File size: 1,775 Bytes
b3722a8
d390139
b3722a8
 
d390139
b3722a8
 
 
 
 
 
 
 
 
49d0cfc
b3722a8
49d0cfc
b3722a8
 
49d0cfc
 
 
 
 
b3722a8
 
 
 
49d0cfc
b3722a8
49d0cfc
b3722a8
49d0cfc
 
 
 
 
b3722a8
 
 
 
 
49d0cfc
b3722a8
 
49d0cfc
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
import os

import yaml
from huggingface_hub import HfApi, HfFileSystem, hf_hub_download

from mlip_arena.models import MLIP
from mlip_arena.models import REGISTRY as MODEL_REGISTRY

with open(os.path.join(os.path.dirname(__file__), "registry.yaml")) as f:
    REGISTRY = yaml.load(f, Loader=yaml.FullLoader)


class Task:
    def __init__(self):
        self.name: str = self.__class__.__name__  # display name on the leaderboard

    def run_local(self, model: MLIP):
        """Run the task using the given model and return the results"""
        raise NotImplementedError

    def run_hf(self, model: MLIP):
        """Run the task using the given model and return the results"""
        raise NotImplementedError

        # Calcualte evaluation metrics and postprocessed data
        api = HfApi()
        api.upload_file(
            path_or_fileobj="results.json",
            path_in_repo=f"{self.__class__.__name__}/{model.__class__.__name__}/results.json",  # Upload to a specific folder
            repo_id="atomind/mlip-arena",
            repo_type="dataset",
        )

    def run_nersc(self, model: MLIP):
        """Run the task using the given model and return the results"""
        raise NotImplementedError

    def get_results(self):
        """Get the results from the task"""

        # fs = HfFileSystem()
        # files = fs.glob(f"datasets/atomind/mlip-arena/{self.__class__.__name__}/*/*.json")

        for model, metadata in MODEL_REGISTRY.items():
            results = hf_hub_download(
                repo_id="atomind/mlip-arena",
                filename="results.json",
                subfolder=f"{self.__class__.__name__}/{model}",
                repo_type="dataset",
                revision=None,
            )

        return results