metadata

license: apache-2.0
base_model: google/mobilebert-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: mobilebert_trajectory_proof0
    results: []

mobilebert_trajectory_proof0

This model is a fine-tuned version of google/mobilebert-uncased on the None dataset. It achieves the following results on the evaluation set:

Loss: 16478.7676
Accuracy: 0.7

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-06
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	13	3802792.75	0.7
No log	2.0	26	3156379.5	0.7
No log	3.0	39	2625222.75	0.7
No log	4.0	52	2124862.75	0.7
No log	5.0	65	1716339.0	0.7
No log	6.0	78	1375630.5	0.7
No log	7.0	91	1055760.0	0.7
No log	8.0	104	741251.0	0.7
No log	9.0	117	451555.9375	0.7
No log	10.0	130	227678.9219	0.7
No log	11.0	143	82893.1172	0.7
No log	12.0	156	48836.0117	0.7
No log	13.0	169	33417.5430	0.7
No log	14.0	182	21012.9023	0.7
No log	15.0	195	17473.7871	0.7
No log	16.0	208	16478.7676	0.7

Framework versions

Transformers 4.37.2
Pytorch 2.2.0+cu118
Datasets 2.17.1
Tokenizers 0.15.2