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: 3.2824
Accuracy: 0.1864
Precision: 0.1087
Recall: 0.1864
F1: 0.1114
Binary: 0.4189

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	4.2455	0.0533	0.0410	0.0533	0.0255	0.3252
No log	1.72	100	3.9851	0.0630	0.0179	0.0630	0.0163	0.3249
No log	2.59	150	3.7681	0.0726	0.0401	0.0726	0.0283	0.3421
No log	3.45	200	3.7448	0.0678	0.0383	0.0678	0.0252	0.3167
No log	4.31	250	3.5775	0.0896	0.0447	0.0896	0.0413	0.3453
No log	5.17	300	3.5078	0.0969	0.0380	0.0969	0.0400	0.3482
No log	6.03	350	3.4008	0.1283	0.0445	0.1283	0.0587	0.3777
No log	6.9	400	3.3443	0.1550	0.1089	0.1550	0.0850	0.3983
No log	7.76	450	3.2965	0.1792	0.1052	0.1792	0.1056	0.4177
3.768	8.62	500	3.2824	0.1864	0.1087	0.1864	0.1114	0.4189
3.768	9.48	550	3.2609	0.2058	0.1339	0.2058	0.1251	0.4332

Framework versions

Transformers 4.38.2
Pytorch 2.3.0
Datasets 2.19.1
Tokenizers 0.15.1