refactored into a python package

.gitignore

@@ -7,6 +7,7 @@ __pycache__/
 *.so
 
 # Distribution / packaging
+_version.py
 .Python
 build/
 develop-eggs/

pyproject.toml

@@ -0,0 +1,47 @@
+[build-system]
+requires = ["setuptools", "setuptools_scm[toml]>7.0"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "timeserieslib"
+description = "A library for time series analysis"
+readme = "README.md"
+requires-python = ">=3.8"
+authors = [
+
+]
+keywords = ["regression", "python", "sklearn"]
+license = {text = "BSD3 License"} #??? unclear
+classifiers = [
+    "Programming Language :: Python :: 3",
+]
+dependencies = [
+    "einops>=0.4.0",
+	"matplotlib>=3.7.0",
+	"numpy>=1.23.5",
+	"pandas>=1.5.3",
+	"patool>=1.12",
+	"reformer-pytorch>=1.4.4",
+	"scikit-learn>=1.2.2",
+	"scipy>=1.10.1",
+	"sktime>=0.16.1",
+	"sympy>=1.11.1",
+	"torch>=1.7.1",
+	"tqdm>=4.64.1,"
+]
+dynamic = ["version"]
+
+[tool.setuptools]
+# packages = ["timeserieslib"]
+include-package-data = true
+
+[tool.setuptools.packages.find]
+where = ["src"]
+include = ["vfdt*"]
+namespaces = true
+
+[tool.setuptools_scm]
+write_to = "src/timeserieslib/_version.py"
+
+[tool.setuptools.dynamic]
+version = {attr = "timeserieslib._version.__version__"}

run.py

@@ -1,12 +1,12 @@
 import argparse
 import os
 import torch
-from exp.exp_long_term_forecasting import Exp_Long_Term_Forecast
-from exp.exp_imputation import Exp_Imputation
-from exp.exp_short_term_forecasting import Exp_Short_Term_Forecast
-from exp.exp_anomaly_detection import Exp_Anomaly_Detection
-from exp.exp_classification import Exp_Classification
-from utils.print_args import print_args
+from timeserieslib.exp.exp_long_term_forecasting import Exp_Long_Term_Forecast
+from timeserieslib.exp.exp_imputation import Exp_Imputation
+from timeserieslib.exp.exp_short_term_forecasting import Exp_Short_Term_Forecast
+from timeserieslib.exp.exp_anomaly_detection import Exp_Anomaly_Detection
+from timeserieslib.exp.exp_classification import Exp_Classification
+from timeserieslib.utils.print_args import print_args
 import random
 import numpy as np
 

src/timeserieslib/data_provider/__init__.py

@@ -0,0 +1 @@
+

data_provider/data_factory.py → src/timeserieslib/data_provider/data_factory.py

@@ -1,6 +1,6 @@
-from data_provider.data_loader import Dataset_ETT_hour, Dataset_ETT_minute, Dataset_Custom, Dataset_M4, PSMSegLoader, \
+from timeserieslib.data_provider.data_loader import Dataset_ETT_hour, Dataset_ETT_minute, Dataset_Custom, Dataset_M4, PSMSegLoader, \
     MSLSegLoader, SMAPSegLoader, SMDSegLoader, SWATSegLoader, UEAloader
-from data_provider.uea import collate_fn
+from timeserieslib.data_provider.uea import collate_fn
 from torch.utils.data import DataLoader
 
 data_dict = {
@@ -31,7 +31,7 @@ def data_provider(args, flag):
     if args.task_name == 'anomaly_detection':
         drop_last = False
         data_set = Data(
-            args = args,
+            args=args,
             root_path=args.root_path,
             win_size=args.seq_len,
             flag=flag,
@@ -47,7 +47,7 @@ def data_provider(args, flag):
     elif args.task_name == 'classification':
         drop_last = False
         data_set = Data(
-            args = args,
+            args=args,
             root_path=args.root_path,
             flag=flag,
         )
@@ -65,7 +65,7 @@ def data_provider(args, flag):
         if args.data == 'm4':
             drop_last = False
         data_set = Data(
-            args = args,
+            args=args,
             root_path=args.root_path,
             data_path=args.data_path,
             flag=flag,

data_provider/data_loader.py → src/timeserieslib/data_provider/data_loader.py

@@ -6,12 +6,12 @@
 import torch
 from torch.utils.data import Dataset, DataLoader
 from sklearn.preprocessing import StandardScaler
-from utils.timefeatures import time_features
-from data_provider.m4 import M4Dataset, M4Meta
-from data_provider.uea import subsample, interpolate_missing, Normalizer
+from timeserieslib.utils.timefeatures import time_features
+from timeserieslib.data_provider.m4 import M4Dataset, M4Meta
+from timeserieslib.data_provider.uea import subsample, interpolate_missing, Normalizer
 from sktime.datasets import load_from_tsfile_to_dataframe
 import warnings
-from utils.augmentation import run_augmentation_single
+from timeserieslib.utils.augmentation import run_augmentation_single
 
 warnings.filterwarnings('ignore')
 
