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_256
    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: 0.6911764705882353
          - name: F1
            type: f1
            value: 0.7967741935483871

mobilebert_sa_GLUE_Experiment_logit_kd_mrpc_256

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: 0.4961
Accuracy: 0.6912
F1: 0.7968
Combined Score: 0.7440

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
0.6315	1.0	29	0.5588	0.6838	0.8122	0.7480
0.6098	2.0	58	0.5552	0.6838	0.8122	0.7480
0.6099	3.0	87	0.5544	0.6838	0.8122	0.7480
0.6084	4.0	116	0.5541	0.6838	0.8122	0.7480
0.603	5.0	145	0.5497	0.6838	0.8122	0.7480
0.5758	6.0	174	0.5335	0.7059	0.8171	0.7615
0.4984	7.0	203	0.4961	0.6912	0.7968	0.7440
0.4329	8.0	232	0.5478	0.6814	0.7743	0.7278
0.3876	9.0	261	0.5450	0.6838	0.7861	0.7349
0.3286	10.0	290	0.5792	0.6814	0.7628	0.7221
0.2833	11.0	319	0.5819	0.6446	0.7249	0.6847
0.2611	12.0	348	0.6755	0.6936	0.7913	0.7425

Framework versions

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