Merge pull request #588 from WenjieDu/dev

Remove data prerprocessing pipelines and update docs

Author: WenjieDu

docs/index.rst

@@ -1,6 +1,4 @@
-.. PyPOTS documentation index page
-Created by Wenjie Du <[email protected]>
-License: BSD-3-Clause
+.. PyPOTS documentation index page created by Wenjie Du <[email protected]>
 
 Welcome to PyPOTS docs!
 ===============================

docs/references.bib

@@ -800,4 +800,17 @@ @inproceedings{zhou2023gpt4ts
 booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
 year={2023},
 url={https://openreview.net/forum?id=gMS6FVZvmF}
+}
+
+@article{yue2022ts2vec,
+title={TS2Vec: Towards Universal Representation of Time Series},
+volume={36},
+url={https://ojs.aaai.org/index.php/AAAI/article/view/20881},
+DOI={10.1609/aaai.v36i8.20881},
+number={8},
+journal={Proceedings of the AAAI Conference on Artificial Intelligence},
+author={Yue, Zhihan and Wang, Yujing and Duan, Juanyong and Yang, Tianmeng and Huang, Congrui and Tong, Yunhai and Xu, Bixiong},
+year={2022},
+month={Jun.},
+pages={8980-8987}
 }

pypots/data/__init__.py

@@ -6,7 +6,6 @@
 # License: BSD-3-Clause
 
 from .dataset import BaseDataset, SUPPORTED_DATASET_FILE_FORMATS
-from .generating import gene_random_walk
 from .saving import (
     save_dict_into_h5,
     load_dict_from_h5,
@@ -18,15 +17,11 @@
     sliding_window,
     inverse_sliding_window,
 )
-from .load_specific_datasets import load_specific_dataset
 
 __all__ = [
     # base dataset classes
     "BaseDataset",
     "SUPPORTED_DATASET_FILE_FORMATS",
-    # dataset generation functions
-    "gene_random_walk",
-    "load_specific_dataset",
     # utils
     "parse_delta",
     "sliding_window",

pypots/data/generating.py

@@ -14,7 +14,7 @@
 from ...data.dataset import BaseDataset
 
 
-class DatasetForForecastingCSDI(BaseDataset):
+class DatasetForCSDI(BaseDataset):
     """Dataset for CSDI forecasting model."""
 
     def __init__(
@@ -194,7 +194,7 @@ def _fetch_data_from_file(self, idx: int) -> Iterable:
         return sample
 
 
-class TestDatasetForForecastingCSDI(DatasetForForecastingCSDI):
+class TestDatasetForCSDI(DatasetForCSDI):
     """Test dataset for CSDI forecasting model."""
 
     def __init__(

pypots/data/load_specific_datasets.py

@@ -19,7 +19,7 @@
     pass
 
 from .core import _CSDI
-from .data import DatasetForForecastingCSDI, TestDatasetForForecastingCSDI
+from .data import DatasetForCSDI, TestDatasetForCSDI
 from ..base import BaseNNForecaster
 from ...data.checking import key_in_data_set
 from ...optim.adam import Adam
@@ -357,7 +357,7 @@ def fit(
         n_sampling_times: int = 1,
     ) -> None:
         # Step 1: wrap the input data with classes Dataset and DataLoader
-        training_set = DatasetForForecastingCSDI(
+        training_set = DatasetForCSDI(
             train_set,
             file_type=file_type,
         )
@@ -371,7 +371,7 @@ def fit(
         if val_set is not None:
             if not key_in_data_set("X_pred", val_set):
                 raise ValueError("val_set must contain 'X_pred' for model validation.")
-            val_set = DatasetForForecastingCSDI(
+            val_set = DatasetForCSDI(
                 val_set,
                 file_type=file_type,
             )
@@ -426,7 +426,7 @@ def predict(
 
