galb-dai's picture
Update tiers.
985eb9a
from dataclasses import dataclass
from enum import Enum
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 modification is needed.
@dataclass
class ColumnContent:
name: str
type: str
displayed_by_default: bool
hidden: bool = False
never_hidden: bool = False
@dataclass(frozen=True)
class AutoEvalColumn:
system = ColumnContent("System Name", "markdown", True, never_hidden=True)
organization = ColumnContent("Organization", "str", True, never_hidden=True)
success_rate_overall = ColumnContent("Overall Success (%)", "number", True)
success_rate_tier1 = ColumnContent("Deeper Tier Success (%)", "number", True)
success_rate_tier2 = ColumnContent("Deepest Tier Success (%)", "number", True)
submitted_on = ColumnContent("Submitted On", "datetime", True)
# For the queue columns in the submission tab
@dataclass(frozen=True)
class EvalQueueColumn: # Queue column
model = ColumnContent("model", "markdown", True)
revision = ColumnContent("revision", "str", True)
private = ColumnContent("private", "bool", True)
precision = ColumnContent("precision", "str", True)
weight_type = ColumnContent("weight_type", "str", True)
status = ColumnContent("status", "str", True)
# All the model information that we might need
@dataclass
class ModelDetails:
name: str
display_name: str = ""
class ModelType(Enum):
LLM = ModelDetails(name="LLM")
AgenticLLM = ModelDetails(name="AgenticLLM")
Other = ModelDetails(name="Other")
def to_str(self):
return self.value.name
@staticmethod
def from_str(type: str):
if type == "AgenticLLM":
return ModelType.AgenticLLM
if type == "LLM":
return ModelType.LLM
return ModelType.Other
class Precision(Enum):
float16 = ModelDetails("float16")
bfloat16 = ModelDetails("bfloat16")
Unknown = ModelDetails("?")
COLS = [c.name 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)]