@@ -23,7 +23,7 @@ def __init__(self, args, root_path, flag='train', size=None,
         # size [seq_len, label_len, pred_len]
         self.args = args
         # info
-        if size == None:
+        if size is None:
             self.seq_len = 24 * 4 * 4
             self.label_len = 24 * 4
             self.pred_len = 24 * 4
@@ -79,14 +79,14 @@ def __read_data__(self):
             data_stamp = df_stamp.drop(['date'], 1).values
         elif self.timeenc == 1:
             data_stamp = time_features(pd.to_datetime(df_stamp['date'].values), freq=self.freq)
-            data_stamp = data_stamp.transpose(1, 0) 
+            data_stamp = data_stamp.transpose(1, 0)
 
         self.data_x = data[border1:border2]
         self.data_y = data[border1:border2]
 
         if self.set_type == 0 and self.args.augmentation_ratio > 0:
             self.data_x, self.data_y, augmentation_tags = run_augmentation_single(self.data_x, self.data_y, self.args)
-            
+
         self.data_stamp = data_stamp
 
     def __getitem__(self, index):
@@ -116,7 +116,7 @@ def __init__(self, args, root_path, flag='train', size=None,
         # size [seq_len, label_len, pred_len]
         self.args = args
         # info
-        if size == None:
+        if size is None:
             self.seq_len = 24 * 4 * 4
             self.label_len = 24 * 4
             self.pred_len = 24 * 4
@@ -211,7 +211,7 @@ def __init__(self, args, root_path, flag='train', size=None,
         # size [seq_len, label_len, pred_len]
         self.args = args
         # info
-        if size == None:
+        if size is None:
             self.seq_len = 24 * 4 * 4
             self.label_len = 24 * 4
             self.pred_len = 24 * 4
@@ -359,7 +359,7 @@ def __getitem__(self, index):
         insample[-len(insample_window):, 0] = insample_window
         insample_mask[-len(insample_window):, 0] = 1.0
         outsample_window = sampled_timeseries[
-                           cut_point - self.label_len:min(len(sampled_timeseries), cut_point + self.pred_len)]
+            cut_point - self.label_len:min(len(sampled_timeseries), cut_point + self.pred_len)]
         outsample[:len(outsample_window), 0] = outsample_window
         outsample_mask[:len(outsample_window), 0] = 1.0
         return insample, outsample, insample_mask, outsample_mask
@@ -676,7 +676,7 @@ def load_all(self, root_path, file_list=None, flag=None):
             data_paths = list(filter(lambda x: re.search(flag, x), data_paths))
         input_paths = [p for p in data_paths if os.path.isfile(p) and p.endswith('.ts')]
         if len(input_paths) == 0:
-            pattern='*.ts'
+            pattern = '*.ts'
             raise Exception("No .ts files found using pattern: '{}'".format(pattern))
 
         all_df, labels_df = self.load_single(input_paths[0])  # a single file contains dataset
@@ -685,7 +685,7 @@ def load_all(self, root_path, file_list=None, flag=None):
 
     def load_single(self, filepath):
         df, labels = load_from_tsfile_to_dataframe(filepath, return_separate_X_and_y=True,
-                                                             replace_missing_vals_with='NaN')
+                                                   replace_missing_vals_with='NaN')
         labels = pd.Series(labels, dtype="category")
         self.class_names = labels.cat.categories
         labels_df = pd.DataFrame(labels.cat.codes,
@@ -742,7 +742,7 @@ def __getitem__(self, ind):
             batch_x = batch_x.reshape((1 * seq_len, num_columns))
 
         return self.instance_norm(torch.from_numpy(batch_x)), \
-               torch.from_numpy(labels)
+            torch.from_numpy(labels)
 
     def __len__(self):
         return len(self.all_IDs)

