Correctly Print Task Versioning (#1173)

* change version field formatting in metadata

* mention versioning in new task guide

* add instructions for changelog

* run linters

lm_eval/tasks/hellaswag/hellaswag.yaml

@@ -19,4 +19,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/hendrycks_ethics/commonsense.yaml

@@ -12,4 +12,4 @@ doc_to_choice: ['no', 'yes']
 metric_list:
   - metric: acc
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/hendrycks_ethics/deontology.yaml

@@ -5,5 +5,5 @@ doc_to_text: "Question: Would most people believe this reasonable or unreasonabl
 doc_to_target: label
 doc_to_choice: ['unreasonable', 'reasonable']
 metadata:
-  - version: 1.0
+  version: 1.0
 # TODO: implement exact-match metric for this subset

lm_eval/tasks/hendrycks_ethics/justice.yaml

@@ -6,4 +6,4 @@ dataset_name: justice
 doc_to_text: "Question: Would most people believe this reasonable or unreasonable to say? \"{{scenario}}\"\nAnswer:"
 # TODO: impl. exact match for this and deontology
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/hendrycks_ethics/utilitarianism.yaml

@@ -9,4 +9,4 @@ doc_to_choice: ['no', 'yes']
 metric_list:
   - metric: acc
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/hendrycks_ethics/utilitarianism_original_yaml

@@ -13,4 +13,4 @@
 #   - metric: acc
 # TODO: we want this to be implemented as a winograd_schema task type, actually
 # metadata:
-#   - version: 1.0
+#   version: 1.0

lm_eval/tasks/hendrycks_ethics/virtue.yaml

@@ -7,4 +7,4 @@ doc_to_text: "Sentence: {{scenario}}\nQuestion: Does the character in this sente
 doc_to_target: label
 doc_to_choice: ['no', 'yes']
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/ifeval/ifeval.yaml

@@ -26,4 +26,4 @@ metric_list:
     aggregation: !function utils.agg_inst_level_acc
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/lambada/lambada_openai.yaml

@@ -17,4 +17,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/lambada/lambada_standard.yaml

@@ -18,4 +18,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/lambada_cloze/lambada_openai_cloze.yaml

@@ -17,4 +17,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/lambada_cloze/lambada_standard_cloze.yaml

@@ -18,4 +18,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/lambada_multilingual/lambada_mt_en.yaml

@@ -17,4 +17,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/logiqa/logiqa.yaml

@@ -18,4 +18,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/logiqa2/logieval.yaml

@@ -24,4 +24,4 @@ filter_list:
         regex_pattern: "^\\s*([A-D])"
       - function: "take_first"
 metadata:
-  - version: 0.0
+  version: 0.0

lm_eval/tasks/logiqa2/logiqa2.yaml

@@ -18,4 +18,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 0.0
+  version: 0.0

lm_eval/tasks/mathqa/mathqa.yaml

@@ -19,4 +19,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/mc_taco/default.yaml

@@ -12,4 +12,4 @@ metric_list:
   - metric: acc
   - metric: f1
 metadata:
-  - version: 1.0
+  version: 1.0

lm_eval/tasks/mgsm/direct/direct_yaml

@@ -26,4 +26,4 @@ metric_list:
     ignore_case: true
     ignore_punctuation: true
 metadata:
-  - version: 0.0
+  version: 0.0

lm_eval/tasks/mgsm/en_cot/cot_yaml

@@ -28,4 +28,4 @@ filter_list:
         regex_pattern: "The answer is (\\-?[0-9\\.\\,]+)"
       - function: "take_first"
 metadata:
-  - version: 0.0
+  version: 0.0