Update README.md

README.md

@@ -14,15 +14,12 @@ model-index:
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# PreDA_t5-base
+![framework](preda_architecture_digram.png)
+
+# PreDA-base (Prefix-Based Dream Reports Annotation)
+
+This model is a fine-tuned version of [google-t5/t5-base](https://huggingface.co/google-t5/t5-small) on the annotated [Dreambank.net](https://dreambank.net/) dataset.It achieves the following results on the evaluation set:
 
-This model is a fine-tuned version of [google-t5/t5-base](https://huggingface.co/google-t5/t5-base) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 2.4362
-- Rouge1: 0.7866
-- Rouge2: 0.6744
-- Rougel: 0.7633
-- Rougelsum: 0.7631
 
 ## Model description
 
@@ -38,7 +35,24 @@ More information needed
 
 ## Training procedure
 
-### Training hyperparameters
+the overall idea of our approach is to disentangle each dream report from its annotation as a whole and to create an augmented set of (dream report; single 
+feature annotation). To make sure that, given the same report, the model would produce a specific HVDC feature, we simply append at 
+the beginning of each report a string of the form ``HVDC-Feature :'', in a manner that closely mimics T5 task-specific prefix fine-tuning. 
+
+After this procedure to the original dataset (\~1.8K) we obtain approximately 6.6K items. %For this work, we focused on six HVDC features, namely Characters, 
+Activities, Emotion, Friendliness, Misfortune, and Good Fortune. We did so to exclude features that amounted to less than 10\% of the total instance. Indeed, 
+this would have excluded Good Fortune (see Figure \ref{fig:train_set_prefix_dist}). We include either way the feature to control for 
+memorisation and counterbalance the Misfortune feature. 
+In the present study, we focused on a subset of six HVDC features: Characters, Activities, Emotion, Friendliness, Misfortune, and Good Fortune. 
+This selection was made to exclude features that represented less than 10\% of the total instances. Notably, Good Fortune would have been excluded under this 
+criterion (refer to Figure \ref{fig:train_set_prefix_dist}), but we intentionally retained this feature to control against potential 
+memorisation effects and to provide a counterbalance to the Misfortune feature. After filtering out instances whose annotation 
+feature is not one of the six selected features, we are left with \~5.3K %5389 
+dream reports. We then generate a random split of 80\%-20\% for the training (i.e., 4,311 reports) and testing (i.e. 1,078 reports) sets.
+
+### Training
+
+#### Hyperparameters
 
 The following hyperparameters were used during training:
 - learning_rate: 0.001
@@ -52,7 +66,7 @@ The following hyperparameters were used during training:
 - mixed_precision_training: Native AMP
 - label_smoothing_factor: 0.1
 
-### Training results
+#### Metrics
 
 | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
@@ -77,6 +91,19 @@ The following hyperparameters were used during training:
 | 2.4518        | 19.0  | 2565 | 2.4362          | 0.7866 | 0.6753 | 0.7635 | 0.7635    |
 | 2.4663        | 20.0  | 2700 | 2.4362          | 0.7866 | 0.6744 | 0.7633 | 0.7631    |
 
+### Evaluation
+
+We selected the best model via validation loss. The table below reposts overall and feature-specific scores.
+
+| Feature      |   rouge1 |   rouge2 |   rougeL |   rougeLsum |
+|:-------------|---------:|---------:|---------:|------------:|
+
+
+### Disclaimer
+Dream reports and their annotation have been used in clinical settings and applied for diagnostic purposes. 
+This does not apply in any way to our experimental results and output. 
+Our work aims to provide experimental evidence of the feasibility of using PLMs to support humans
+in annotating dream reports for research purposes, as well as detailing their strengths and limitations when approaching such a task.
 
 ### Framework versions