Spaces:
Running
Running
yangzhitao
commited on
Commit
·
ba366ad
1
Parent(s):
58bbf33
refactor: update benchmark display in submit tab
Browse files- app.py +4 -4
- src/schemas/meta_toml.py +10 -1
app.py
CHANGED
|
@@ -466,11 +466,11 @@ def create_submit_tab(tab_id: int, demo: gr.Blocks):
|
|
| 466 |
for benchmark in benchmarks_list:
|
| 467 |
with gr.Row():
|
| 468 |
benchmark_checkbox = gr.Checkbox(
|
| 469 |
-
label=f"{benchmark.title} ({benchmark.
|
| 470 |
value=False,
|
| 471 |
)
|
| 472 |
result_input = gr.Number(
|
| 473 |
-
label="Result Value",
|
| 474 |
value=None,
|
| 475 |
interactive=True,
|
| 476 |
visible=False,
|
|
@@ -567,8 +567,8 @@ def create_submit_tab(tab_id: int, demo: gr.Blocks):
|
|
| 567 |
strict=True,
|
| 568 |
):
|
| 569 |
if checkbox_checked:
|
| 570 |
-
|
| 571 |
-
results[benchmark.key] = {
|
| 572 |
|
| 573 |
if not results:
|
| 574 |
raise ValueError("At least one benchmark result is required.")
|
|
|
|
| 466 |
for benchmark in benchmarks_list:
|
| 467 |
with gr.Row():
|
| 468 |
benchmark_checkbox = gr.Checkbox(
|
| 469 |
+
label=f"{benchmark.title} ({benchmark.default_metric_label})",
|
| 470 |
value=False,
|
| 471 |
)
|
| 472 |
result_input = gr.Number(
|
| 473 |
+
label="Result Value (in %, e.g. 42.0 for 42.0%)",
|
| 474 |
value=None,
|
| 475 |
interactive=True,
|
| 476 |
visible=False,
|
|
|
|
| 567 |
strict=True,
|
| 568 |
):
|
| 569 |
if checkbox_checked:
|
| 570 |
+
metric_key = benchmark.default_metric
|
| 571 |
+
results[benchmark.key] = {metric_key: float(result_value) if result_value is not None else None}
|
| 572 |
|
| 573 |
if not results:
|
| 574 |
raise ValueError("At least one benchmark result is required.")
|
src/schemas/meta_toml.py
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
from functools import cached_property
|
| 2 |
|
| 3 |
-
from pydantic import BaseModel, ConfigDict
|
| 4 |
from typing_extensions import Self
|
| 5 |
|
| 6 |
|
|
@@ -84,6 +84,15 @@ class _HashableComparableMixin(BaseModel):
|
|
| 84 |
class MetaToml_Benchmark(_HashableComparableMixin):
|
| 85 |
disabled: bool = False
|
| 86 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
|
| 88 |
class MetaToml_Model(_HashableComparableMixin): ...
|
| 89 |
|
|
|
|
| 1 |
from functools import cached_property
|
| 2 |
|
| 3 |
+
from pydantic import BaseModel, ConfigDict, computed_field
|
| 4 |
from typing_extensions import Self
|
| 5 |
|
| 6 |
|
|
|
|
| 84 |
class MetaToml_Benchmark(_HashableComparableMixin):
|
| 85 |
disabled: bool = False
|
| 86 |
|
| 87 |
+
@computed_field
|
| 88 |
+
@property
|
| 89 |
+
def default_metric(self) -> str:
|
| 90 |
+
return "caa" if self.key.startswith("site") else "acc"
|
| 91 |
+
|
| 92 |
+
@property
|
| 93 |
+
def default_metric_label(self) -> str:
|
| 94 |
+
return "CAA" if self.default_metric == "caa" else "Acc."
|
| 95 |
+
|
| 96 |
|
| 97 |
class MetaToml_Model(_HashableComparableMixin): ...
|
| 98 |
|