changes for pre-commit

lm_eval/tasks/pile/pile_opensubtitles.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_openwebtext2.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_philpapers.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_pile-cc.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_pubmed-abstracts.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_pubmed-central.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_stackexchange.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_ubuntu-irc.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_uspto.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_wikipedia.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/pile/pile_youtubesubtitles.yaml

@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: false
   - metric: bits_per_byte
     aggregation: bits_per_byte
-    higher_is_better: false
+    higher_is_better: false
\ No newline at end of file

lm_eval/tasks/piqa/piqa.yaml

-group: 
+group:
   - piqa_yaml_grp
 task: piqa_yaml
 dataset_path: piqa
@@ -9,7 +9,7 @@ validation_split: validation
 test_split: null
 template_aliases: "{% set question = goal %}{% set answer_choices = [sol1, sol2] %}{% set gold = label %}" # set the list of possible answer choices, and set what this doc's gold label idx is
 doc_to_text: "Question: {{question}}\nAnswer:"
-doc_to_target: "{{gold}}" # this will be cast to an int. 
+doc_to_target: "{{gold}}" # this will be cast to an int.
 metric_list:
   - metric: acc
     aggregation: mean

lm_eval/tasks/sciq/sciq.yaml

-group: 
+group:
   - sciq_yaml_grp
 task: sciq_yaml
 dataset_path: sciq
@@ -9,7 +9,7 @@ validation_split: validation
 test_split: test
 template_aliases: "{% set answer_choices = [distractor1, distractor2, distractor3, correct_answer] %}{% set gold = 3 %}" # set the list of possible answer choices, and set what this doc's gold label idx is
 doc_to_text: "{{support.lstrip()}}\nQuestion: {{question}}\nAnswer:"
-doc_to_target: "{{gold}}" # this will be cast to an int. 
+doc_to_target: "{{gold}}" # this will be cast to an int.
 metric_list:
   - metric: acc
     aggregation: mean
@@ -19,4 +19,4 @@ metric_list:
     higher_is_better: true
   - metric: acc_mutual_info
     aggregation: mean
-    higher_is_better: true
+    higher_is_better: true
\ No newline at end of file

lm_eval/tasks/wikitext/README.md

@@ -33,4 +33,4 @@ Homepage: https://www.salesforce.com/products/einstein/ai-research/the-wikitext-
 - [x] Is in Eval-harness v1.0 ?
 - [x] Has been checked for regression from v1.0?
 - [ ] Has been checked for equivalence with original paper methodology?
-- [ ] "Main" checked variant clearly denoted?
+- [ ] "Main" checked variant clearly denoted?
\ No newline at end of file

lm_eval/tasks/wikitext/preprocess_wikitext.py

 import re
 
+
 def wikitext_detokenizer(doc):
     string = doc["page"]
     # contractions

lm_eval/utils.py

@@ -158,7 +158,15 @@ def make_table(result_dict):
     md_writer = MarkdownTableWriter()
     latex_writer = LatexTableWriter()
     md_writer.headers = ["Task", "Version", "Filter", "Metric", "Value", "", "Stderr"]
-    latex_writer.headers = ["Task", "Version", "Filter", "Metric", "Value", "", "Stderr"]
+    latex_writer.headers = [
+        "Task",
+        "Version",
+        "Filter",
+        "Metric",
+        "Value",
+        "",
+        "Stderr",
+    ]
 
     values = []
 
@@ -166,7 +174,7 @@ def make_table(result_dict):
         version = result_dict["versions"][k]
         for (mf), v in dic.items():
             m, _, f = mf.partition(",")
-            print(m,f)
+            print(m, f)
             if m.endswith("_stderr"):
                 continue