MCPcopy Create free account

hub / github.com/thuml/Time-Series-Library / functions

Functions768 in github.com/thuml/Time-Series-Library

⨍ Functions 768◇ Types & classes 221
↓ 56 callersMethodencoder
(self, x)
models/SegRNN.py:52
↓ 32 callersMethodload
Load cached dataset. :param training: Load training part if training is True, test part otherwise.
data_provider/m4.py:104
↓ 20 callersMethodtransform
(self, data)
utils/tools.py:76
↓ 11 callersFunctionsfb1d
1D synthesis filter bank of an image tensor
layers/DWT_Decomposition.py:481
↓ 10 callersMethod__init__
(self)
layers/MSGBlock.py:155
↓ 10 callersMethod__init__
(self, input_size, output_size)
models/TemporalFusionTransformer.py:93
↓ 9 callersMethod__init__
(self, layers)
layers/ETSformer_EncDec.py:251
↓ 8 callersMethod__init__
(self, c_in, d_model, embed_type='fixed', freq='h', dropout=0.1)
layers/Embed.py:110
↓ 8 callersFunctionafb1d
1D analysis filter bank (along one dimension only) of an image Inputs: x (tensor): 4D input with the last two dimensions the spatial inp
layers/DWT_Decomposition.py:334
↓ 7 callersMethod__init__
(self, attn_layers, conv_layers=None, norm_layer=None)
layers/Autoformer_EncDec.py:114
↓ 7 callersFunctiondata_provider
(args, flag)
data_provider/data_factory.py:22
↓ 7 callersMethodstep
(self, hidden_states, conv_state, ssm_state)
layers/MambaBlock.py:273
↓ 6 callersMethod__init__
(self, f_in, f_out, hidden_dim=128, hi
models/Koopa.py:30
↓ 6 callersMethod_get_data
(self, flag)
exp/exp_classification.py:34
↓ 6 callersFunctionarchADF
(root_path, data_path)
utils/ADFtest.py:33
↓ 6 callersMethodbackward
(ctx, low, highs)
layers/DWT_Decomposition.py:616
↓ 6 callersMethodinverse_transform
(self, data)
utils/tools.py:79
↓ 6 callersMethodout_projection
(self, dec_out, i, out_res)
models/TimeMixer.py:269
↓ 6 callersFunctionroll
(x, n, dim, make_even=False)
layers/DWT_Decomposition.py:252
↓ 5 callersMethod__init__
(self, dim, n_heads)
layers/TimeFilter_layers.py:8
↓ 5 callersMethod__init__
(self, mask_flag=True, factor=5, scale=None, attention_dropout=0.1, output_attention=False)
layers/SelfAttention_Family.py:13
↓ 5 callersMethod_get_data
(self, flag)
exp/exp_anomaly_detection.py:31
↓ 5 callersMethodsummarize_groups
Re-group scores respecting M4 rules. :param scores: Scores per group. :return: Grouped scores.
utils/m4_summary.py:113
↓ 4 callersMethod__init__
(self, configs, window_size, inner_size)
layers/Pyraformer_EncDec.py:103
↓ 4 callersMethod__init__
(self, k=3, alpha=64, L=0, c=1, base='legendre',
layers/MultiWaveletCorrelation.py:507
↓ 4 callersMethod__init__
(self, attn_layers)
layers/Crossformer_EncDec.py:62
↓ 4 callersMethod__init__
(self, attn_layers, conv_layers=None, norm_layer=None)
layers/Transformer_EncDec.py:55
↓ 4 callersMethod__init__
(self, configs)
models/TimeMixer.py:189
↓ 4 callersMethod__init__
(self, input_seq=[], out_seq=[], batch_size=[],
models/WPMixer.py:41
↓ 4 callersMethod__init__
(self, configs)
models/SCINet.py:103
↓ 4 callersMethod__init__
(self, configs)
models/TimeXer.py:116
↓ 4 callersMethod__multi_scale_process_inputs
(self, x_enc, x_mark_enc)
models/TimeMixer.py:289
↓ 4 callersMethod_get_data
(self, flag)
exp/exp_long_term_forecasting.py:29
↓ 4 callersMethod_get_data
(self, flag)
exp/exp_short_term_forecasting.py:35
↓ 4 callersMethod_get_data
(self, flag)
