Merge pull request #524 from WenjieDu/dev

Fix model saving issue and update docs

.gitignore

@@ -19,3 +19,8 @@ docs/_build
 # ignore specific kinds of files like all PDFs
 *.pdf
 *.ipynb
+*TODO
+
+# ignore executable and config files
+*.sh
+*.yml

README.md

@@ -77,7 +77,8 @@ algorithms, PyPOTS is going to have unified APIs together with detailed document
 algorithms as tutorials.
 
 🤗 **Please** star this repo to help others notice PyPOTS if you think it is a useful toolkit.
-**Please** kindly [cite PyPOTS](https://github.com/WenjieDu/PyPOTS#-citing-pypots) in your publications if it helps with your research.
+**Please** kindly [cite PyPOTS](https://github.com/WenjieDu/PyPOTS#-citing-pypots) in your publications if it helps with
+your research.
 This really means a lot to our open-source research. Thank you!
 
 The rest of this readme file is organized as follows:
@@ -98,9 +99,9 @@ corresponding task (note that models will be continuously updated in the future
 currently supported. Stay tuned❗️).
 
 🌟 Since **v0.2**, all neural-network models in PyPOTS has got hyperparameter-optimization support.
-This functionality is implemented with the [Microsoft NNI](https://github.com/microsoft/nni) framework. You may want to refer to our time-series
-imputation survey repo [Awesome_Imputation](https://github.com/WenjieDu/Awesome_Imputation) to see how to config and
-tune the hyperparameters.
+This functionality is implemented with the [Microsoft NNI](https://github.com/microsoft/nni) framework. You may want to
+refer to our time-series imputation survey repo [Awesome_Imputation](https://github.com/WenjieDu/Awesome_Imputation)
+to see how to config and tune the hyperparameters.
 
 🔥 Note that all models whose name with `🧑‍🔧` in the table (e.g. Transformer, iTransformer, Informer etc.) are not
 originally proposed as algorithms for POTS data in their papers, and they cannot directly accept time series with
@@ -225,7 +226,9 @@ for a guideline with more details.
 
 PyPOTS is available on both [PyPI](https://pypi.python.org/pypi/pypots)
 and [Anaconda](https://anaconda.org/conda-forge/pypots).
-You can install PyPOTS like below as well as TSDB and PyGrinder:
+You can install PyPOTS like below as well as
+[TSDB](https://github.com/WenjieDu/TSDB),[PyGrinder](https://github.com/WenjieDu/PyGrinder),
+[BenchPOTS](https://github.com/WenjieDu/BenchPOTS), and [AI4TS](https://github.com/WenjieDu/AI4TS):
 
 ``` bash
 # via pip
@@ -235,8 +238,8 @@ pip install pypots --upgrade  # update pypots to the latest version
 pip install https://github.com/WenjieDu/PyPOTS/archive/main.zip
 
 # via conda
-conda install -c conda-forge pypots  # the first time installation
-conda update  -c conda-forge pypots  # update pypots to the latest version
+conda install conda-forge::pypots  # the first time installation
+conda update  conda-forge::pypots  # update pypots to the latest version
 ```
 
 ## ❖ Usage

docs/install.rst

@@ -14,8 +14,8 @@ It is recommended to use **pip** or **conda** for PyPOTS installation as shown b
     pip install https://github.com/WenjieDu/PyPOTS/archive/main.zip
 
     # via conda
-    conda install -c conda-forge pypots  # the first time installation
-    conda update  -c conda-forge pypots  # update pypots to the latest version
+    conda install conda-forge::pypots  # the first time installation
+    conda update  conda-forge::pypots  # update pypots to the latest version
 
 
 Required Dependencies

docs/references.bib

@@ -765,8 +765,17 @@ @article{bai2018tcn
 }
 
 @article{zhan2024tefn,
-  title={Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting},
-  author={Zhan, Tianxiang and He, Yuanpeng and Li, Zhen and Deng, Yong},
-  journal={arXiv preprint arXiv:2405.06419},
-  year={2024}
-}
+title={Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting},
+author={Zhan, Tianxiang and He, Yuanpeng and Li, Zhen and Deng, Yong},
+journal={arXiv preprint arXiv:2405.06419},
+year={2024}
+}
+
+@inproceedings{jin2024timellm,
+title={Time-{LLM}: Time Series Forecasting by Reprogramming Large Language Models},
+author={Ming Jin and Shiyu Wang and Lintao Ma and Zhixuan Chu and James Y. Zhang and Xiaoming Shi and Pin-Yu Chen and Yuxuan Liang and Yuan-Fang Li and Shirui Pan and Qingsong Wen},
+booktitle={The Twelfth International Conference on Learning Representations},
+year={2024},
+url={https://openreview.net/forum?id=Unb5CVPtae}
+}
+

pypots/classification/base.py

@@ -347,8 +347,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):

pypots/classification/brits/model.py

@@ -228,7 +228,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/classification/grud/model.py

@@ -204,7 +204,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/classification/raindrop/model.py

@@ -248,7 +248,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/clustering/crli/model.py

@@ -295,7 +295,7 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
                     saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
@@ -354,7 +354,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/clustering/vader/model.py

@@ -303,8 +303,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):
@@ -367,7 +367,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/forecasting/base.py

@@ -346,8 +346,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):

pypots/forecasting/csdi/model.py

@@ -307,8 +307,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):
@@ -379,7 +379,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/autoformer/model.py

@@ -234,7 +234,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/base.py

@@ -346,8 +346,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):

pypots/imputation/brits/model.py

@@ -216,7 +216,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/crossformer/model.py

@@ -240,7 +240,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/csdi/model.py

@@ -287,8 +287,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):
@@ -363,7 +363,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/dlinear/model.py

@@ -211,7 +211,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/etsformer/model.py

@@ -234,7 +234,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/fedformer/model.py

@@ -248,7 +248,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/film/model.py

@@ -228,7 +228,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/frets/model.py

@@ -210,7 +210,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/gpvae/model.py

@@ -312,8 +312,8 @@ def _train_model(
 
                 # save the model if necessary
                 self._auto_save_model_if_necessary(
-                    confirm_saving=self.best_epoch == epoch,
-                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss}",
+                    confirm_saving=self.best_epoch == epoch and self.model_saving_strategy == "better",
+                    saving_name=f"{self.__class__.__name__}_epoch{epoch}_loss{mean_loss:.4f}",
                 )
 
                 if os.getenv("enable_tuning", False):
@@ -377,7 +377,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/grud/model.py

@@ -201,7 +201,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/imputeformer/model.py

@@ -253,7 +253,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/informer/model.py

@@ -228,7 +228,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/itransformer/model.py

@@ -255,7 +255,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/koopa/model.py

@@ -241,7 +241,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/micn/model.py

@@ -222,7 +222,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/moderntcn/model.py

@@ -252,7 +252,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,

pypots/imputation/mrnn/model.py

@@ -218,7 +218,7 @@ def fit(
         self.model.eval()  # set the model as eval status to freeze it.
 
         # Step 3: save the model if necessary
-        self._auto_save_model_if_necessary(confirm_saving=True)
+        self._auto_save_model_if_necessary(confirm_saving=self.model_saving_strategy == "best")
 
     def predict(
         self,