dfmodel_without_freezing_epoch2

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

• Loss: 0.0269
• Accuracy: 0.9958

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
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1175	0.0126	10	0.0409	0.9954
0.0935	0.0252	20	0.1893	0.9344
0.0568	0.0378	30	0.0232	0.9961
0.0211	0.0503	40	0.0949	0.9849
0.0669	0.0629	50	0.0416	0.9942
0.204	0.0755	60	0.1481	0.9729
0.2098	0.0881	70	0.0513	0.9876
0.0483	0.1007	80	0.0227	0.9954
0.0102	0.1133	90	0.0491	0.9923
0.0715	0.1259	100	0.0621	0.9895
0.1262	0.1385	110	0.0329	0.9947
0.0255	0.1510	120	0.0625	0.9891
0.0391	0.1636	130	0.0259	0.9956
0.0222	0.1762	140	0.0380	0.9931
0.0027	0.1888	150	0.0518	0.9909
0.0556	0.2014	160	0.0349	0.9932
0.0217	0.2140	170	0.0269	0.9953
0.0033	0.2266	180	0.0371	0.9940
0.0224	0.2392	190	0.0418	0.9935
0.0411	0.2517	200	0.0342	0.9945
0.095	0.2643	210	0.0353	0.9951
0.0699	0.2769	220	0.0272	0.9954
0.0443	0.2895	230	0.1465	0.9626
0.0692	0.3021	240	0.0273	0.9950
0.0561	0.3147	250	0.0385	0.9918
0.0326	0.3273	260	0.0292	0.9945
0.0444	0.3398	270	0.0250	0.9959
0.0034	0.3524	280	0.0268	0.9956
0.0373	0.3650	290	0.0262	0.9956
0.0205	0.3776	300	0.0289	0.9953
0.0421	0.3902	310	0.0276	0.9950
0.0254	0.4028	320	0.0252	0.9923
0.0043	0.4154	330	0.0560	0.9873
0.0108	0.4280	340	0.0549	0.9884
0.0671	0.4405	350	0.0487	0.9904
0.011	0.4531	360	0.1066	0.9825
0.2087	0.4657	370	0.1894	0.9534
0.1495	0.4783	380	0.0815	0.9827
0.0112	0.4909	390	0.0308	0.9951
0.0217	0.5035	400	0.0453	0.9932
0.0602	0.5161	410	0.0772	0.9871
0.0809	0.5287	420	0.0539	0.9945
0.0136	0.5412	430	0.0303	0.9951
0.0245	0.5538	440	0.0336	0.9950
0.0211	0.5664	450	0.0335	0.9950
0.051	0.5790	460	0.0431	0.9932
0.0219	0.5916	470	0.0703	0.9879
0.0618	0.6042	480	0.0517	0.9904
0.0365	0.6168	490	0.0760	0.9852
0.0777	0.6294	500	0.0465	0.9913
0.0422	0.6419	510	0.0376	0.9943
0.0466	0.6545	520	0.0984	0.9843
0.0763	0.6671	530	0.0509	0.9906
0.0431	0.6797	540	0.0340	0.9945
0.0214	0.6923	550	0.0300	0.9953
0.0214	0.7049	560	0.0294	0.9954
0.051	0.7175	570	0.0269	0.9959
0.0048	0.7300	580	0.0322	0.9950
0.0212	0.7426	590	0.0422	0.9932
0.021	0.7552	600	0.0488	0.9921
0.0089	0.7678	610	0.0258	0.9961
0.0394	0.7804	620	0.0400	0.9934
0.0541	0.7930	630	0.0489	0.9921
0.0208	0.8056	640	0.0259	0.9961
0.0208	0.8182	650	0.0262	0.9959
0.0027	0.8307	660	0.0268	0.9959
0.0573	0.8433	670	0.0267	0.9959
0.0389	0.8559	680	0.0252	0.9961
0.0032	0.8685	690	0.0259	0.9959
0.0032	0.8811	700	0.0270	0.9958
0.038	0.8937	710	0.0270	0.9958
0.0031	0.9063	720	0.0265	0.9958
0.003	0.9189	730	0.0264	0.9959
0.0028	0.9314	740	0.0265	0.9959
0.0223	0.9440	750	0.0273	0.9958
0.0206	0.9566	760	0.0271	0.9958
0.0564	0.9692	770	0.0270	0.9958
0.0206	0.9818	780	0.0269	0.9958
0.0206	0.9944	790	0.0269	0.9958

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.1.2
• Datasets 2.21.1.dev0
• Tokenizers 0.19.1