exp/exp_imputation.py:27
↓ 4 callersMethod_get_mask
(self, device)
models/TimeFilter.py:75
↓ 4 callersFunction_traceback
(DTW, slope_constraint)
utils/dtw.py:12
↓ 4 callersFunctionadjust_learning_rate
(optimizer, epoch, args)
utils/tools.py:12
↓ 4 callersMethodencoder
(self, x)
models/DLinear.py:55
↓ 4 callersMethodencoder
(self, x)
models/LightTS.py:102
↓ 4 callersFunctionint_to_mode
(mode)
layers/DWT_Decomposition.py:558
↓ 4 callersFunctionprep_filt_afb2d
Prepares the filters to be of the right form for the afb2d function. In particular, makes the tensors the right shape. It takes mirror image
layers/DWT_Decomposition.py:1198
↓ 4 callersFunctionpsi
(psi1, psi2, i, inp)
layers/MultiWaveletCorrelation.py:141
↓ 4 callersFunctionreflect
Reflect the values in matrix *x* about the scalar values *minx* and *maxx*. Hence a vector *x* containing a long linearly increasing series is
layers/DWT_Decomposition.py:1251
↓ 4 callersFunctionrun_augmentation_single
(x, y, args)
utils/augmentation.py:350
↓ 4 callersFunctionvisual
Results visualization
utils/tools.py:83
↓ 3 callersMethod__init__
(self, configs)
models/MambaSimple.py:18
↓ 3 callersFunctionapply_mask_to_region
(adj_mask, n)
layers/TimeFilter_layers.py:81
↓ 3 callersMethodcompl_mul1d
(self, order, x, weights)
layers/FourierCorrelation.py:108
↓ 3 callersFunctiondivide_no_nan
a/b where the resulted NaN or Inf are replaced by 0.
utils/losses.py:25
↓ 3 callersFunctionget_frequency_modes
get modes on frequency domain: 'random' means sampling randomly; 'else' means sampling the lowest modes;
layers/FourierCorrelation.py:10
↓ 3 callersFunctiongroup_values
(values, groups, group_name)
utils/m4_summary.py:28
↓ 3 callersMethodinverse_transform
(self, data)
data_provider/data_loader.py:393
↓ 3 callersFunctionmetric
(pred, true)
utils/metrics.py:34
↓ 3 callersFunctionmypad
Function to do numpy like padding on tensors. Only works for 2-D padding. Inputs: x (tensor): tensor to pad pad (tuple): tup
layers/DWT_Decomposition.py:271
↓ 3 callersFunctionprep_filt_afb1d
Prepares the filters to be of the right form for the afb2d function. In particular, makes the tensors the right shape. It takes mirror image
layers/DWT_Decomposition.py:1229
↓ 3 callersFunctionprep_filt_sfb1d
Prepares the filters to be of the right form for the sfb1d function. In particular, makes the tensors the right shape. It does not mirror ima
layers/DWT_Decomposition.py:1175
↓ 3 callersFunctiontime_features
(dates, freq='h')
utils/timefeatures.py:147
↓ 3 callersMethodwavelet_transform
(self, x)
layers/MultiWaveletCorrelation.py:373
↓ 2 callersMethodFreMLP
(self, B, nd, dimension, x, r, i, rb, ib)
models/FreTS.py:75
↓ 2 callersFunctionMSE
(pred, true)
utils/metrics.py:18
↓ 2 callersMethod__init__
(self)
utils/losses.py:36
↓ 2 callersMethod__init__
(self, J=1, wave='db1', mode='zero', use_amp=False)
layers/DWT_Decomposition.py:151
↓ 2 callersMethod__init__
(self, configs)
models/FiLM.py:95
↓ 2 callersMethod__init__
(self, configs)
models/MultiPatchFormer.py:75
↓ 2 callersMethod__init__
patch_len: int, patch len for patch_embedding stride: int, stride for patch_embedding
models/PatchTST.py:36
↓ 2 callersMethod__init__
(self, feature_size=512, n_heads=8, dropout=0.05, decomp_kernel=[32], conv_kernel=[24], isome
models/MICN.py:13
↓ 2 callersMethod__init__
(self, configs, bias=True, feature_encode_dim=2)
