moderate_severe_depression_model

This model is a fine-tuned version of allenai/longformer-scico on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4276
• Macro F1: 0.8927
• Accuracy: 0.8932

Results on Test set:

-Accuracy: 0.8817204301075269

-F1 score: 0.8819253137510324

-Precision: 0.8855477220587717

-Recall : 0.8817204301075269

-Matthews Correlation Coefficient: 0.8242972089300715

-Precision of each class: [0.98420129 0.85636693 0.81607495]

-Recall of each class: [0.93548387 0.78921023 0.92046719]

-F1 score of each class: [0.95922441 0.82141823 0.8651333 ]

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 12
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Macro F1	Accuracy
0.3476	1.0	1798	0.3343	0.8765	0.8782
0.2658	2.0	3596	0.3190	0.8856	0.8859
0.2157	3.0	5394	0.3607	0.8938	0.8939
0.1749	4.0	7192	0.4276	0.8927	0.8932

Framework versions

• Transformers 4.27.1
• Pytorch 2.0.1+cu118
• Datasets 2.9.0
• Tokenizers 0.13.3