Stylistic fixes and addition of multi-modal scaffolding

flood_forecast/basic/linear_regression.py

@@ -112,7 +112,7 @@ def simple_decode(model: Type[torch.nn.Module],
             else:
                 # residual = output_len if max_seq_len - output_len - i >= 0 else max_seq_len % output_len
                 if output_len != out.shape[1]:
-                    raise ValueError("Output length should laways equal the output shape")
+                    raise ValueError("Output length should always equal the output shape")
                 real_target2[:, i:i + residual, 0:multi_targets] = out[:, :residual]
                 src = torch.cat((src[:, residual:, :], real_target2[:, i:i + residual, :]), 1)
                 ys = torch.cat((ys, real_target2[:, i:i + residual, :]), 1)

flood_forecast/evaluator.py

@@ -3,7 +3,7 @@
 Description:
     This module contains functions for evaluating models. The basic logic flow is as follows:
     1. `evaluate_model` is called from `trainer.py` at the end of training. It calls `infer_on_torch_model` which does the actual inference. # noqa
-    2. `infer_on_torch_model` calls `generate_predictions` which calls `generate_decoded_predictions` or `generate_predictions_non_decoded` depending on whether the model uses a decoder or not.
+    2. `infer_on_torch_mode` calls `generate_predictions` which calls `generate_decoded_predictions` or `generate_predictions_non_decoded` depending on whether the model uses a decoder or not.
     3. `generate_decoded_predictions` calls `decoding_functions` which calls `greedy_decode` or `beam_decode` depending on the decoder function specified in the config file.
     4. The returned value from `generate_decoded_predictions` is then used to calculate the evaluation metrics in `run_evaluation`.
     5. `run_evaluation` returns the evaluation metrics to `evaluate_model` which returns them to `trainer.py`.

flood_forecast/model_dict_function.py

@@ -1,4 +1,4 @@
-from flood_forecast.multi_models.crossvivit import RoCrossViViT
+from flood_forecast.multimodal_models.crossvivit import RoCrossViViT
 from flood_forecast.transformer_xl.multi_head_base import MultiAttnHeadSimple
 from flood_forecast.transformer_xl.transformer_basic import SimpleTransformer, CustomTransformerDecoder
 from flood_forecast.transformer_xl.informer import Informer