Spaces:
Running
Running
Upload folder using huggingface_hub
Browse files
.ruff_cache/0.14.5/10030214861837525307
CHANGED
|
Binary files a/.ruff_cache/0.14.5/10030214861837525307 and b/.ruff_cache/0.14.5/10030214861837525307 differ
|
|
|
.ruff_cache/0.14.5/1217244776251154506
CHANGED
|
Binary files a/.ruff_cache/0.14.5/1217244776251154506 and b/.ruff_cache/0.14.5/1217244776251154506 differ
|
|
|
.ruff_cache/0.14.5/17154366104450809768
CHANGED
|
Binary files a/.ruff_cache/0.14.5/17154366104450809768 and b/.ruff_cache/0.14.5/17154366104450809768 differ
|
|
|
.ruff_cache/0.14.5/17246995203982082100
CHANGED
|
Binary files a/.ruff_cache/0.14.5/17246995203982082100 and b/.ruff_cache/0.14.5/17246995203982082100 differ
|
|
|
.ruff_cache/0.14.5/18101337634414237271
CHANGED
|
Binary files a/.ruff_cache/0.14.5/18101337634414237271 and b/.ruff_cache/0.14.5/18101337634414237271 differ
|
|
|
src/display/utils.py
CHANGED
|
@@ -11,7 +11,7 @@ def fields(raw_class):
|
|
| 11 |
# These classes are for user facing column names,
|
| 12 |
# to avoid having to change them all around the code
|
| 13 |
# when a modif is needed
|
| 14 |
-
@dataclass
|
| 15 |
class ColumnContent:
|
| 16 |
name: str
|
| 17 |
type: str
|
|
@@ -24,23 +24,23 @@ class ColumnContent:
|
|
| 24 |
auto_eval_column_dict = []
|
| 25 |
# Init
|
| 26 |
auto_eval_column_dict.append(
|
| 27 |
-
|
| 28 |
)
|
| 29 |
-
auto_eval_column_dict.append(
|
| 30 |
# Scores
|
| 31 |
for idx, task in enumerate(Tasks):
|
| 32 |
# Only show overall ANLS (first task) by default
|
| 33 |
displayed_by_default = idx == 0
|
| 34 |
auto_eval_column_dict.append(
|
| 35 |
-
|
| 36 |
)
|
| 37 |
# Cost/Efficiency metrics
|
| 38 |
-
auto_eval_column_dict.append(
|
| 39 |
-
auto_eval_column_dict.append(
|
| 40 |
# Model information
|
| 41 |
-
auto_eval_column_dict.append(
|
| 42 |
-
auto_eval_column_dict.append(
|
| 43 |
-
auto_eval_column_dict.append(
|
| 44 |
|
| 45 |
# We use make dataclass to dynamically fill the scores from Tasks
|
| 46 |
AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
|
|
|
|
| 11 |
# These classes are for user facing column names,
|
| 12 |
# to avoid having to change them all around the code
|
| 13 |
# when a modif is needed
|
| 14 |
+
@dataclass(frozen=True)
|
| 15 |
class ColumnContent:
|
| 16 |
name: str
|
| 17 |
type: str
|
|
|
|
| 24 |
auto_eval_column_dict = []
|
| 25 |
# Init
|
| 26 |
auto_eval_column_dict.append(
|
| 27 |
+
("model_type_symbol", ColumnContent, ColumnContent("T", "markdown", True, never_hidden=True))
|
| 28 |
)
|
| 29 |
+
auto_eval_column_dict.append(("model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)))
|
| 30 |
# Scores
|
| 31 |
for idx, task in enumerate(Tasks):
|
| 32 |
# Only show overall ANLS (first task) by default
|
| 33 |
displayed_by_default = idx == 0
|
| 34 |
auto_eval_column_dict.append(
|
| 35 |
+
(task.name, ColumnContent, ColumnContent(task.value.col_name, "number", displayed_by_default))
|
| 36 |
)
|
| 37 |
# Cost/Efficiency metrics
|
| 38 |
+
auto_eval_column_dict.append(("agent_steps", ColumnContent, ColumnContent("Agent Steps", "number", True)))
|
| 39 |
+
auto_eval_column_dict.append(("cost_usd", ColumnContent, ColumnContent("Cost (USD)", "number", True)))
|
| 40 |
# Model information
|
| 41 |
+
auto_eval_column_dict.append(("model_type", ColumnContent, ColumnContent("Model Type", "str", True)))
|
| 42 |
+
auto_eval_column_dict.append(("submitted_by", ColumnContent, ColumnContent("Submitted By", "str", False)))
|
| 43 |
+
auto_eval_column_dict.append(("submission_date", ColumnContent, ColumnContent("Submission Date", "str", False)))
|
| 44 |
|
| 45 |
# We use make dataclass to dynamically fill the scores from Tasks
|
| 46 |
AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
|