models/TiDE.py:46
↓ 2 callersMethod__init__
(self, configs)
models/MambaSingleLayer.py:47
↓ 2 callersFunction_cummulative_matrix
(cost, slope_constraint, window)
utils/dtw.py:79
↓ 2 callersMethod_dummy_forward
(self, input_length)
layers/DWT_Decomposition.py:81
↓ 2 callersFunction_traceback
(D)
utils/dtw_metric.py:100
↓ 2 callersFunctionaccelerated_dtw
Computes Dynamic Time Warping (DTW) of two sequences in a faster way. Instead of iterating through each element and calculating each distance
utils/dtw_metric.py:58
↓ 2 callersFunctionafb1d_atrous
1D analysis filter bank (along one dimension only) of an image without downsampling. Does the a trous algorithm. Inputs: x (tensor):
layers/DWT_Decomposition.py:430
↓ 2 callersFunctionaugment
(x, y, args)
utils/augmentation.py:386
↓ 2 callersFunctioncal_accuracy
(y_pred, y_true)
utils/tools.py:119
↓ 2 callersMethodcompl_mul1d
(self, order, x, weights)
layers/MultiWaveletCorrelation.py:404
↓ 2 callersFunctionconv1d_fft
(f, g, dim=-1)
layers/ETSformer_EncDec.py:28
↓ 2 callersFunctionget_filter
(base, k)
layers/MultiWaveletCorrelation.py:140
↓ 2 callersFunctionget_known_len
(embed_type, freq)
models/TemporalFusionTransformer.py:20
↓ 2 callersMethodget_position_encoding
(self, x)
models/SCINet.py:179
↓ 2 callersMethodjitter
(self, x)
layers/ETSformer_EncDec.py:18
↓ 2 callersMethodlast_insample_window
The last window of insample size of all timeseries. This function does not support batching and does not reshuffle timeseries.
data_provider/data_loader.py:396
↓ 2 callersFunctionmase
(forecast, insample, outsample, frequency)
utils/m4_summary.py:32
↓ 2 callersFunctionmode_to_int
(mode)
layers/DWT_Decomposition.py:539
↓ 2 callersMethodnormalize
Args: df: input dataframe Returns: df: normalized dataframe
data_provider/uea.py:78
↓ 2 callersFunctionprep_filt_afb2d_nonsep
Prepares the filters to be of the right form for the afb2d_nonsep function. In particular, makes 2d point spread functions, and mirror images
layers/DWT_Decomposition.py:1074
↓ 2 callersFunctionprep_filt_sfb2d
Prepares the filters to be of the right form for the sfb2d function. In particular, makes the tensors the right shape. It does not mirror im
layers/DWT_Decomposition.py:1143
↓ 2 callersFunctionprep_filt_sfb2d_nonsep
Prepares the filters to be of the right form for the sfb2d_nonsep function. In particular, makes 2d point spread functions. Does not mirror i
layers/DWT_Decomposition.py:1109
↓ 2 callersMethodsave_checkpoint
(self, val_loss, model, path)
utils/tools.py:57
↓ 2 callersMethodscale
(self, x)
layers/ETSformer_EncDec.py:21
↓ 2 callersFunctionsmape_2
(forecast, target)
utils/m4_summary.py:36
↓ 2 callersMethodtest
(self)
exp/exp_basic.py:75
↓ 2 callersMethodvali
(self, vali_data, vali_loader, criterion)
exp/exp_classification.py:47
↓ 2 callersMethodvali
(self, vali_data, vali_loader, criterion)
exp/exp_long_term_forecasting.py:42
↓ 2 callersMethodvali
(self, vali_data, vali_loader, criterion)
exp/exp_anomaly_detection.py:43
↓ 2 callersMethodvali
(self, vali_data, vali_loader, criterion)
exp/exp_imputation.py:39
↓ 2 callersFunctionwindow_slice
(x, reduce_ratio=0.9)
utils/augmentation.py:74
↓ 2 callersMethodzip_up_the_pants
(self, even, odd)
models/SCINet.py:86
↓ 1 callersFunctionFFT_for_Period
(x, k=2)
models/TimesNet.py:9
next →1–100 of 768, ranked by callers