Fix initial lint errors #27

- remove unused imports
- fix trailing white spaces and indentation
- update workflow to ignore version import

.github/workflows/main.yml

@@ -18,4 +18,4 @@ jobs:
           pip install flake8
       - name: Lint with flake8
         run: | 
-          flake8 --max-line-length 80 dora_exp_pipeline
+          flake8 --max-line-length 80 --per-file-ignores="__init__.py:F401" dora_exp_pipeline

dora_exp_pipeline/dora_config.py

@@ -67,19 +67,19 @@ def log_configs(self, config_file):
     def verify_config_parameters(self):
         # Verify `data_type` field
         if not isinstance(self.data_type, str):
-            raise RuntimeError(f'data_type field must be a string')
+            raise RuntimeError('data_type field must be a string')
 
         # Verify `data_to_fit`
         if not isinstance(self.data_to_fit, str):
-            raise RuntimeError(f'data_to_fit field must be a string')
+            raise RuntimeError('data_to_fit field must be a string')
 
         if not os.path.exists(self.data_to_fit):
             raise RuntimeError(f'data_to_fit not found: '
                                f'{os.path.abspath(self.data_to_fit)}')
 
         # Verify `data_to_score`
         if not isinstance(self.data_to_score, str):
-            raise RuntimeError(f'data_to_score field must be a string')
+            raise RuntimeError('data_to_score field must be a string')
 
         if (len(self.data_to_score) != 0 and
                 not os.path.exists(self.data_to_score)):
@@ -88,15 +88,15 @@ def verify_config_parameters(self):
 
         # Verify `features`
         if not isinstance(self.features, dict):
-            raise RuntimeError(f'features field must be a dictionary')
+            raise RuntimeError('features field must be a dictionary')
 
         # Verify `outlier_detection`
         if not isinstance(self.outlier_detection, dict):
-            raise RuntimeError(f'outlier_detection field must be a dictionary')
+            raise RuntimeError('outlier_detection field must be a dictionary')
 
         # Verify `results`
         if not isinstance(self.results, dict):
-            raise RuntimeError(f'results field must be a dictionary')
+            raise RuntimeError('results field must be a dictionary')
 
 
 # Copyright (c) 2021 California Institute of Technology ("Caltech").

dora_exp_pipeline/dora_data_loader.py

@@ -173,22 +173,22 @@ def _load(self, dir_path: str) -> dict:
         data_dict.setdefault('id', [])
         data_dict.setdefault('data', [])
 
-        # TODO: add support for other data types 
+        # TODO: add support for other data types
         # (e.g., .h5 dataframes, .npy)
         if dir_path.endswith('.csv'):
             # Load the csv data
             with open(dir_path, 'r') as csv_file:
                 csv_reader = csv.reader(csv_file, quoting=csv.QUOTE_NONNUMERIC)
                 # Read each row into the data_dict
                 for row in csv_reader:
-                    # Assumes the first column is the ID 
+                    # Assumes the first column is the ID
                     # and all other columns are time steps
                     data_dict['id'].append(row[0])
                     data_dict['data'].append(np.array(row[1:]))
         else:
             raise RuntimeError(f'File extension not supported. '
                                f'Valid file extensions: '
-                               f'{file_types}') 
+                               f'{file_types}')
 
         return data_dict
 

dora_exp_pipeline/dora_exp.py

@@ -12,7 +12,6 @@
 import sys
 from dora_exp_pipeline.dora_config import DoraConfig
 from dora_exp_pipeline.dora_data_loader import get_data_loader_by_name
-from dora_exp_pipeline.dora_feature import z_score_normalize
 from dora_exp_pipeline.outlier_detection import register_od_alg
 # from src.demud_outlier_detection import DEMUDOutlierDetection
 from dora_exp_pipeline.iforest_outlier_detection import IForestOutlierDetection

dora_exp_pipeline/dora_feature.py

@@ -3,10 +3,7 @@
 #
 # Steven Lu, June 29, 2020
 
-import os
 import numpy as np
-from scipy.stats import skew
-from scipy.stats import kurtosis
 from skimage import transform
 from six import add_metaclass
 from abc import ABCMeta, abstractmethod

dora_exp_pipeline/lrx_outlier_detection.py

@@ -11,9 +11,7 @@
 import sys
 import numpy as np
 from dora_exp_pipeline.outlier_detection import OutlierDetection
-from dora_exp_pipeline.util import load_images
 from dora_exp_pipeline.util import DEFAULT_DATA_DIR
-from dora_exp_pipeline.util import get_image_file_list
 
 
 class LocalRXOutlierDetection(OutlierDetection):
@@ -36,7 +34,7 @@ def _rank_internal(self, files, rank_data, prior_data, config, seed,
         features_1d = config.features['one_dimensional']
         features_keys = features_1d.keys()
         if 'flattened_pixel_values' not in features_keys or \
-                        len(features_keys) > 1:
+                len(features_keys) > 1:
             raise RuntimeError(
                 'LRX must be used with `flattened_pixel_values` feature alone.'
             )
@@ -91,7 +89,7 @@ def _rank_targets(self, data_test, files, inner_window, outer_window, bands,
 
 # Local RX (LRX)
 def get_LRX_scores(images, w_inner, w_outer, bands):
-    # Images has shape N x M where N is number of images and 
+    # Images has shape N x M where N is number of images and
     # M is flattened image dimension. Divide by bands to get
     # image dimensions.
     rows, cols = images.shape
@@ -112,7 +110,7 @@ def get_LRX_scores(images, w_inner, w_outer, bands):
             for j in range(s, im.shape[1]-s):
                 scores[idx, i, j] = lrx(im[i - s: i + s + 1, j - s: j + s + 1],
                                         w_inner)
-    return np.mean(scores[:,s:-s,s:-s], axis=(1, 2)), scores
+    return np.mean(scores[:, s:-s, s:-s], axis=(1, 2)), scores
 
 
 def lrx(patch, w_in):
@@ -183,7 +181,7 @@ def main():
                         help='size of inner window (default 3)')
     parser.add_argument('-u', '--outer_window', type=int, default=5,
                         help='size of outer window (default 5)')
-    parser.add_argument('-b', '--bands', type=int, default=1, 
+    parser.add_argument('-b', '--bands', type=int, default=1,
                         help='number of bands in input images (default 1)')
     parser.add_argument('--seed', type=int, default=1234,
                         help='Integer used to seed the random generator. This '

dora_exp_pipeline/random_outlier_detection.py

@@ -9,10 +9,10 @@
 # Steven Lu, May 13, 2020, refactored the code to extract common functionalities
 #                          out to util.py.
 
-import sys
 import numpy as np
 from dora_exp_pipeline.outlier_detection import OutlierDetection
 
+
 class RandomOutlierDetection(OutlierDetection):
     def __init__(self):
         super(RandomOutlierDetection, self).__init__('random')
@@ -23,8 +23,8 @@ def _random(self, train, test, seed):
         random_state = np.random.RandomState(seed)
         random_state.shuffle(indices)
 
-        # This interprets the indices as the scores so when 
-        # the scores are sorted later they will have the 
+        # This interprets the indices as the scores so when
+        # the scores are sorted later they will have the
         # random order.
         return indices