Probabilistic Time Series Forecasting
The following code gave me an error
# perform forward pass
outputs = model(
past_values=batch["past_values"],
past_time_features=batch["past_time_features"],
past_observed_mask=batch["past_observed_mask"],
static_categorical_features=batch["static_categorical_features"]
if config.num_static_categorical_features > 0
else None,
static_real_features=batch["static_real_features"]
if config.num_static_real_features > 0
else None,
future_values=batch["future_values"],
future_time_features=batch["future_time_features"],
future_observed_mask=batch["future_observed_mask"],
output_hidden_states=True,
)
=====================================================
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
Cell In[40], line 2
1 # perform forward pass
----> 2 outputs = model(
3 past_values=batch["past_values"],
4 past_time_features=batch["past_time_features"],
5 past_observed_mask=batch["past_observed_mask"],
6 static_categorical_features=batch["static_categorical_features"]
7 if config.num_static_categorical_features > 0
8 else None,
9 static_real_features=batch["static_real_features"]
10 if config.num_static_real_features > 0
11 else None,
12 future_values=batch["future_values"],
13 future_time_features=batch["future_time_features"],
14 future_observed_mask=batch["future_observed_mask"],
15 output_hidden_states=True,
16 )
File ~\anaconda3\lib\site-packages\torch\nn\modules\module.py:1130, in Module._call_impl(self, *input, **kwargs)
1126 # If we don't have any hooks, we want to skip the rest of the logic in
1127 # this function, and just call forward.
1128 if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks
1129 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1130 return forward_call(*input, **kwargs)
1131 # Do not call functions when jit is used
1132 full_backward_hooks, non_full_backward_hooks = [], []
File ~\anaconda3\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py:1813, in TimeSeriesTransformerForPrediction.forward(self, past_values, past_time_features, past_observed_mask, static_categorical_features, static_real_features, future_values, future_time_features, future_observed_mask, decoder_attention_mask, head_mask, decoder_head_mask, cross_attn_head_mask, encoder_outputs, past_key_values, output_hidden_states, output_attentions, use_cache, return_dict)
1810 if future_values is not None:
1811 use_cache = False
-> 1813 outputs = self.model(
1814 past_values=past_values,
1815 past_time_features=past_time_features,
1816 past_observed_mask=past_observed_mask,
1817 static_categorical_features=static_categorical_features,
1818 static_real_features=static_real_features,
1819 future_values=future_values,
1820 future_time_features=future_time_features,
1821 decoder_attention_mask=decoder_attention_mask,
1822 head_mask=head_mask,
1823 decoder_head_mask=decoder_head_mask,
1824 cross_attn_head_mask=cross_attn_head_mask,
1825 encoder_outputs=encoder_outputs,
1826 past_key_values=past_key_values,
1827 output_hidden_states=output_hidden_states,
1828 output_attentions=output_attentions,
1829 use_cache=use_cache,
1830 return_dict=return_dict,
1831 )
1833 prediction_loss = None
1834 params = None
File ~\anaconda3\lib\site-packages\torch\nn\modules\module.py:1130, in Module._call_impl(self, *input, **kwargs)
1126 # If we don't have any hooks, we want to skip the rest of the logic in
1127 # this function, and just call forward.
1128 if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks
1129 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1130 return forward_call(*input, **kwargs)
1131 # Do not call functions when jit is used
1132 full_backward_hooks, non_full_backward_hooks = [], []
File ~\anaconda3\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py:1626, in TimeSeriesTransformerModel.forward(self, past_values, past_time_features, past_observed_mask, static_categorical_features, static_real_features, future_values, future_time_features, decoder_attention_mask, head_mask, decoder_head_mask, cross_attn_head_mask, encoder_outputs, past_key_values, output_hidden_states, output_attentions, use_cache, return_dict)
1623 use_cache = use_cache if use_cache is not None else self.config.use_cache
1624 return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-> 1626 transformer_inputs, scale, static_feat = self.create_network_inputs(
1627 past_values=past_values,
1628 past_time_features=past_time_features,
1629 past_observed_mask=past_observed_mask,
1630 static_categorical_features=static_categorical_features,
1631 static_real_features=static_real_features,
1632 future_values=future_values,
1633 future_time_features=future_time_features,
1634 )
1636 if encoder_outputs is None:
1637 enc_input = transformer_inputs[:, : self.config.context_length, ...]
File ~\anaconda3\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py:1535, in TimeSeriesTransformerModel.create_network_inputs(self, past_values, past_time_features, static_categorical_features, static_real_features, past_observed_mask, future_values, future_time_features)
1533 # static features
1534 log_scale = scale.log() if self.config.input_size == 1 else scale.squeeze(1).log()
-> 1535 static_feat = torch.cat((embedded_cat, static_real_features, log_scale), dim=1)
1536 expanded_static_feat = static_feat.unsqueeze(1).expand(-1, time_feat.shape[1], -1)
1538 # all features
TypeError: expected Tensor as element 1 in argument 0, but got NoneType
My Anaconda has Python version :slight_smile:
!python --version
Python 3.10.9