MINGYISU commited on
Commit
7db36fb
·
verified ·
1 Parent(s): 3a7a40d

Update utils.py

Browse files
Files changed (1) hide show
  1. utils.py +13 -27
utils.py CHANGED
@@ -65,11 +65,12 @@ CITATION_BUTTON_TEXT = r"""@article{jiang2024vlm2vec,
65
  year={2024}
66
  }"""
67
 
68
- SUBMIT_INTRODUCTION = """# Submit on MMEB Leaderboard Introduction
69
-
70
- ## Please refer to the [**GitHub page**](https://github.com/TIGER-AI-Lab/VLM2Vec) for detailed instructions about evaluating your model. \n
71
- ## After running the pipelines, please use the script we provided \(e.g., [report_score_v2.py](https://github.com/TIGER-AI-Lab/VLM2Vec/blob/main/experiments/report_score_v2.py)\) to generate the unified score sheet which is like the following format.
72
- ## ⚠️ Please note that you need to submit the JSON file with the following format:
 
73
  ```json
74
  {
75
  "metadata": {
@@ -91,33 +92,18 @@ SUBMIT_INTRODUCTION = """# Submit on MMEB Leaderboard Introduction
91
  },
92
  ... ...
93
  },
94
- "visdoc": {
95
- "ViDoRe": {
96
- "hit@1": 0.5,
97
- "ndcg@1": 0.5,
98
- ... ...
99
- },
100
- ... ...
101
- },
102
  "video": {
103
- "DiDeMo": {
104
- "hit@1": 0.5,
105
- "ndcg@1": 0.5,
106
- ... ...
107
- },
108
- "MSR-VTT": {
109
- ... ...
110
- },
111
  ... ...
112
- }
 
113
  }
114
  }
115
  ```
116
- ## ⚠️ To submit, create a pull request and upload the generated JSON file to the ***scores*** folder, then inform us on [our discord server](https://discord.gg/njyKubdtry), or send us an email at m7su@uwaterloo.ca, including your model's information.\n
117
- Please email us or leave a simple message (simply @ us) in the PR to indicate that you are ready to merge your PR. We will not merge your PR without informing us since we might think you are still editing and not yet ready to merge.
118
- We will review your submission and update the leaderboard accordingly. \n\n
119
- ## Special Instructions for submitting to MMEB Image (Previously MMEB-V1) Leaderboard
120
- We understand that some researchers want to exclusively submit to the Image leaderboard, but unfortunately our current code doesn't support that. To do so, simply run the 36 image datasets only and ignore the other datasets.
121
  The leaderboard will automatically assign a 0 to the missing datasets.
122
  We might be able to hide your model from the other leaderboards in the next leaderboard updates, but for now your model will be shown on all leaderboards and might have a lower rank.
123
  We highly recommend joining our [discord server](https://discord.gg/njyKubdtry), which provides a convenient way to stay informed with latest updates, or share any feedback you have for improving the leaderboard experience. We appreciate your contributions to the MMEB community!
 
65
  year={2024}
66
  }"""
67
 
68
+ SUBMIT_INTRODUCTION = """# Submit on MMEB Leaderboard Introduction \n
69
+ # <span style="color:#FFA500; font-weight:bold;">!!!IMPORTANT: Please follow the steps below to submit your model.</span> \n
70
+ 1. **Step 1️⃣:** Please refer to the [**GitHub page**](https://github.com/TIGER-AI-Lab/VLM2Vec) for detailed instructions about evaluating your model. \n
71
+ 2. **Step 2️⃣:** After running the evaluation pipelines, please use the script we provided \(e.g., [report_score_v2.py](https://github.com/TIGER-AI-Lab/VLM2Vec/blob/main/experiments/report_score_v2.py)\) to generate the unified score sheet.
72
+ - Note: Please double check that your score sheet should look like the following format: \n
73
+ - (⚠️ Remaining steps to be continued after this code block) \n
74
  ```json
75
  {
76
  "metadata": {
 
92
  },
93
  ... ...
94
  },
 
 
 
 
 
 
 
 
95
  "video": {
 
 
 
 
 
 
 
 
96
  ... ...
97
+ },
98
+ ... ...
99
  }
100
  }
101
  ```
102
+ 3. **Step 3️⃣:** Finally, create a pull request and upload the generated JSON file to the ***scores*** folder, then inform us on [our discord server](https://discord.gg/njyKubdtry), or send us an email at m7su@uwaterloo.ca, indicating that your PR is ready to merge.\n
103
+ - Note: Please email us or leave a simple message (simply @ us) in the PR to indicate that you are ready to merge your PR. We will then review your submission and update the leaderboard accordingly. \n\n
104
+ ## ⚠️ Special Instructions for submitting to MMEB Image (Previously MMEB-V1) Leaderboard
105
+ We understand that some researchers want to exclusively submit to the Image leaderboard, but unfortunately our current leaderboard system cannot exclude your model from other modalities' leaderboards.
106
+ To do so, please run the 36 image datasets only and simply ignore the other datasets.
107
  The leaderboard will automatically assign a 0 to the missing datasets.
108
  We might be able to hide your model from the other leaderboards in the next leaderboard updates, but for now your model will be shown on all leaderboards and might have a lower rank.
109
  We highly recommend joining our [discord server](https://discord.gg/njyKubdtry), which provides a convenient way to stay informed with latest updates, or share any feedback you have for improving the leaderboard experience. We appreciate your contributions to the MMEB community!