[Format] Add config lints (#892)

configs/datasets/leval/levalgovreportsumm/leval_gov_report_summ_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_govreport_summ_infer_cfg = dict(
 )
 
 LEval_govreport_summ_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levallegalcontractqa/leval_legalcontractqa_gen_68a2ac.py

@@ -27,7 +27,7 @@ LEval_legalqa_infer_cfg = dict(
 )
 
 LEval_legalqa_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalmeetingsumm/leval_meetingsumm_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_meetingsumm_infer_cfg = dict(
 )
 
 LEval_meetingsumm_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalmultidocqa/leval_multidocqa_gen_96bf3f.py

@@ -27,7 +27,7 @@ LEval_multidocqa_infer_cfg = dict(
 )
 
 LEval_multidocqa_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalnarrativeqa/leval_narrativeqa_gen_766dd0.py

@@ -27,7 +27,7 @@ LEval_narrativeqa_infer_cfg = dict(
 )
 
 LEval_narrativeqa_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalnaturalquestion/leval_naturalquestion_gen_52c33f.py

@@ -27,7 +27,7 @@ LEval_nq_infer_cfg = dict(
 )
 
 LEval_nq_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalnewssumm/leval_newssumm_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_newssumm_infer_cfg = dict(
 )
 
 LEval_newssumm_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalpaperassistant/leval_paper_assistant_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_ps_summ_infer_cfg = dict(
 )
 
 LEval_ps_summ_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalpatentsumm/leval_patent_summ_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_patent_summ_infer_cfg = dict(
 )
 
 LEval_patent_summ_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalquality/leval_quality_gen_36a006.py

@@ -28,7 +28,7 @@ LEval_quality_infer_cfg = dict(
 )
 
 LEval_quality_eval_cfg = dict(
-    evaluator=dict(type=AccEvaluator), 
+    evaluator=dict(type=AccEvaluator),
     pred_postprocessor=dict(type=first_capital_postprocess),
     pred_role='BOT'
 )

configs/datasets/leval/levalreviewsumm/leval_review_summ_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_review_summ_infer_cfg = dict(
 )
 
 LEval_review_summ_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levalscientificqa/leval_scientificqa_gen_96bf3f.py

@@ -27,7 +27,7 @@ LEval_scientificqa_infer_cfg = dict(
 )
 
 LEval_scientificqa_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/leval/levaltopicretrieval/leval_topic_retrieval_gen_bf433f.py

@@ -28,7 +28,7 @@ LEval_tr_infer_cfg = dict(
 )
 
 LEval_tr_eval_cfg = dict(
-    evaluator=dict(type=LEvalEMEvaluator), 
+    evaluator=dict(type=LEvalEMEvaluator),
     pred_postprocessor=dict(type=general_postprocess),
     pred_role='BOT'
 )

configs/datasets/leval/levaltpo/leval_tpo_gen_36a006.py

@@ -28,7 +28,7 @@ LEval_tpo_infer_cfg = dict(
 )
 
 LEval_tpo_eval_cfg = dict(
-    evaluator=dict(type=AccEvaluator), 
+    evaluator=dict(type=AccEvaluator),
     pred_postprocessor=dict(type=first_capital_postprocess),
     pred_role='BOT'
 )

configs/datasets/leval/levaltvshowsumm/leval_tvshow_summ_gen_b03798.py

@@ -27,7 +27,7 @@ LEval_tvshow_summ_infer_cfg = dict(
 )
 
 LEval_tvshow_summ_eval_cfg = dict(
-    evaluator=dict(type=RougeEvaluator), 
+    evaluator=dict(type=RougeEvaluator),
     pred_role='BOT'
 )
 

configs/datasets/llm_compression/README.md

@@ -4,7 +4,7 @@
 
 The following introduction comes from the abstract of [Compression Represents Intelligence Linearly](https://arxiv.org/abs/2404.09937):
 
