metadata

license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: hubert-classifier
    results: []

hubert-classifier

This model is a fine-tuned version of facebook/hubert-base-ls960 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.4231
Accuracy: 0.0993
Precision: 0.0191
Recall: 0.0993
F1: 0.0296
Binary: 0.3603

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: 3e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Binary
No log	0.86	50	4.4031	0.0218	0.0042	0.0218	0.0061	0.1886
No log	1.72	100	4.1752	0.0363	0.0204	0.0363	0.0100	0.3102
No log	2.59	150	3.9334	0.0363	0.0159	0.0363	0.0125	0.3102
No log	3.45	200	3.7751	0.0557	0.0270	0.0557	0.0268	0.3266
No log	4.31	250	3.6877	0.0775	0.0241	0.0775	0.0314	0.3383
No log	5.17	300	3.5839	0.0920	0.0220	0.0920	0.0329	0.3550
No log	6.03	350	3.5119	0.0920	0.0219	0.0920	0.0327	0.3586
No log	6.9	400	3.4851	0.0872	0.0274	0.0872	0.0283	0.3508
No log	7.76	450	3.4593	0.0920	0.0221	0.0920	0.0312	0.3552
3.8751	8.62	500	3.4240	0.1017	0.0223	0.1017	0.0328	0.3610
3.8751	9.48	550	3.4231	0.0993	0.0191	0.0993	0.0296	0.3603

Framework versions

Transformers 4.38.2
Pytorch 2.3.0
Datasets 2.19.1
Tokenizers 0.15.1