qa-persian-albert-fa-zwnj-base-v2

This model is a fine-tuned version of makhataei/qa-persian-albert-fa-zwnj-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.9720

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: 7.8125e-10
• train_batch_size: 14
• eval_batch_size: 14
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss
6.0547	1.0	9	5.9759
6.1559	2.0	18	5.9758
6.1407	3.0	27	5.9757
6.2374	4.0	36	5.9756
6.1228	5.0	45	5.9755
6.0274	6.0	54	5.9754
6.175	7.0	63	5.9753
6.1089	8.0	72	5.9752
5.9715	9.0	81	5.9751
6.0709	10.0	90	5.9750
6.0287	11.0	99	5.9749
6.0229	12.0	108	5.9748
6.0991	13.0	117	5.9747
6.177	14.0	126	5.9746
6.0944	15.0	135	5.9745
6.1154	16.0	144	5.9744
6.1325	17.0	153	5.9743
6.032	18.0	162	5.9742
6.0753	19.0	171	5.9741
6.0302	20.0	180	5.9741
6.0986	21.0	189	5.9740
6.0651	22.0	198	5.9739
6.0975	23.0	207	5.9738
6.1147	24.0	216	5.9738
6.0799	25.0	225	5.9737
6.1036	26.0	234	5.9736
6.0992	27.0	243	5.9735
6.0736	28.0	252	5.9735
6.1135	29.0	261	5.9734
6.0838	30.0	270	5.9734
6.166	31.0	279	5.9733
6.1658	32.0	288	5.9732
6.0511	33.0	297	5.9732
6.0808	34.0	306	5.9731
6.3259	35.0	315	5.9731
6.5127	36.0	324	5.9730
6.1636	37.0	333	5.9730
6.0474	38.0	342	5.9729
6.0181	39.0	351	5.9729
6.2572	40.0	360	5.9729
6.1826	41.0	369	5.9728
6.0872	42.0	378	5.9728
6.5487	43.0	387	5.9728
6.0156	44.0	396	5.9727
6.0935	45.0	405	5.9727
6.1265	46.0	414	5.9727
6.0836	47.0	423	5.9726
6.3812	48.0	432	5.9726
6.1809	49.0	441	5.9726
6.189	50.0	450	5.9725
6.3619	51.0	459	5.9725
6.1027	52.0	468	5.9725
6.0734	53.0	477	5.9724
6.1045	54.0	486	5.9724
6.0697	55.0	495	5.9724
6.3692	56.0	504	5.9724
6.0849	57.0	513	5.9723
6.0133	58.0	522	5.9723
6.1366	59.0	531	5.9723
6.0863	60.0	540	5.9723
6.0564	61.0	549	5.9723
6.1513	62.0	558	5.9722
6.1085	63.0	567	5.9722
6.071	64.0	576	5.9722
6.0008	65.0	585	5.9722
6.1798	66.0	594	5.9722
6.14	67.0	603	5.9722
6.0828	68.0	612	5.9721
6.4297	69.0	621	5.9721
6.0881	70.0	630	5.9721
6.5834	71.0	639	5.9721
6.3729	72.0	648	5.9721
6.214	73.0	657	5.9721
6.1623	74.0	666	5.9721
6.7243	75.0	675	5.9721
6.0237	76.0	684	5.9721
6.099	77.0	693	5.9721
6.0	78.0	702	5.9721
6.0503	79.0	711	5.9721
6.3524	80.0	720	5.9721
6.4079	81.0	729	5.9720
6.0968	82.0	738	5.9720
5.9891	83.0	747	5.9720
6.2035	84.0	756	5.9720
6.0786	85.0	765	5.9720
6.0148	86.0	774	5.9720
6.2646	87.0	783	5.9720
6.1343	88.0	792	5.9720
6.0886	89.0	801	5.9720
6.0741	90.0	810	5.9720
6.0167	91.0	819	5.9720
6.0372	92.0	828	5.9720
6.1389	93.0	837	5.9720
6.2138	94.0	846	5.9720
6.0838	95.0	855	5.9720
6.0839	96.0	864	5.9720
6.3859	97.0	873	5.9720
6.1075	98.0	882	5.9720
6.0148	99.0	891	5.9720
6.1987	100.0	900	5.9720

Framework versions

• Transformers 4.35.2
• Pytorch 2.0.1+cu117
• Datasets 2.15.0
• Tokenizers 0.15.0