exp/exp_anomaly_detection.py → src/timeserieslib/exp/exp_anomaly_detection.py

@@ -1,18 +1,18 @@
-from data_provider.data_factory import data_provider
-from exp.exp_basic import Exp_Basic
-from utils.tools import EarlyStopping, adjust_learning_rate, adjustment
+import numpy as np
+import warnings
+import time
+import os
+from torch import optim
+import torch.nn as nn
+import torch
+from timeserieslib.data_provider.data_factory import data_provider
+from timeserieslib.exp.exp_basic import Exp_Basic
+from timeserieslib.utils.tools import EarlyStopping, adjust_learning_rate, adjustment
 from sklearn.metrics import precision_recall_fscore_support
 from sklearn.metrics import accuracy_score
 import torch.multiprocessing
 
 torch.multiprocessing.set_sharing_strategy('file_system')
-import torch
-import torch.nn as nn
-from torch import optim
-import os
-import time
-import warnings
-import numpy as np
 
 warnings.filterwarnings('ignore')
 

exp/exp_basic.py → src/timeserieslib/exp/exp_basic.py

@@ -1,8 +1,12 @@
 import os
 import torch
-from models import Autoformer, Transformer, TimesNet, Nonstationary_Transformer, DLinear, FEDformer, \
+from timeserieslib.models import Autoformer, Transformer, TimesNet, Nonstationary_Transformer, DLinear, FEDformer, \
     Informer, LightTS, Reformer, ETSformer, Pyraformer, PatchTST, MICN, Crossformer, FiLM, iTransformer, \
-    Koopa, TiDE, FreTS, TimeMixer, TSMixer, SegRNN, MambaSimple, Mamba, TemporalFusionTransformer
+    Koopa, TiDE, FreTS, TimeMixer, TSMixer, SegRNN, MambaSimple, TemporalFusionTransformer
+try:
+    from timeserieslib.models import Mamba  # Mamba does not work without cuda
+except ModuleNotFoundError:
+    Mamba = None
 
 
 class Exp_Basic(object):
@@ -29,12 +33,15 @@ def __init__(self, args):
             'TiDE': TiDE,
             'FreTS': FreTS,
             'MambaSimple': MambaSimple,
-            'Mamba': Mamba,
             'TimeMixer': TimeMixer,
             'TSMixer': TSMixer,
             'SegRNN': SegRNN,
             'TemporalFusionTransformer': TemporalFusionTransformer
         }
+        if Mamba is not None:
+            self.model_dict['Mamba'] = Mamba
+        elif self.args.model == 'Mamba':
+            raise ModuleNotFoundError('Mamba is not installed')
         self.device = self._acquire_device()
         self.model = self._build_model().to(self.device)
 

exp/exp_classification.py → src/timeserieslib/exp/exp_classification.py

@@ -1,6 +1,6 @@
-from data_provider.data_factory import data_provider
-from exp.exp_basic import Exp_Basic
-from utils.tools import EarlyStopping, adjust_learning_rate, cal_accuracy
+from timeserieslib.data_provider.data_factory import data_provider
+from timeserieslib.exp.exp_basic import Exp_Basic
+from timeserieslib.utils.tools import EarlyStopping, adjust_learning_rate, cal_accuracy
 import torch
 import torch.nn as nn
 from torch import optim
@@ -183,8 +183,8 @@ def test(self, setting, test=0):
             os.makedirs(folder_path)
 
         print('accuracy:{}'.format(accuracy))
-        file_name='result_classification.txt'
-        f = open(os.path.join(folder_path,file_name), 'a')
+        file_name = 'result_classification.txt'
+        f = open(os.path.join(folder_path, file_name), 'a')
         f.write(setting + "  \n")
         f.write('accuracy:{}'.format(accuracy))
         f.write('\n')

exp/exp_imputation.py → src/timeserieslib/exp/exp_imputation.py

@@ -1,7 +1,7 @@
-from data_provider.data_factory import data_provider
-from exp.exp_basic import Exp_Basic
-from utils.tools import EarlyStopping, adjust_learning_rate, visual
-from utils.metrics import metric
+from timeserieslib.data_provider.data_factory import data_provider
+from timeserieslib.exp.exp_basic import Exp_Basic
+from timeserieslib.utils.tools import EarlyStopping, adjust_learning_rate, visual
+from timeserieslib.utils.metrics import metric
 import torch
 import torch.nn as nn
 from torch import optim
@@ -186,7 +186,7 @@ def test(self, setting, test=0):
                 f_dim = -1 if self.args.features == 'MS' else 0
                 outputs = outputs[:, :, f_dim:]
 