->There is a belief that learning to compress well will lead to intelligence. Recently, language modeling has been shown to be equivalent to compression, which offers a compelling rationale for the success of large language models (LLMs): the development of more advanced language models is essentially enhancing compression which facilitates intelligence. ...our findings suggest that compression efficiency, as an unsupervised metric derived from raw text corpora, serves as a reliable evaluation measure that is linearly associated with the model capabilities. We open-source our compression datasets as well as our data collection pipelines to facilitate future researchers to assess compression properly. 
+>There is a belief that learning to compress well will lead to intelligence. Recently, language modeling has been shown to be equivalent to compression, which offers a compelling rationale for the success of large language models (LLMs): the development of more advanced language models is essentially enhancing compression which facilitates intelligence. ...our findings suggest that compression efficiency, as an unsupervised metric derived from raw text corpora, serves as a reliable evaluation measure that is linearly associated with the model capabilities. We open-source our compression datasets as well as our data collection pipelines to facilitate future researchers to assess compression properly.
 
 
 ## Official Links
@@ -13,7 +13,7 @@ The following introduction comes from the abstract of [Compression Represents In
 - GitHub Repository: [llm-compression-intelligence](https://github.com/hkust-nlp/llm-compression-intelligence)
 
 
-## Overview and Usage 
+## Overview and Usage
 
 ### Dataset
 The dataset, which consists of three external corpora, can be downloaded using the following python script:
@@ -95,11 +95,11 @@ Token indices sequence length is longer than the specified maximum sequence leng
 ## Reference
 ```
 @misc{huang2024compression,
-      title={Compression Represents Intelligence Linearly}, 
+      title={Compression Represents Intelligence Linearly},
       author={Yuzhen Huang and Jinghan Zhang and Zifei Shan and Junxian He},
       year={2024},
       eprint={2404.09937},
       archivePrefix={arXiv},
       primaryClass={cs.CL}
 }
-```
\ No newline at end of file
+```

configs/datasets/llm_compression/llm_compression.py

@@ -20,10 +20,10 @@ for _name in subset_mapping.keys():
     llm_cmp_infer_cfg = dict(
         prompt_template=dict(
             type=PromptTemplate,
-            template="{content}",
+            template='{content}',
         ),
         # No in-context example, using ZeroRetriever
-        retriever=dict(type=ZeroRetriever),  
+        retriever=dict(type=ZeroRetriever),
         # Calculates cross entropy loss for each batch based on a sliding context window
         # Setting block_size=1900 and stride=512 according to the original paper
         inferencer=dict(type=SWCELossInferencer, block_size=1900, stride=512),
@@ -34,13 +34,13 @@ for _name in subset_mapping.keys():
 
     llm_compression_datasets.append(
         dict(
-            abbr=f"llm_compression-{_name}",
+            abbr=f'llm_compression-{_name}',
             type=LLMCompressionDataset,
-            path="./data/llm-compression",
+            path='./data/llm-compression',
             name=_name,
             samples=None,  # Set small samples for testing
             reader_cfg=dict(
-                input_columns=["content"],
+                input_columns=['content'],
                 output_column=None,
             ),
             infer_cfg=llm_cmp_infer_cfg,

configs/datasets/longbench/longbench.py

@@ -22,5 +22,5 @@ with read_base():
     from .longbenchlsht.longbench_lsht_gen import LongBench_lsht_datasets
     from .longbenchmulti_news.longbench_multi_news_gen import LongBench_multi_news_datasets
     from .longbenchsamsum.longbench_samsum_gen import LongBench_samsum_datasets
-    
-longbench_datasets = sum((v for k, v in locals().items() if k.endswith('_datasets')), [])
\ No newline at end of file
+
+longbench_datasets = sum((v for k, v in locals().items() if k.endswith('_datasets')), [])

configs/datasets/lveval/lveval.md

@@ -162,4 +162,4 @@ LVEval_factrecall_zh_32k                   ef3320     f1             gen     38.
 LVEval_factrecall_zh_64k                   ef3320     f1             gen     20.50
 LVEval_factrecall_zh_128k                  ef3320     f1             gen     11.00
 LVEval_factrecall_zh_256k                  ef3320     f1             gen     5.00
-```
\ No newline at end of file
+```

configs/datasets/lveval/lveval.py

@@ -34,5 +34,5 @@ with read_base():
     )
 
 LVEval_datasets = sum(
-    (v for k, v in locals().items() if k.endswith("_datasets")), []
+    (v for k, v in locals().items() if k.endswith('_datasets')), []
 )