predict-perception-xlmr-focus-assassin

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3264
• Rmse: 0.9437
• Rmse Focus::a Sull'assassino: 0.9437
• Mae: 0.7093
• Mae Focus::a Sull'assassino: 0.7093
• R2: 0.6145
• R2 Focus::a Sull'assassino: 0.6145
• Cos: 0.7391
• Pair: 0.0
• Rank: 0.5
• Neighbors: 0.6131
• Rsa: nan

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: 20
• eval_batch_size: 8
• seed: 1996
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Rmse	Rmse Focus::a Sull'assassino	Mae	Mae Focus::a Sull'assassino	R2	R2 Focus::a Sull'assassino	Cos	Pair	Rank	Neighbors	Rsa
1.0403	1.0	15	1.1576	1.7771	1.7771	1.6028	1.6028	-0.3670	-0.3670	-0.2174	0.0	0.5	0.2379	nan
0.9818	2.0	30	0.8916	1.5596	1.5596	1.4136	1.4136	-0.0529	-0.0529	0.3913	0.0	0.5	0.3793	nan
0.9276	3.0	45	0.9277	1.5909	1.5909	1.4560	1.4560	-0.0955	-0.0955	0.3913	0.0	0.5	0.3742	nan
0.8395	4.0	60	0.7958	1.4734	1.4734	1.3032	1.3032	0.0603	0.0603	0.5652	0.0	0.5	0.4598	nan
0.7587	5.0	75	0.4647	1.1259	1.1259	0.9316	0.9316	0.4513	0.4513	0.6522	0.0	0.5	0.5087	nan
0.696	6.0	90	0.5368	1.2101	1.2101	1.0847	1.0847	0.3661	0.3661	0.7391	0.0	0.5	0.5302	nan
0.548	7.0	105	0.3110	0.9211	0.9211	0.7896	0.7896	0.6328	0.6328	0.6522	0.0	0.5	0.5261	nan
0.4371	8.0	120	0.3392	0.9619	0.9619	0.8132	0.8132	0.5995	0.5995	0.6522	0.0	0.5	0.5261	nan
0.355	9.0	135	0.3938	1.0366	1.0366	0.8153	0.8153	0.5349	0.5349	0.7391	0.0	0.5	0.6131	nan
0.2919	10.0	150	0.3484	0.9749	0.9749	0.7487	0.7487	0.5886	0.5886	0.7391	0.0	0.5	0.6131	nan
0.2595	11.0	165	0.2812	0.8759	0.8759	0.6265	0.6265	0.6679	0.6679	0.7391	0.0	0.5	0.6131	nan
0.2368	12.0	180	0.2534	0.8314	0.8314	0.6402	0.6402	0.7008	0.7008	0.7391	0.0	0.5	0.6131	nan
0.227	13.0	195	0.2878	0.8861	0.8861	0.6769	0.6769	0.6601	0.6601	0.7391	0.0	0.5	0.6131	nan
0.1979	14.0	210	0.2405	0.8100	0.8100	0.6113	0.6113	0.7160	0.7160	0.7391	0.0	0.5	0.6131	nan
0.1622	15.0	225	0.2575	0.8382	0.8382	0.6017	0.6017	0.6959	0.6959	0.8261	0.0	0.5	0.6622	nan
0.1575	16.0	240	0.2945	0.8963	0.8963	0.6741	0.6741	0.6523	0.6523	0.8261	0.0	0.5	0.6622	nan
0.1479	17.0	255	0.3563	0.9859	0.9859	0.7367	0.7367	0.5792	0.5792	0.8261	0.0	0.5	0.6622	nan
0.1269	18.0	270	0.2806	0.8750	0.8750	0.6665	0.6665	0.6686	0.6686	0.8261	0.0	0.5	0.6622	nan
0.1257	19.0	285	0.3267	0.9441	0.9441	0.6739	0.6739	0.6142	0.6142	0.8261	0.0	0.5	0.6622	nan
0.134	20.0	300	0.3780	1.0155	1.0155	0.7331	0.7331	0.5536	0.5536	0.7391	0.0	0.5	0.5302	nan
0.1171	21.0	315	0.3890	1.0301	1.0301	0.7444	0.7444	0.5406	0.5406	0.8261	0.0	0.5	0.6622	nan
0.0934	22.0	330	0.3131	0.9242	0.9242	0.6923	0.6923	0.6303	0.6303	0.8261	0.0	0.5	0.6622	nan
0.1112	23.0	345	0.2912	0.8913	0.8913	0.6610	0.6610	0.6561	0.6561	0.8261	0.0	0.5	0.6622	nan
0.1038	24.0	360	0.3109	0.9209	0.9209	0.7019	0.7019	0.6329	0.6329	0.8261	0.0	0.5	0.6622	nan
0.085	25.0	375	0.3469	0.9728	0.9728	0.7383	0.7383	0.5904	0.5904	0.8261	0.0	0.5	0.6622	nan
0.0843	26.0	390	0.3017	0.9073	0.9073	0.6848	0.6848	0.6437	0.6437	0.7391	0.0	0.5	0.6131	nan
0.093	27.0	405	0.3269	0.9443	0.9443	0.7042	0.7042	0.6140	0.6140	0.7391	0.0	0.5	0.6131	nan
0.0846	28.0	420	0.3161	0.9286	0.9286	0.6937	0.6937	0.6267	0.6267	0.7391	0.0	0.5	0.6131	nan
0.0764	29.0	435	0.3244	0.9408	0.9408	0.7079	0.7079	0.6169	0.6169	0.7391	0.0	0.5	0.6131	nan
0.0697	30.0	450	0.3264	0.9437	0.9437	0.7093	0.7093	0.6145	0.6145	0.7391	0.0	0.5	0.6131	nan

Framework versions

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