         # Step 1: wrap the input data with classes Dataset and DataLoader
         self.model.eval()  # set the model as eval status to freeze it.
-        test_set = TestDatasetForForecastingCSDI(
+        test_set = TestDatasetForCSDI(
             test_set,
             self.n_pred_steps,
             self.n_pred_features,

pypots/forecasting/csdi/data.py

@@ -34,7 +34,7 @@ class Crossformer(BaseNNImputer):
         The number of features in the time-series data sample.
 
     n_layers :
-        The number of layers in the 1st and 2nd DMSA blocks in the SAITS model.
+        The number of layers in the Crossformer model.
 
     d_model :
         The dimension of the model.

pypots/forecasting/csdi/model.py

@@ -22,8 +22,8 @@ def __init__(
         self,
         n_steps,
         n_features,
-        n_e_layers,
-        n_d_layers,
+        n_encoder_layers,
+        n_decoder_layers,
         d_model,
         n_heads,
         d_ffn,
@@ -57,7 +57,7 @@ def __init__(
                     dropout=dropout,
                     activation=activation,
                 )
-                for _ in range(n_e_layers)
+                for _ in range(n_encoder_layers)
             ]
         )
         # Decoder
@@ -70,7 +70,7 @@ def __init__(
                     n_steps,
                     dropout=dropout,
                 )
-                for _ in range(n_d_layers)
+                for _ in range(n_decoder_layers)
             ],
         )
 

pypots/imputation/crossformer/model.py

@@ -33,10 +33,10 @@ class ETSformer(BaseNNImputer):
     n_features :
         The number of features in the time-series data sample.
 
-    n_e_layers :
+    n_encoder_layers :
         The number of layers in the ETSformer encoder.
 
-    n_d_layers :
+    n_decoder_layers :
         The number of layers in the ETSformer decoder.
 
     d_model :
@@ -115,8 +115,8 @@ def __init__(
         self,
         n_steps: int,
         n_features: int,
-        n_e_layers: int,
-        n_d_layers: int,
+        n_encoder_layers: int,
+        n_decoder_layers: int,
         d_model: int,
         n_heads: int,
         d_ffn: int,
@@ -152,8 +152,8 @@ def __init__(
         self.n_features = n_features
         # model hype-parameters
         self.n_heads = n_heads
-        self.n_e_layers = n_e_layers
-        self.n_d_layers = n_d_layers
+        self.n_encoder_layers = n_encoder_layers
+        self.n_decoder_layers = n_decoder_layers
         self.d_model = d_model
         self.d_ffn = d_ffn
         self.dropout = dropout
@@ -165,8 +165,8 @@ def __init__(
         self.model = _ETSformer(
             self.n_steps,
             self.n_features,
-            self.n_e_layers,
-            self.n_d_layers,
+            self.n_encoder_layers,
+            self.n_decoder_layers,
             self.d_model,
             self.n_heads,
             self.d_ffn,

pypots/imputation/etsformer/core.py

@@ -36,7 +36,7 @@ class ImputeFormer(BaseNNImputer):
         The number of features in the time-series data sample.
 
     n_layers :
-        The number of layers in the 1st and 2nd DMSA blocks in the SAITS model.
+        The number of layers in the ImputeFormer model.
 
     d_input_embed :
         The dimension of the input embedding.

pypots/imputation/etsformer/model.py

@@ -36,7 +36,7 @@ class iTransformer(BaseNNImputer):
         The number of features in the time-series data sample.
 
     n_layers :
-        The number of layers in the 1st and 2nd DMSA blocks in the SAITS model.
+        The number of layers in the iTransformer model.
 
     d_model :
         The dimension of the model's backbone.

pypots/imputation/imputeformer/model.py

@@ -37,7 +37,7 @@ class Transformer(BaseNNImputer):
         The number of features in the time-series data sample.
 
     n_layers :
-        The number of layers in the 1st and 2nd DMSA blocks in the SAITS model.
+        The number of layers in the Transformer model.
 
     d_model :
         The dimension of the model's backbone.