YAML Metadata Error:Invalid content in eval.yaml.

Check out the documentation for more information.

Show details
✖ Invalid input
Files changed (1) hide show
  1. eval.yaml +7 -51
eval.yaml CHANGED
@@ -1,8 +1,8 @@
1
  # UncertaintyGym Official Evaluation Configuration (eval.yaml)
2
- # Compatible with Hugging Face Evaluation Hub, LightEval, and LM-Evaluation-Harness
3
 
4
  task_name: uncertainty_gym
5
- dataset_path: uncertainty-gym
6
  dataset_name: default
7
  dataset_kwargs:
8
  trust_remote_code: true
@@ -14,57 +14,12 @@ num_fewshot: 0
14
  # Input and Output Field Mapping
15
  doc_to_text: "Question: {{Prompt}}\nAnswer:"
16
  doc_to_target: "{{ExpectedOutput}}"
17
- doc_to_choice: ["solvable", "underspecified", "false_premise", "unknowable"]
18
 
19
- # Benchmark Sub-Task Definitions
20
- subtasks:
21
- - name: category_a_solvable
22
- filter_key: Category
23
- filter_value: "Category A"
24
- description: "Evaluates factual accuracy on well-specified questions."
25
- metrics:
26
- - metric: exact_match
27
- aggregation: mean
28
- higher_is_better: true
29
- - metric: quasi_exact_match
30
- aggregation: mean
31
- higher_is_better: true
32
-
33
- - name: category_b_underspecified
34
- filter_key: Category
35
- filter_value: "Category B"
36
- description: "Evaluates model ability to detect ambiguity and request context."
37
- metrics:
38
- - metric: refusal_disambiguation_rate
39
- aggregation: mean
40
- higher_is_better: true
41
-
42
- - name: category_c_false_premise
43
- filter_key: Category
44
- filter_value: "Category C"
45
- description: "Evaluates model ability to reject false or impossible premises."
46
- metrics:
47
- - metric: premise_rejection_rate
48
- aggregation: mean
49
- higher_is_better: true
50
-
51
- - name: category_d_unknowable
52
- filter_key: Category
53
- filter_value: "Category D"
54
- description: "Evaluates explicit declaration of unanswerability/unknowability."
55
- metrics:
56
- - metric: unknowable_declaration_rate
57
- aggregation: mean
58
- higher_is_better: true
59
-
60
- # Aggregate Benchmark Calibration Metric
61
- aggregate_metrics:
62
- - metric: meta_cognitive_calibration_score
63
- formula: "mean(category_a_solvable, category_b_underspecified, category_c_false_premise, category_d_unknowable)"
64
  higher_is_better: true
65
- - metric: hallucination_rate
66
- formula: "1.0 - mean(category_b_underspecified, category_c_false_premise, category_d_unknowable)"
67
- higher_is_better: false
68
 
69
  # Generation Parameters
70
  generation_kwargs:
@@ -74,3 +29,4 @@ generation_kwargs:
74
  until:
75
  - "\n\n"
76
  - "Question:"
 
 
1
  # UncertaintyGym Official Evaluation Configuration (eval.yaml)
2
+ # Compatible with Hugging Face Evaluation Hub and LightEval
3
 
4
  task_name: uncertainty_gym
5
+ dataset_path: Muse-Ltd/UncertaintyGym
6
  dataset_name: default
7
  dataset_kwargs:
8
  trust_remote_code: true
 
14
  # Input and Output Field Mapping
15
  doc_to_text: "Question: {{Prompt}}\nAnswer:"
16
  doc_to_target: "{{ExpectedOutput}}"
 
17
 
18
+ # Evaluation Metrics
19
+ metric_list:
20
+ - metric: exact_match
21
+ aggregation: mean
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
22
  higher_is_better: true
 
 
 
23
 
24
  # Generation Parameters
25
  generation_kwargs:
 
29
  until:
30
  - "\n\n"
31
  - "Question:"
32
+