You can deploy this model and start earning money today!

mhuzaifa5/
results

Name: results
Availability: InStock

generated_from_trainer

Hate-speech-CNERG/urdu-abusive-MuRIL

finetune

afl-3.0

text-embeddings-inference

endpoints_compatible

Views

No views yet

results

This model is a fine-tuned version of Hate-speech-CNERG/urdu-abusive-MuRIL on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.7248
Accuracy: 0.8862
F1: 0.8860
Precision: 0.8859
Recall: 0.8860
Tp: 265
Tn: 241
Fp: 32
Fn: 33

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: 32
eval_batch_size: 32
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 577
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Tp	Tn	Fp	Fn
1.7600	1.0	161	1.3736	0.8021	0.8018	0.8017	0.8020	240	218	55	58
1.1108	2.0	322	0.8446	0.8231	0.8230	0.8229	0.8235	243	227	46	55
0.8090	3.0	483	0.7538	0.8406	0.8405	0.8405	0.8412	247	233	40	51
0.6990	4.0	644	0.6882	0.8599	0.8589	0.8625	0.8579	269	222	51	29
0.5774	5.0	805	0.6841	0.8634	0.8625	0.8661	0.8614	270	223	50	28
0.4871	6.0	966	0.7191	0.8616	0.8610	0.8629	0.8602	266	226	47	32
0.4043	7.0	1127	0.7483	0.8704	0.8702	0.8702	0.8702	261	236	37	37
0.3410	8.0	1288	0.8218	0.8651	0.8651	0.8652	0.8659	253	241	32	45
0.3137	9.0	1449	1.0352	0.8634	0.8626	0.8652	0.8618	268	225	48	30
0.2681	10.0	1610	1.0923	0.8792	0.8786	0.8802	0.8779	270	232	41	28
0.2411	11.0	1771	0.9847	0.8651	0.8648	0.8650	0.8647	261	233	40	37
0.1948	12.0	1932	1.3305	0.8599	0.8598	0.8600	0.8607	251	240	33	47
0.1797	13.0	2093	1.3197	0.8564	0.8564	0.8575	0.8578	246	243	30	52
0.1753	14.0	2254	1.3560	0.8616	0.8615	0.8613	0.8618	256	236	37	42
0.1089	15.0	2415	1.5423	0.8634	0.8633	0.8634	0.8641	253	240	33	45
0.1257	16.0	2576	1.3678	0.8757	0.8753	0.8757	0.8750	265	235	38	33
0.1053	17.0	2737	1.4366	0.8792	0.8790	0.8788	0.8793	261	241	32	37
0.0899	18.0	2898	1.7645	0.8476	0.8476	0.8515	0.8500	237	247	26	61
0.0921	19.0	3059	1.6728	0.8599	0.8598	0.8654	0.8627	238	253	20	60
0.0792	20.0	3220	1.5444	0.8704	0.8702	0.8702	0.8702	261	236	37	37
0.0724	21.0	3381	1.6630	0.8669	0.8669	0.8677	0.8682	250	245	28	48
0.0736	22.0	3542	1.6494	0.8669	0.8669	0.8675	0.8680	251	244	29	47
0.0481	23.0	3703	1.6977	0.8722	0.8721	0.8724	0.8731	254	244	29	44
0.0589	24.0	3864	1.6649	0.8722	0.8720	0.8718	0.8721	260	238	35	38
0.0696	25.0	4025	1.8603	0.8616	0.8616	0.8641	0.8636	244	248	25	54
0.0291	26.0	4186	1.7914	0.8757	0.8752	0.8764	0.8746	268	232	41	30
0.0514	27.0	4347	1.8855	0.8757	0.8750	0.8774	0.8741	271	229	44	27
0.0418	28.0	4508	1.7515	0.8774	0.8773	0.8772	0.8778	259	242	31	39
0.0343	29.0	4669	2.0288	0.8669	0.8667	0.8666	0.8668	259	236	37	39
0.0370	30.0	4830	1.7248	0.8862	0.8860	0.8859	0.8860	265	241	32	33

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.8.3
Tokenizers 0.22.2

mhuzaifa5/results

results

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

mhuzaifa5/
results