Update README.md
Browse files
README.md
CHANGED
@@ -15,33 +15,3 @@ thumbnail: >-
|
|
15 |
https://cdn-uploads.huggingface.co/production/uploads/5f19edf678d261307936f4c8/4v6TPbN8qa6JptyCFUy-J.png
|
16 |
short_description: Solbench Leaderboard
|
17 |
---
|
18 |
-
|
19 |
-
# Start the configuration
|
20 |
-
|
21 |
-
Most of the variables to change for a default leaderboard are in `src/env.py` (replace the path for your leaderboard) and `src/about.py` (for tasks).
|
22 |
-
|
23 |
-
Results files should have the following format and be stored as json files:
|
24 |
-
```json
|
25 |
-
{
|
26 |
-
"config": {
|
27 |
-
"model_dtype": "torch.float16", # or torch.bfloat16 or 8bit or 4bit
|
28 |
-
"model_name": "path of the model on the hub: org/model",
|
29 |
-
"model_sha": "revision on the hub",
|
30 |
-
},
|
31 |
-
"results": {
|
32 |
-
"task_name": {
|
33 |
-
"metric_name": score,
|
34 |
-
},
|
35 |
-
"task_name2": {
|
36 |
-
"metric_name": score,
|
37 |
-
}
|
38 |
-
}
|
39 |
-
}
|
40 |
-
```
|
41 |
-
|
42 |
-
# Code logic for more complex edits
|
43 |
-
|
44 |
-
You'll find
|
45 |
-
- the main table' columns names and properties in `src/display/utils.py`
|
46 |
-
- the logic to read all results and request files, then convert them in dataframe lines, in `src/leaderboard/read_evals.py`, and `src/populate.py`
|
47 |
-
- teh logic to allow or filter submissions in `src/submission/submit.py` and `src/submission/check_validity.py`
|
|
|
15 |
https://cdn-uploads.huggingface.co/production/uploads/5f19edf678d261307936f4c8/4v6TPbN8qa6JptyCFUy-J.png
|
16 |
short_description: Solbench Leaderboard
|
17 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|