Skip to content

docs: bit misspell in README comments #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

docs: bit misspell in README comments #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
094b4ec
docs: bit misspell in comments
guspan-tanadi Sep 18, 2023
8b32e97
docs: bit misspell in README
guspan-tanadi Sep 18, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ Author: Kaicheng Yang, Jiankang Deng, Xiang An, Jiawei Li, Ziyong Feng, Jia Guo,

1、Download YFCC15M

The YFCC15M dataset we used is [YFCC15M-DeCLIP](https://arxiv.org/abs/2110.05208), we download it from the [repo](https://github.com/AdamRain/YFCC15M_downloader), finally we successful donwload 15061515 image-text pairs.
The YFCC15M dataset we used is [YFCC15M-DeCLIP](https://arxiv.org/abs/2110.05208), we download it from the [repo](https://github.com/AdamRain/YFCC15M_downloader), finally we successful download 15061515 image-text pairs.

2、Generate synthetic caption

Expand All @@ -40,7 +40,7 @@ Author: Kaicheng Yang, Jiankang Deng, Xiang An, Jiawei Li, Ziyong Feng, Jia Guo,

- ### Pretrained Model Weight

You can download the pretrained model weight from [Google Drive](https://drive.google.com/file/d/1AqSHisCKZOZ16Q3sYguK6zIZIuwwEriE/view?usp=share_link) or [BaiduYun](https://pan.baidu.com/s/10dFfvGMWeaTXUyrZlZlCEw?pwd=xftg), and you can find the traning log in [Google Drive](https://drive.google.com/file/d/1I8gdSQCJAfFamDcVztwW8EQIc_OOK8Xh/view?usp=share_link) or [BaiduYun](https://pan.baidu.com/s/1oz0UVzX2N0Sri7MfwR-kog?pwd=7ki7)
You can download the pretrained model weight from [Google Drive](https://drive.google.com/file/d/1AqSHisCKZOZ16Q3sYguK6zIZIuwwEriE/view?usp=share_link) or [BaiduYun](https://pan.baidu.com/s/10dFfvGMWeaTXUyrZlZlCEw?pwd=xftg), and you can find the training log in [Google Drive](https://drive.google.com/file/d/1I8gdSQCJAfFamDcVztwW8EQIc_OOK8Xh/view?usp=share_link) or [BaiduYun](https://pan.baidu.com/s/1oz0UVzX2N0Sri7MfwR-kog?pwd=7ki7)

- ### Training

Expand All @@ -51,7 +51,7 @@ Author: Kaicheng Yang, Jiankang Deng, Xiang An, Jiawei Li, Ziyong Feng, Jia Guo,

- ### Evaluation

Evaluate zero shot cross-modal retireval
Evaluate zero shot cross-modal retrieval

```
bash run_retrieval.sh
Expand Down
6 changes: 3 additions & 3 deletions text_image_retrieval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
'flickr': flickr30k}

def compute_retrieval(similarity_scores, txt2img, img2txt):
# comput text -> image
# compute text -> image
t2i_similarity_score = similarity_scores.t()
t2i_ranks = torch.zeros(t2i_similarity_score.shape[0])

Expand All @@ -23,7 +23,7 @@ def compute_retrieval(similarity_scores, txt2img, img2txt):
tr10 = 100.0 * len(torch.where(t2i_ranks < 10)[0]) / len(t2i_ranks)
t2i_report_dict = {"r1": tr1, "r5": tr5, "r10": tr10}

#comput image -> text
#compute image -> text
i2t_similarity_score = similarity_scores
i2t_ranks = torch.zeros(i2t_similarity_score.shape[0])
for index, score in enumerate(i2t_similarity_score):
Expand Down Expand Up @@ -136,4 +136,4 @@ def main(args):
parser.add_argument("--model-weight", default= "/mnt/laion/clip/vit_b_16-laion400m_e32-55e67d44.pt")
parser.add_argument("--input-size", default=224, type=int, help="Image resolution.")
args = parser.parse_args()
main(args)
main(args)