error with lstm_ae

[StefanBloemheuvel]

When i try to run the lstm_ae i get the following error:

IndexError                                Traceback (most recent call last)
[c:\Users\sdblo\Mijn](file:///C:/Users/sdblo/Mijn) Drive\PhD\Publicaties\graph_node_autoencoder\sequitur_example.py in line 56
     [31](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=30) # torch.use_deterministic_algorithms(True)
     [32](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=31) 
     [33](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=32) # train_data, test_data = train_test_split(data, test_size=0.1, shuffle=False, random_state=42)
   (...)
     [53](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=52) 
     [54](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=53) # train_set = torch.tensor(train_data, dtype=torch.float32)
     [55](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=54) train_set = [torch.randn(10, 5, 5) for _ in range(100)]
---> [56](file:///c%3A/Users/sdblo/Mijn%20Drive/PhD/Publicaties/graph_node_autoencoder/sequitur_example.py?line=55) encoder, decoder, _, _ = quick_train(LSTM_AE, train_set, encoding_dim=64, verbose=True, epochs=500, )

File [c:\Users\sdblo\miniconda3\envs\tsl\lib\site-packages\sequitur\quick_train.py:76](file:///C:/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py:76), in quick_train(model, train_set, encoding_dim, verbose, lr, epochs, denoise, **kwargs)
     [74](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=73) def quick_train(model, train_set, encoding_dim, verbose=False, lr=1e-3,
     [75](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=74)                 epochs=50, denoise=False, **kwargs):
---> [76](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=75)     model = instantiate_model(model, train_set, encoding_dim, **kwargs)
     [77](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=76)     losses = train_model(model, train_set, verbose, lr, epochs, denoise)
     [78](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=77)     encodings = get_encodings(model, train_set)

File [c:\Users\sdblo\miniconda3\envs\tsl\lib\site-packages\sequitur\quick_train.py:16](file:///C:/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py:16), in instantiate_model(model, train_set, encoding_dim, **kwargs)
     [14](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=13) def instantiate_model(model, train_set, encoding_dim, **kwargs):
     [15](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=14)     if model.__name__ in ("LINEAR_AE", "LSTM_AE"):
---> [16](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=15)         return model(train_set[-1].shape[-1], encoding_dim, **kwargs)
     [17](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=16)     elif model.__name__ == "CONV_LSTM_AE":
     [18](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/quick_train.py?line=17)         if len(train_set[-1].shape) == 3: # 2D elements

File [c:\Users\sdblo\miniconda3\envs\tsl\lib\site-packages\sequitur\models\lstm_ae.py:87](file:///C:/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py:87), in LSTM_AE.__init__(self, input_dim, encoding_dim, h_dims, h_activ, out_activ)
     [83](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=82) super(LSTM_AE, self).__init__()
     [85](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=84) self.encoder = Encoder(input_dim, encoding_dim, h_dims, h_activ,
     [86](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=85)                        out_activ)
---> [87](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=86) self.decoder = Decoder(encoding_dim, input_dim, h_dims[::-1],
     [88](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=87)                        h_activ)

File [c:\Users\sdblo\miniconda3\envs\tsl\lib\site-packages\sequitur\models\lstm_ae.py:46](file:///C:/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py:46), in Decoder.__init__(self, input_dim, out_dim, h_dims, h_activ)
     [43](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=42) def __init__(self, input_dim, out_dim, h_dims, h_activ):
     [44](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=43)     super(Decoder, self).__init__()
---> [46](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=45)     layer_dims = [input_dim] + h_dims + [h_dims[-1]]
     [47](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=46)     self.num_layers = len(layer_dims) - 1
     [48](file:///c%3A/Users/sdblo/miniconda3/envs/tsl/lib/site-packages/sequitur/models/lstm_ae.py?line=47)     self.layers = nn.ModuleList()

IndexError: list index out of range

[SecantZhang]

same error here.