Merge pull request #166 from jon-tow/remove-unused-imports

Remove unused imports and format imports

lm_eval/tasks/superglue.py

@@ -4,11 +4,11 @@ To-do:
     - ReCoRD
 """
 import numpy as np
+import sklearn
+import transformers.data.metrics.squad_metrics as squad_metrics
 from . common import HFTask, yesno
 from lm_eval.base import rf
 from ..metrics import mean, acc_all, metric_max_over_ground_truths
-import sklearn
-import transformers.data.metrics.squad_metrics as squad_metrics
 from ..utils import general_detokenize
 
 

lm_eval/tasks/translation.py

-import abc
-import json
-import random
-import os
-from collections import Iterable
-from pprint import pprint
-
 import pycountry
+from pprint import pprint
 from sacrebleu import sacrebleu
-import logging
-
 from lm_eval import metrics
 from lm_eval.base import Task, rf
 

lm_eval/tasks/triviaqa.py

 import os
 import json
-import random
 from lm_eval.base import Task, rf
 from ..metrics import mean
 from ..utils import sh
 
+
 class TriviaQA(Task):
     def download(self):
         if not os.path.exists('data/triviaqa'):

lm_eval/tasks/unscramble.py

 import gzip
 import json
-import random
 import shutil
 from pathlib import Path
 from best_download import download_file

lm_eval/tasks/wikitext.py

-import numpy as np
-from scipy.stats import pearsonr, spearmanr
-from sklearn.metrics import f1_score, matthews_corrcoef
-from tqdm import auto as tqdm_lib
-from . common import NLP_TASK, simple_accuracy_metric, yesno
+from . common import HFTask
 
-class WikiText103(NLP_TASK):
+
+class WikiText103(HFTask):
     NLP_PATH = "wikitext"
     NLP_NAME = "wikitext-103-raw-v1"
 
@@ -66,7 +63,7 @@ class WikiText103(NLP_TASK):
         raise NotImplementedError('Evaluation not implemented')
 
 
-class WikiText2(NLP_TASK):
+class WikiText2(HFTask):
     NLP_PATH = "wikitext"
     NLP_NAME = "wikitext-2-raw-v1"
 

main.py

@@ -2,8 +2,6 @@ import argparse
 import json
 import numpy as np
 import random
-import itertools
-import collections
 import logging
 
 from lm_eval import models, tasks, evaluator, base

scripts/cost_estimate.py

-import argparse
-import json
-import numpy as np
 import random
-import itertools
-import collections
-import logging
-
-from lm_eval import models, tasks, evaluator, base
-import random
-from lm_eval.base import LM
 import transformers
+from lm_eval import tasks, evaluator
+from lm_eval.base import LM
 
 
 class DryrunLM(LM):

scripts/fewshot_description_experiment.py

-import argparse
 import json
 import numpy as np
 import random
-import itertools
-import collections
 import logging
-
 from lm_eval import models, tasks, evaluator, base
 
 logging.getLogger("openai").setLevel(logging.WARNING)

scripts/write_out.py

@@ -2,7 +2,6 @@ import argparse
 import numpy as np
 import os
 import random
-
 from lm_eval import tasks
 from lm_eval.utils import join_iters
 

tests/test_models.py

 import lm_eval.models as models
-import lm_eval.base as base
+
 
 def test_gpt2():
     gpt2 = models.get_model('gpt2').create_from_arg_string("device=cpu")

tests/test_tasks.py

 import lm_eval.tasks as tasks
 import lm_eval.base as base
-from itertools import islice
 import pytest
+from itertools import islice
 
 
 @pytest.mark.parametrize("taskname,Task", tasks.TASK_REGISTRY.items())