Update space
Browse files- src/display/utils.py +3 -2
- src/leaderboard/read_evals.py +3 -2
src/display/utils.py
CHANGED
@@ -51,7 +51,7 @@ auto_eval_column_dict = []
|
|
51 |
|
52 |
# Init
|
53 |
auto_eval_column_dict.append(["model", ColumnContent, field(default_factory=lambda: ColumnContent("Model", "markdown", True, never_hidden=True))])
|
54 |
-
auto_eval_column_dict.append(["license", ColumnContent, field(default_factory=lambda: ColumnContent("
|
55 |
|
56 |
# new columns
|
57 |
for domain in Domains:
|
@@ -59,7 +59,8 @@ for domain in Domains:
|
|
59 |
|
60 |
auto_eval_column_dict.append(["organization", ColumnContent, field(default_factory=lambda: ColumnContent("Organization", "str", False))])
|
61 |
auto_eval_column_dict.append(["knowledge_cutoff", ColumnContent, field(default_factory=lambda: ColumnContent("Knowledge cutoff", "str", False))])
|
62 |
-
auto_eval_column_dict.append(["score", ColumnContent, field(default_factory=lambda: ColumnContent("Score", "number", True))])
|
|
|
63 |
auto_eval_column_dict.append(["rank", ColumnContent, field(default_factory=lambda: ColumnContent("Rank", "number", True))])
|
64 |
|
65 |
|
|
|
51 |
|
52 |
# Init
|
53 |
auto_eval_column_dict.append(["model", ColumnContent, field(default_factory=lambda: ColumnContent("Model", "markdown", True, never_hidden=True))])
|
54 |
+
auto_eval_column_dict.append(["license", ColumnContent, field(default_factory=lambda: ColumnContent("License", "str", False))])
|
55 |
|
56 |
# new columns
|
57 |
for domain in Domains:
|
|
|
59 |
|
60 |
auto_eval_column_dict.append(["organization", ColumnContent, field(default_factory=lambda: ColumnContent("Organization", "str", False))])
|
61 |
auto_eval_column_dict.append(["knowledge_cutoff", ColumnContent, field(default_factory=lambda: ColumnContent("Knowledge cutoff", "str", False))])
|
62 |
+
auto_eval_column_dict.append(["score", ColumnContent, field(default_factory=lambda: ColumnContent("Average Score", "number", True))])
|
63 |
+
auto_eval_column_dict.append(["score_sd", ColumnContent, field(default_factory=lambda: ColumnContent("Score SD", "number", True))])
|
64 |
auto_eval_column_dict.append(["rank", ColumnContent, field(default_factory=lambda: ColumnContent("Rank", "number", True))])
|
65 |
|
66 |
|
src/leaderboard/read_evals.py
CHANGED
@@ -63,6 +63,7 @@ class ModelResult:
|
|
63 |
AutoEvalColumn.rank.name: None, # placeholder for the rank
|
64 |
AutoEvalColumn.model.name: self.model,
|
65 |
AutoEvalColumn.score.name: self.results[Domains.dim0.value.dimension],
|
|
|
66 |
AutoEvalColumn.license.name: self.license,
|
67 |
AutoEvalColumn.organization.name: self.org,
|
68 |
AutoEvalColumn.knowledge_cutoff.name: self.knowledge_cutoff,
|
@@ -82,8 +83,8 @@ class ModelResult:
|
|
82 |
# for task in Tasks:
|
83 |
# data_dict[task.value.col_name] = self.results[task.value.benchmark]
|
84 |
|
85 |
-
for domain in Domains:
|
86 |
-
|
87 |
|
88 |
return data_dict
|
89 |
|
|
|
63 |
AutoEvalColumn.rank.name: None, # placeholder for the rank
|
64 |
AutoEvalColumn.model.name: self.model,
|
65 |
AutoEvalColumn.score.name: self.results[Domains.dim0.value.dimension],
|
66 |
+
AutoEvalColumn.score_sd.name: None, # placeholder for the score sd
|
67 |
AutoEvalColumn.license.name: self.license,
|
68 |
AutoEvalColumn.organization.name: self.org,
|
69 |
AutoEvalColumn.knowledge_cutoff.name: self.knowledge_cutoff,
|
|
|
83 |
# for task in Tasks:
|
84 |
# data_dict[task.value.col_name] = self.results[task.value.benchmark]
|
85 |
|
86 |
+
# for domain in Domains:
|
87 |
+
# data_dict[domain.value.col_name] = self.results[domain.value.dimension]
|
88 |
|
89 |
return data_dict
|
90 |
|