Modelpublic

alex-miller/curated-gender-equality-weighted

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes14downloads

Model Card

curated-gender-equality-weighted

This model is a fine-tuned version of alex-miller/ODABert on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.3517
—Accuracy: 0.9298
—F1: 0.8728
—Precision: 0.8629
—Recall: 0.8829

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-06
—trainbatchsize: 24
—evalbatchsize: 24
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.7794	1.0	342	0.5844	0.7357	0.6625	0.5084	0.9508
0.5228	2.0	684	0.4504	0.8588	0.7793	0.6795	0.9133
0.4184	3.0	1026	0.3897	0.8957	0.8242	0.7627	0.8963
0.3696	4.0	1368	0.3621	0.9083	0.8423	0.7937	0.8972
0.3354	5.0	1710	0.3667	0.9064	0.8369	0.7976	0.8803
0.3079	6.0	2052	0.3525	0.9122	0.8459	0.8118	0.8829
0.2927	7.0	2394	0.3373	0.9181	0.8552	0.8260	0.8865
0.2741	8.0	2736	0.3219	0.9252	0.8668	0.8423	0.8928
0.2627	9.0	3078	0.3576	0.9147	0.8481	0.8245	0.8731
0.2528	10.0	3420	0.3306	0.9266	0.8685	0.8496	0.8883
0.2398	11.0	3762	0.3336	0.9281	0.8703	0.8564	0.8847
0.2318	12.0	4104	0.3289	0.9308	0.8751	0.8615	0.8892
0.2234	13.0	4446	0.3438	0.9271	0.8683	0.8559	0.8811
0.2169	14.0	4788	0.3311	0.9327	0.8787	0.8643	0.8937
0.2126	15.0	5130	0.3444	0.9288	0.8716	0.8580	0.8856
0.2078	16.0	5472	0.3442	0.9298	0.8734	0.8597	0.8874
0.206	17.0	5814	0.3520	0.9281	0.8699	0.8589	0.8811
0.1986	18.0	6156	0.3473	0.9305	0.8743	0.8632	0.8856
0.1958	19.0	6498	0.3502	0.9300	0.8732	0.8636	0.8829
0.1958	20.0	6840	0.3517	0.9298	0.8728	0.8629	0.8829

Framework versions

—Transformers 4.41.2
—Pytorch 2.0.1
—Datasets 2.20.0
—Tokenizers 0.19.1