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: 3.2561
Accuracy: 0.2155
Precision: 0.1870
Recall: 0.2155
F1: 0.1453
Binary: 0.4380

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	4.2865	0.0291	0.0034	0.0291	0.0043	0.2661
No log	1.72	100	4.0141	0.0460	0.0306	0.0460	0.0187	0.3199
No log	2.59	150	3.8172	0.0630	0.0204	0.0630	0.0214	0.3378
No log	3.45	200	3.7045	0.0896	0.0534	0.0896	0.0533	0.3431
No log	4.31	250	3.5695	0.1308	0.0844	0.1308	0.0796	0.3785
No log	5.17	300	3.4883	0.1622	0.1140	0.1622	0.1002	0.3954
No log	6.03	350	3.4447	0.1622	0.1145	0.1622	0.1005	0.3925
No log	6.9	400	3.3482	0.1671	0.0972	0.1671	0.1026	0.4077
No log	7.76	450	3.3042	0.1961	0.1424	0.1961	0.1295	0.4228
3.7796	8.62	500	3.2724	0.2131	0.1898	0.2131	0.1472	0.4370
3.7796	9.48	550	3.2561	0.2155	0.1870	0.2155	0.1453	0.4380

Framework versions

Transformers 4.38.2
Pytorch 2.3.0
Datasets 2.19.1
Tokenizers 0.15.1