024 | 0.0078125 | 4.0 | [z6ngefw7_0.0078125_1024_256_cifar100_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/z6ngefw7_0.0078125_1024_256_cifar100_model.pth) | 64.32 | 72.51 |
152
- | `TinyImageNet` | 1024 | 0.0009765 | 4.0 | [kxlkigsv_0.0009765_1024_256_tiny_imagenet_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/kxlkigsv_0.0009765_1024_256_tiny_imagenet_model.pth) | 42.21 | 51.84 |
153
- | `STL-10` | 1024 | 0.0078125 | 2.0 | [pbknx38b_0.0078125_1024_256_stl10_model.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/pbknx38b_0.0078125_1024_256_stl10_model.pth) | 87.79 | 91.70 |
154
 
155
  **On `ImageNet`**
156
  | # Epochs | $d$ | $\lambda_{BT}$ | $\lambda_{reg}$ | Download Link to Pretrained Model | Linear Acc. |
157
  | ---------- | --- | ---------- | ---------- | ------------------------ | ----------- |
158
- | 300 | 8192 | 0.0051 | 0.0 (BT) | [3on0l4wl_0.0000_8192_1024_imagenet_resnet50.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/3on0l4wl_0.0000_8192_1024_imagenet_resnet50.pth) | 71.3 |
159
- | 300 | 8192 | 0.0051 | 0.0025 | [l418b9zw_0.0025_8192_1024_imagenet_resnet50.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/l418b9zw_0.0025_8192_1024_imagenet_resnet50.pth) | 70.9 |
160
- | 300 | 8192 | 0.0051 | 0.1 | [13awtq23_0.1000_8192_1024_imagenet_resnet50](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/13awtq23_0.1000_8192_1024_imagenet_resnet50.pth) | 71.6 |
161
- | 300 | 8192 | 0.0051 | 1.0 | [3fb1op86_1.0000_8192_1024_imagenet_resnet50.pth](https://github.com/wgcban/mix-bt/releases/download/v1.0.0/3fb1op86_1.0000_8192_1024_imagenet_resnet50.pth) | **72.2** |
162
  | 300 | 8192 | 0.0051 | 3.0 | [TBU]() | TBU |
163
  | 300 | 8192 | 0.0051 | 5.0 | [TBU]() | TBU |
164
 
 
134
  ```
135
 
136
  ## 3 Pre-Trained Checkpoints
137
+ Download the pre-trained models from `checkpoints/` and store them in `checkpoints/`. This repository provides pre-trained checkpoints for both [`ResNet-18`](https://arxiv.org/abs/1512.03385) and [`ResNet-50`](https://arxiv.org/abs/1512.03385) architectures.
138
 
139
  #### 3.1 ResNet-18
140
  | Dataset | $d$ | $\lambda_{BT}$ | $\lambda_{reg}$ | Download Link to Pretrained Model | KNN Acc. | Linear Acc. |
141
  | ---------- | --- | ---------- | ---------- | ------------------------ | -------- | ----------- |
142
+ | `CIFAR-10` | 1024 | 0.0078125 | 4.0 | [4wdhbpcf_0.0078125_1024_256_cifar10_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/4wdhbpcf_0.0078125_1024_256_cifar10_model.pth) | 90.52 | 92.58 |
143
+ | `CIFAR-100` | 1024 | 0.0078125 | 4.0 | [76kk7scz_0.0078125_1024_256_cifar100_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/76kk7scz_0.0078125_1024_256_cifar100_model.pth) | 61.25 | 69.31 |
144
+ | `TinyImageNet` | 1024 | 0.0009765 | 4.0 | [02azq6fs_0.0009765_1024_256_tiny_imagenet_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/02azq6fs_0.0009765_1024_256_tiny_imagenet_model.pth) | 38.11 | 51.67 |
145
+ | `STL-10` | 1024 | 0.0078125 | 2.0 | [i7det4xq_0.0078125_1024_256_stl10_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/i7det4xq_0.0078125_1024_256_stl10_model.pth) | 88.94 | 91.02 |
146
 
147
  #### 3.2 ResNet-50
148
  | Dataset | $d$ | $\lambda_{BT}$ | $\lambda_{reg}$ | Download Link to Pretrained Model | KNN Acc. | Linear Acc. |
149
  | ---------- | --- | ---------- | ---------- | ------------------------ | -------- | ----------- |
150
+ | `CIFAR-10` | 1024 | 0.0078125 | 4.0 | [v3gwgusq_0.0078125_1024_256_cifar10_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/v3gwgusq_0.0078125_1024_256_cifar10_model.pth) | 91.39 | 93.89 |
151
+ | `CIFAR-100` | 1024 | 0.0078125 | 4.0 | [z6ngefw7_0.0078125_1024_256_cifar100_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/z6ngefw7_0.0078125_1024_256_cifar100_model.pth) | 64.32 | 72.51 |
152
+ | `TinyImageNet` | 1024 | 0.0009765 | 4.0 | [kxlkigsv_0.0009765_1024_256_tiny_imagenet_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/kxlkigsv_0.0009765_1024_256_tiny_imagenet_model.pth) | 42.21 | 51.84 |
153
+ | `STL-10` | 1024 | 0.0078125 | 2.0 | [pbknx38b_0.0078125_1024_256_stl10_model.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/pbknx38b_0.0078125_1024_256_stl10_model.pth) | 87.79 | 91.70 |
154
 
155
  **On `ImageNet`**
156
  | # Epochs | $d$ | $\lambda_{BT}$ | $\lambda_{reg}$ | Download Link to Pretrained Model | Linear Acc. |
157
  | ---------- | --- | ---------- | ---------- | ------------------------ | ----------- |
158
+ | 300 | 8192 | 0.0051 | 0.0 (BT) | [3on0l4wl_0.0000_8192_1024_imagenet_resnet50.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/3on0l4wl_0.0000_8192_1024_imagenet_resnet50.pth) | 71.3 |
159
+ | 300 | 8192 | 0.0051 | 0.0025 | [l418b9zw_0.0025_8192_1024_imagenet_resnet50.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/l418b9zw_0.0025_8192_1024_imagenet_resnet50.pth) | 70.9 |
160
+ | 300 | 8192 | 0.0051 | 0.1 | [13awtq23_0.1000_8192_1024_imagenet_resnet50.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/13awtq23_0.1000_8192_1024_imagenet_resnet50.pth) | 71.6 |
161
+ | 300 | 8192 | 0.0051 | 1.0 | [3fb1op86_1.0000_8192_1024_imagenet_resnet50.pth](https://huggingface.co/wgcban/mix-bt/blob/main/checkpoints/3fb1op86_1.0000_8192_1024_imagenet_resnet50.pth) | **72.2** |
162
  | 300 | 8192 | 0.0051 | 3.0 | [TBU]() | TBU |
163
  | 300 | 8192 | 0.0051 | 5.0 | [TBU]() | TBU |
164