File size: 3,590 Bytes
a76b907 |
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 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 |
from dataclasses import dataclass, make_dataclass
from src.display.about import create_task_list
def fields(raw_class):
return [v for k, v in raw_class.__dict__.items() if k[:2] != "__" and k[-2:] != "__"]
# These classes are for user facing column names,
# to avoid having to change them all around the code
# when a modif is needed
@dataclass
class ColumnContent:
name: str
type: str
displayed_by_default: bool
hidden: bool = False
never_hidden: bool = False
dummy: bool = False
Tasks, Groups = create_task_list()
## Leaderboard columns
auto_eval_column_dict = []
# Init
auto_eval_column_dict.append(["model_submission_date", ColumnContent, ColumnContent("Submission Date", "str", True, never_hidden=True)])
auto_eval_column_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
#Scores
auto_eval_column_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
for task in Tasks:
auto_eval_column_dict.append([task.benchmark, ColumnContent, ColumnContent(task.col_name, "number", True)])
# Dummy column for the search bar (hidden by the custom CSS)
auto_eval_column_dict.append(["dummy", ColumnContent, ColumnContent("model_name_for_query", "str", False, dummy=True)])
# We use make dataclass to dynamically fill the scores from Tasks
AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
## For the queue columns in the submission tab
@dataclass(frozen=True)
class EvalQueueColumn: # Queue column
model = ColumnContent("model", "markdown", True)
submitted_time = ColumnContent("submitted_time", "str", True)
status = ColumnContent("status", "str", True)
# Column selection
COLS = [c.name for c in fields(AutoEvalColumn) if not c.hidden]
TYPES = [c.type for c in fields(AutoEvalColumn) if not c.hidden]
EVAL_COLS = [c.name for c in fields(EvalQueueColumn)]
EVAL_TYPES = [c.type for c in fields(EvalQueueColumn)]
BENCHMARK_COLS = [t.col_name for t in Tasks]
#for grouping
## Leaderboard columns
auto_eval_group_dict = []
# Init
auto_eval_group_dict.append(["model_submission_date", ColumnContent, ColumnContent("Submission Date", "str", True, never_hidden=True)])
auto_eval_group_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
#Scores
auto_eval_group_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
for task in Groups:
auto_eval_group_dict.append([task.benchmark, ColumnContent, ColumnContent(task.col_name, "number", True)])
# Dummy column for the search bar (hidden by the custom CSS)
auto_eval_group_dict.append(["dummy", ColumnContent, ColumnContent("model_name_for_query", "str", False, dummy=True)])
# We use make dataclass to dynamically fill the scores from Tasks
AutoEvalColumnGroup = make_dataclass("AutoEvalColumnGroup", auto_eval_group_dict, frozen=True)
## For the queue columns in the submission tab
@dataclass(frozen=True)
class EvalQueueColumnGroup: # Queue column
model = ColumnContent("model", "markdown", True)
submitted_time = ColumnContent("submitted_time", "str", True)
status = ColumnContent("status", "str", True)
# Column selection
COLS_GROUP = [c.name for c in fields(AutoEvalColumnGroup) if not c.hidden]
TYPES_GROUP = [c.type for c in fields(AutoEvalColumnGroup) if not c.hidden]
EVAL_COLS_GROUP = [c.name for c in fields(EvalQueueColumnGroup)]
EVAL_TYPES_GROUP = [c.type for c in fields(EvalQueueColumnGroup)]
BENCHMARK_COLS_GROUP = [t.col_name for t in Groups] |