Informer代码详细注释版

# Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting (AAAI'21 Best Paper)     This is the origin Pytorch implementation of Informer in the following paper: [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://arxiv.org/abs/2012.07436). Special thanks to `Jieqi Peng`@[cookieminions](https://github.com/cookieminions) for building this repo. :triangular_flag_on_post:**News**(Mar 25, 2021): We update all experiment [results](#resultslink) with hyperparameter settings. :triangular_flag_on_post:**News**(Feb 22, 2021): We provide [Colab Examples](#colablink) for friendly usage. :triangular_flag_on_post:**News**(Feb 8, 2021): Our Informer paper has been awarded [AAAI'21 Best Paper \[Official\]](https://aaai.org/Conferences/AAAI-21/aaai-outstanding-and-distinguished-papers/)[\[Beihang\]](http://scse.buaa.edu.cn/info/1097/7443.htm)[\[Rutgers\]](https://www.business.rutgers.edu/news/hui-xiong-and-research-colleagues-receive-aaai-best-paper-award)! We will continue this line of research and update on this repo. Please star this repo and [cite](#citelink) our paper if you find our work is helpful for you. <p align="center"> <img src=".\img\informer.png" height = "360" alt="" align=center /> <br><br> <b>Figure 1.</b> The architecture of Informer. </p> ## ProbSparse Attention The self-attention scores form a long-tail distribution, where the "active" queries lie in the "head" scores and "lazy" queries lie in the "tail" area. We designed the ProbSparse Attention to select the "active" queries rather than the "lazy" queries. The ProbSparse Attention with Top-u queries forms a sparse Transformer by the probability distribution. `Why not use Top-u keys?` The self-attention layer's output is the re-represent of input. It is formulated as a weighted combination of values w.r.t. the score of dot-product pairs. The top queries with full keys encourage a complete re-represent of leading components in the input, and it is equivalent to selecting the "head" scores among all the dot-product pairs. If we choose Top-u keys, the full keys just preserve the trivial sum of values within the "long tail" scores but wreck the leading components' re-represent. <p align="center"> <img src=".\img\probsparse_intro.png" height = "320" alt="" align=center /> <br><br> <b>Figure 2.</b> The illustration of ProbSparse Attention. </p> ## Requirements - Python 3.6 - matplotlib == 3.1.1 - numpy == 1.19.4 - pandas == 0.25.1 - scikit_learn == 0.21.3 - torch == 1.8.0 Dependencies can be installed using the following command: ```bash pip install -r requirements.txt ``` ## Data The ETT dataset used in the paper can be downloaded in the repo [ETDataset](https://github.com/zhouhaoyi/ETDataset). The required data files should be put into `data/ETT/` folder. A demo slice of the ETT data is illustrated in the following figure. Note that the input of each dataset is zero-mean normalized in this implementation. <p align="center"> <img src="./img/data.png" height = "168" alt="" align=center /> <br><br> <b>Figure 3.</b> An example of the ETT data. </p> The ECL data and Weather data can be downloaded here. - [Google Drive](https://drive.google.com/drive/folders/1ohGYWWohJlOlb2gsGTeEq3Wii2egnEPR?usp=sharing) - [BaiduPan](https://pan.baidu.com/s/1wyaGUisUICYHnfkZzWCwyA), password: 6gan ## Reproducibility To easily reproduce the results you can follow the next steps: 1. Initialize the docker image using: `make init`. 2. Download the datasets using: `make dataset`. 3. Run each script in `scripts/` using `make run_module module="bash ETTh1.sh"` for each script. 4. Alternatively, run all the scripts at once: ``` for file in `ls scripts`; do make run_module module="bash scripts/$script"; done ``` ## Usage <span id="colablink">Colab Examples:</span> We provide google colabs to help reproduce and customize our repo, which includes `experiments(train and test)`, `prediction`, `visualization` and `custom data`. [](https://colab.research.google.com/drive/1_X7O2BkFLvqyCdZzDZvV2MB0aAvYALLC) Commands for training and testing the model with *ProbSparse* self-attention on Dataset ETTh1, ETTh2 and ETTm1 respectively: ```bash # ETTh1 python -u main_informer.py --model informer --data ETTh1 --attn prob --freq h # ETTh2 python -u main_informer.py --model informer --data ETTh2 --attn prob --freq h # ETTm1 python -u main_informer.py --model informer --data ETTm1 --attn prob --freq t ``` More parameter information please refer to `main_informer.py`. We provide a more detailed and complete command description for training and testing the model: ```python python -u main_informer.py --model <model> --data <data> --root_path <root_path> --data_path <data_path> --features <features> --target <target> --freq <freq> --checkpoints <checkpoints> --seq_len <seq_len> --label_len <label_len> --pred_len <pred_len> --enc_in <enc_in> --dec_in <dec_in> --c_out <c_out> --d_model <d_model> --n_heads <n_heads> --e_layers <e_layers> --d_layers <d_layers> --s_layers <s_layers> --d_ff <d_ff> --factor <factor> --padding <padding> --distil --dropout <dropout> --attn <attn> --embed <embed> --activation <activation> --output_attention --do_predict --mix --cols <cols> --itr <itr> --num_workers <num_workers> --train_epochs <train_epochs> --batch_size <batch_size> --patience <patience> --des <des> --learning_rate <learning_rate> --loss <loss> --lradj <lradj> --use_amp --inverse --use_gpu <use_gpu> --gpu <gpu> --use_multi_gpu --devices <devices> ``` The detailed descriptions about the arguments are as following: | Parameter name | Description of parameter | | --- | --- | | model | The model of experiment. This can be set to `informer`, `informerstack`, `informerlight(TBD)` | | data | The dataset name | | root_path | The root path of the data file (defaults to `./data/ETT/`) | | data_path | The data file name (defaults to `ETTh1.csv`) | | features | The forecasting task (defaults to `M`). This can be set to `M`,`S`,`MS` (M : multivariate predict multivariate, S : univariate predict univariate, MS : multivariate predict univariate) | | target | Target feature in S or MS task (defaults to `OT`) | | freq | Freq for time features encoding (defaults to `h`). This can be set to `s`,`t`,`h`,`d`,`b`,`w`,`m` (s:secondly, t:minutely, h:hourly, d:daily, b:business days, w:weekly, m:monthly).You can also use more detailed freq like 15min or 3h | | checkpoints | Location of model checkpoints (defaults to `./checkpoints/`) | | seq_len | Input sequence length of Informer encoder (defaults to 96) | | label_len | Start token length of Informer decoder (defaults to 48) | | pred_len | Prediction sequence length (defaults to 24) | | enc_in | Encoder input size (defaults to 7) | | dec_in | Decoder input size (defaults to 7) | | c_out | Output size (defaults to 7) | | d_model | Dimension of model (defaults to 512) | | n_heads | Num of heads (defaults to 8) | | e_layers | Num of encoder layers (defaults to 2) | | d_layers | Num of decoder layers (defaults to 1) | | s_layers | Num of stack encoder layers (defaults to `3,2,1`) | | d_ff | Dimension of fcn (defaults to 2048) | | factor | Probsparse attn factor (defaults to 5) | | padding | Padding type(defaults to 0). | | distil | Whether to use distilling in encoder, using this argument means not using distilling (defaults to `True`) | | dropout | The probability of dropout (defaults to 0.05) | | attn | Attention used i