metadata

language:
  - en
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: mobilebert_sa_GLUE_Experiment_logit_kd_mrpc
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE MRPC
          type: glue
          config: mrpc
          split: validation
          args: mrpc
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6740196078431373
          - name: F1
            type: f1
            value: 0.7772194304857621

mobilebert_sa_GLUE_Experiment_logit_kd_mrpc

This model is a fine-tuned version of google/mobilebert-uncased on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

Loss: 0.5133
Accuracy: 0.6740
F1: 0.7772
Combined Score: 0.7256

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: 5e-05
train_batch_size: 128
eval_batch_size: 128
seed: 10
distributed_type: multi-GPU
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.6228	1.0	29	0.5556	0.6838	0.8122	0.7480
0.611	2.0	58	0.5551	0.6838	0.8122	0.7480
0.6095	3.0	87	0.5538	0.6838	0.8122	0.7480
0.6062	4.0	116	0.5503	0.6838	0.8122	0.7480
0.5825	5.0	145	0.5262	0.6985	0.8167	0.7576
0.4981	6.0	174	0.5197	0.6936	0.8038	0.7487
0.468	7.0	203	0.5133	0.6740	0.7772	0.7256
0.3901	8.0	232	0.5382	0.6838	0.7757	0.7297
0.323	9.0	261	0.6140	0.6789	0.7657	0.7223
0.2674	10.0	290	0.5512	0.6740	0.7687	0.7214
0.2396	11.0	319	0.6467	0.6667	0.7631	0.7149
0.2127	12.0	348	0.7811	0.6716	0.7690	0.7203

Framework versions

Transformers 4.26.0
Pytorch 1.14.0a0+410ce96
Datasets 2.9.0
Tokenizers 0.13.2