-                # add support for MS 
+                # add support for MS
                 batch_x = batch_x[:, :, f_dim:]
                 mask = mask[:, :, f_dim:]
 
@@ -200,7 +200,7 @@ def test(self, setting, test=0):
                 if i % 20 == 0:
                     filled = true[0, :, -1].copy()
                     filled = filled * mask[0, :, -1].detach().cpu().numpy() + \
-                             pred[0, :, -1] * (1 - mask[0, :, -1].detach().cpu().numpy())
+                        pred[0, :, -1] * (1 - mask[0, :, -1].detach().cpu().numpy())
                     visual(true[0, :, -1], filled, os.path.join(folder_path, str(i) + '.pdf'))
 
         preds = np.concatenate(preds, 0)

exp/exp_long_term_forecasting.py → src/timeserieslib/exp/exp_long_term_forecasting.py

@@ -1,16 +1,16 @@
-from data_provider.data_factory import data_provider
-from exp.exp_basic import Exp_Basic
-from utils.tools import EarlyStopping, adjust_learning_rate, visual
-from utils.metrics import metric
+from timeserieslib.data_provider.data_factory import data_provider
+from timeserieslib.exp.exp_basic import Exp_Basic
+from timeserieslib.utils.tools import EarlyStopping, adjust_learning_rate, visual
+from timeserieslib.utils.metrics import metric
 import torch
 import torch.nn as nn
 from torch import optim
 import os
 import time
 import warnings
 import numpy as np
-from utils.dtw_metric import dtw,accelerated_dtw
-from utils.augmentation import run_augmentation,run_augmentation_single
+from timeserieslib.utils.dtw_metric import dtw, accelerated_dtw
+from timeserieslib.utils.augmentation import run_augmentation, run_augmentation_single
 
 warnings.filterwarnings('ignore')
 
@@ -223,7 +223,7 @@ def test(self, setting, test=0):
                     shape = outputs.shape
                     outputs = test_data.inverse_transform(outputs.squeeze(0)).reshape(shape)
                     batch_y = test_data.inverse_transform(batch_y.squeeze(0)).reshape(shape)
-        
+
                 outputs = outputs[:, :, f_dim:]
                 batch_y = batch_y[:, :, f_dim:]
 
@@ -252,22 +252,21 @@ def test(self, setting, test=0):
         folder_path = './results/' + setting + '/'
         if not os.path.exists(folder_path):
             os.makedirs(folder_path)
-        
+
         # dtw calculation
         if self.args.use_dtw:
             dtw_list = []
-            manhattan_distance = lambda x, y: np.abs(x - y)
+            def manhattan_distance(x, y): return np.abs(x - y)
             for i in range(preds.shape[0]):
-                x = preds[i].reshape(-1,1)
-                y = trues[i].reshape(-1,1)
+                x = preds[i].reshape(-1, 1)
+                y = trues[i].reshape(-1, 1)
                 if i % 100 == 0:
                     print("calculating dtw iter:", i)
                 d, _, _, _ = accelerated_dtw(x, y, dist=manhattan_distance)
                 dtw_list.append(d)
             dtw = np.array(dtw_list).mean()
         else:
             dtw = -999
-
 
         mae, mse, rmse, mape, mspe = metric(preds, trues)
         print('mse:{}, mae:{}, dtw:{}'.format(mse, mae, dtw))

exp/exp_short_term_forecasting.py → src/timeserieslib/exp/exp_short_term_forecasting.py

@@ -1,9 +1,9 @@
-from data_provider.data_factory import data_provider
-from data_provider.m4 import M4Meta
-from exp.exp_basic import Exp_Basic
-from utils.tools import EarlyStopping, adjust_learning_rate, visual
-from utils.losses import mape_loss, mase_loss, smape_loss
-from utils.m4_summary import M4Summary
+from timeserieslib.data_provider.data_factory import data_provider
+from timeserieslib.data_provider.m4 import M4Meta
+from timeserieslib.exp.exp_basic import Exp_Basic
+from timeserieslib.utils.tools import EarlyStopping, adjust_learning_rate, visual
+from timeserieslib.utils.losses import mape_loss, mase_loss, smape_loss
+from timeserieslib.utils.m4_summary import M4Summary
 import torch
 import torch.nn as nn
 from torch import optim