README.md

metadata

license: mit
base_model: FacebookAI/roberta-base
tags:
  - generated_from_trainer
model-index:
  - name: roberta-petco-fullemailbody-ctr
    results: []

roberta-petco-fullemailbody-ctr

This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4048
• Mse: 0.4048
• Rmse: 0.6363
• Mae: 0.4301
• R2: 0.2106

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Rmse	Mae	R2
1.3784	1.0	15	0.5080	0.5080	0.7127	0.5575	0.0095
0.5017	2.0	30	0.5043	0.5043	0.7101	0.5241	0.0166
0.4986	3.0	45	0.4864	0.4864	0.6974	0.5219	0.0515
0.4621	4.0	60	0.5096	0.5096	0.7139	0.5096	0.0063
0.4483	5.0	75	0.5069	0.5069	0.7120	0.5031	0.0116
0.4396	6.0	90	0.4707	0.4707	0.6861	0.5275	0.0822
0.4145	7.0	105	0.4661	0.4661	0.6828	0.4802	0.0910
0.4293	8.0	120	0.5122	0.5122	0.7157	0.4884	0.0012
0.3681	9.0	135	0.4358	0.4358	0.6601	0.4947	0.1502
0.3349	10.0	150	0.4676	0.4676	0.6838	0.4434	0.0882
0.3003	11.0	165	0.4601	0.4601	0.6783	0.4904	0.1028
0.3132	12.0	180	0.5026	0.5026	0.7089	0.4734	0.0200
0.3446	13.0	195	0.4411	0.4411	0.6641	0.4655	0.1399
0.2935	14.0	210	0.5986	0.5986	0.7737	0.6535	-0.1672
0.2301	15.0	225	0.4409	0.4409	0.6640	0.4506	0.1403
0.2152	16.0	240	0.4048	0.4048	0.6363	0.4301	0.2106
0.2056	17.0	255	0.4115	0.4115	0.6415	0.4429	0.1976
0.1885	18.0	270	0.4058	0.4058	0.6370	0.4467	0.2087
0.1754	19.0	285	0.4282	0.4282	0.6543	0.4765	0.1651
0.1758	20.0	300	0.4076	0.4076	0.6384	0.4487	0.2052

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2