Spaces:
Running
Running
fix
Browse files
tlem.py
CHANGED
@@ -30,7 +30,7 @@ from functools import cached_property
|
|
30 |
TextGenerationPipeline = Callable[[Iterable[str]], list[str]]
|
31 |
|
32 |
|
33 |
-
from evaluate import
|
34 |
|
35 |
|
36 |
@dataclass
|
@@ -40,7 +40,7 @@ class Task:
|
|
40 |
# metrics: list[str] = field(default_factory=list)
|
41 |
metric_name: str | tuple[str, str] = "gsm8k"
|
42 |
input_column: str = "question"
|
43 |
-
label_column: str
|
44 |
prompt: Optional[Callable | str] = None
|
45 |
|
46 |
@cached_property
|
@@ -223,3 +223,8 @@ class ReasoningMetric(evaluate.Metric):
|
|
223 |
# results["scores"] = scores
|
224 |
|
225 |
return results
|
|
|
|
|
|
|
|
|
|
|
|
30 |
TextGenerationPipeline = Callable[[Iterable[str]], list[str]]
|
31 |
|
32 |
|
33 |
+
from evaluate import load
|
34 |
|
35 |
|
36 |
@dataclass
|
|
|
40 |
# metrics: list[str] = field(default_factory=list)
|
41 |
metric_name: str | tuple[str, str] = "gsm8k"
|
42 |
input_column: str = "question"
|
43 |
+
label_column: str = "reference"
|
44 |
prompt: Optional[Callable | str] = None
|
45 |
|
46 |
@cached_property
|
|
|
223 |
# results["scores"] = scores
|
224 |
|
225 |
return results
|
226 |
+
|
227 |
+
# %%
|
228 |
+
|
229 |
+
load("sustech/tlem", "gsm8k")
|
230 |
+
|