Modelpublic

gossminn/predict-perception-bertino-focus-assassin

sourceHugging Facemitupdated 4y agoView on Hugging Face

0likes28downloads

Model Card

predict-perception-bertino-focus-assassin

This model is a fine-tuned version of indigo-ai/BERTino on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.3409
—R2: 0.3205

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: 0.0001
—trainbatchsize: 20
—evalbatchsize: 8
—seed: 1996
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 47

Training results

Training Loss	Epoch	Step	Validation Loss	R2
0.5573	1.0	14	0.4856	0.0321
0.1739	2.0	28	0.4735	0.0562
0.0813	3.0	42	0.3416	0.3191
0.0764	4.0	56	0.3613	0.2799
0.0516	5.0	70	0.3264	0.3495
0.0459	6.0	84	0.4193	0.1643
0.0414	7.0	98	0.3502	0.3019
0.028	8.0	112	0.3361	0.3301
0.0281	9.0	126	0.3610	0.2804
0.027	10.0	140	0.3523	0.2978
0.0216	11.0	154	0.3440	0.3143
0.0181	12.0	168	0.3506	0.3012
0.013	13.0	182	0.3299	0.3424
0.0116	14.0	196	0.3611	0.2803
0.0118	15.0	210	0.3505	0.3013
0.0139	16.0	224	0.3529	0.2967
0.0099	17.0	238	0.3536	0.2952
0.0096	18.0	252	0.3542	0.2941
0.0107	19.0	266	0.3770	0.2486
0.0088	20.0	280	0.3467	0.3091
0.0065	21.0	294	0.3327	0.3369
0.0073	22.0	308	0.3479	0.3066
0.0062	23.0	322	0.3566	0.2893
0.0063	24.0	336	0.3503	0.3019
0.0057	25.0	350	0.3371	0.3282
0.0049	26.0	364	0.3334	0.3355
0.0045	27.0	378	0.3399	0.3225
0.0049	28.0	392	0.3379	0.3266
0.0049	29.0	406	0.3377	0.3268
0.0055	30.0	420	0.3357	0.3309
0.005	31.0	434	0.3394	0.3235
0.0046	32.0	448	0.3432	0.3159
0.0048	33.0	462	0.3427	0.3169
0.0041	34.0	476	0.3450	0.3123
0.0041	35.0	490	0.3436	0.3151
0.0051	36.0	504	0.3394	0.3234
0.0037	37.0	518	0.3370	0.3283
0.004	38.0	532	0.3370	0.3284
0.0033	39.0	546	0.3339	0.3344
0.0034	40.0	560	0.3335	0.3352
0.003	41.0	574	0.3373	0.3276
0.0035	42.0	588	0.3380	0.3264
0.0032	43.0	602	0.3382	0.3259
0.0034	44.0	616	0.3432	0.3158
0.003	45.0	630	0.3421	0.3181
0.0027	46.0	644	0.3410	0.3203
0.0037	47.0	658	0.3409	0.3205

Framework versions

—Transformers 4.16.2
—Pytorch 1.10.2+cu113
—Datasets 1.18.3
—Tokenizers 0.11.0