wav2vec2-5Class-train-test-finetune-V7

This model is a fine-tuned version of anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test on the anderloh/Master5Class dataset. It achieves the following results on the evaluation set:

• Loss: 0.9663
• Accuracy: 0.6538

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: 3e-05
• train_batch_size: 128
• eval_batch_size: 128
• seed: 0
• gradient_accumulation_steps: 4
• total_train_batch_size: 512
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 250.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.92	3	1.6026	0.1608
No log	1.85	6	1.6024	0.1608
No log	2.77	9	1.6021	0.1608
No log	4.0	13	1.6014	0.1608
No log	4.92	16	1.6008	0.1608
No log	5.85	19	1.6001	0.1608
No log	6.77	22	1.5992	0.1608
No log	8.0	26	1.5978	0.1748
No log	8.92	29	1.5965	0.1888
No log	9.85	32	1.5952	0.2098
No log	10.77	35	1.5938	0.2273
No log	12.0	39	1.5916	0.2343
No log	12.92	42	1.5899	0.2692
No log	13.85	45	1.5880	0.2727
No log	14.77	48	1.5860	0.3077
No log	16.0	52	1.5833	0.3566
No log	16.92	55	1.5811	0.3881
No log	17.85	58	1.5788	0.3811
No log	18.77	61	1.5764	0.3671
No log	20.0	65	1.5731	0.3497
No log	20.92	68	1.5702	0.3287
No log	21.85	71	1.5672	0.3252
No log	22.77	74	1.5641	0.3147
No log	24.0	78	1.5597	0.3112
No log	24.92	81	1.5564	0.3077
No log	25.85	84	1.5532	0.3042
No log	26.77	87	1.5499	0.2937
No log	28.0	91	1.5454	0.2902
No log	28.92	94	1.5419	0.2867
No log	29.85	97	1.5383	0.2832
1.5563	30.77	100	1.5349	0.2762
1.5563	32.0	104	1.5304	0.2797
1.5563	32.92	107	1.5273	0.2762
1.5563	33.85	110	1.5247	0.2657
1.5563	34.77	113	1.5223	0.2517
1.5563	36.0	117	1.5194	0.2483
1.5563	36.92	120	1.5178	0.2413
1.5563	37.85	123	1.5168	0.2378
1.5563	38.77	126	1.5162	0.2448
1.5563	40.0	130	1.5162	0.2448
1.5563	40.92	133	1.5167	0.2483
1.5563	41.85	136	1.5181	0.2483
1.5563	42.77	139	1.5203	0.2587
1.5563	44.0	143	1.5227	0.2692
1.5563	44.92	146	1.5243	0.2832
1.5563	45.85	149	1.5239	0.2797
1.5563	46.77	152	1.5224	0.3007
1.5563	48.0	156	1.5170	0.3077
1.5563	48.92	159	1.5103	0.3287
1.5563	49.85	162	1.5032	0.3497
1.5563	50.77	165	1.4959	0.3601
1.5563	52.0	169	1.4857	0.3636
1.5563	52.92	172	1.4788	0.3671
1.5563	53.85	175	1.4713	0.3741
1.5563	54.77	178	1.4642	0.3811
1.5563	56.0	182	1.4553	0.3881
1.5563	56.92	185	1.4481	0.3986
1.5563	57.85	188	1.4421	0.4021
1.5563	58.77	191	1.4357	0.4126
1.5563	60.0	195	1.4284	0.4196
1.5563	60.92	198	1.4218	0.4196
1.3138	61.85	201	1.4167	0.4301
1.3138	62.77	204	1.4091	0.4301
1.3138	64.0	208	1.3995	0.4371
1.3138	64.92	211	1.3911	0.4406
1.3138	65.85	214	1.3825	0.4336
1.3138	66.77	217	1.3735	0.4441
1.3138	68.0	221	1.3632	0.4476
1.3138	68.92	224	1.3556	0.4510
1.3138	69.85	227	1.3492	0.4510
1.3138	70.77	230	1.3441	0.4510
1.3138	72.0	234	1.3352	0.4580
1.3138	72.92	237	1.3269	0.4615
1.3138	73.85	240	1.3186	0.4755
1.3138	74.77	243	1.3105	0.4755
1.3138	76.0	247	1.2992	0.4790
1.3138	76.92	250	1.2896	0.4825
1.3138	77.85	253	1.2797	0.4825
1.3138	78.77	256	1.2707	0.4860
1.3138	80.0	260	1.2587	0.4930
1.3138	80.92	263	1.2494	0.4930
1.3138	81.85	266	1.2407	0.4930
1.3138	82.77	269	1.2314	0.5105
1.3138	84.0	273	1.2205	0.5140
1.3138	84.92	276	1.2124	0.5210
1.3138	85.85	279	1.2043	0.5315
1.3138	86.77	282	1.1973	0.5350
1.3138	88.0	286	1.1870	0.5524
1.3138	88.92	289	1.1788	0.5629
1.3138	89.85	292	1.1700	0.5629
1.3138	90.77	295	1.1613	0.5699
1.3138	92.0	299	1.1498	0.5839
1.047	92.92	302	1.1411	0.5874
1.047	93.85	305	1.1330	0.5944
1.047	94.77	308	1.1261	0.5944
1.047	96.0	312	1.1161	0.6014
1.047	96.92	315	1.1084	0.6014
1.047	97.85	318	1.1003	0.6049
1.047	98.77	321	1.0926	0.6049
1.047	100.0	325	1.0821	0.6084
1.047	100.92	328	1.0754	0.6084
1.047	101.85	331	1.0690	0.6084
1.047	102.77	334	1.0637	0.6154
1.047	104.0	338	1.0549	0.6189
1.047	104.92	341	1.0478	0.6224
1.047	105.85	344	1.0420	0.6259
1.047	106.77	347	1.0370	0.6294
1.047	108.0	351	1.0308	0.6294
1.047	108.92	354	1.0263	0.6259
1.047	109.85	357	1.0231	0.6259
1.047	110.77	360	1.0204	0.6329
1.047	112.0	364	1.0167	0.6294
1.047	112.92	367	1.0145	0.6294
1.047	113.85	370	1.0119	0.6329
1.047	114.77	373	1.0077	0.6294
1.047	116.0	377	1.0012	0.6364
1.047	116.92	380	0.9975	0.6364
1.047	117.85	383	0.9938	0.6364
1.047	118.77	386	0.9913	0.6399
1.047	120.0	390	0.9886	0.6469
1.047	120.92	393	0.9870	0.6469
1.047	121.85	396	0.9861	0.6399
1.047	122.77	399	0.9857	0.6434
0.8183	124.0	403	0.9855	0.6399
0.8183	124.92	406	0.9864	0.6399
0.8183	125.85	409	0.9857	0.6399
0.8183	126.77	412	0.9818	0.6399
0.8183	128.0	416	0.9765	0.6399
0.8183	128.92	419	0.9740	0.6399
0.8183	129.85	422	0.9737	0.6434
0.8183	130.77	425	0.9754	0.6469
0.8183	132.0	429	0.9753	0.6469
0.8183	132.92	432	0.9740	0.6469
0.8183	133.85	435	0.9710	0.6469
0.8183	134.77	438	0.9686	0.6469
0.8183	136.0	442	0.9671	0.6469
0.8183	136.92	445	0.9669	0.6434
0.8183	137.85	448	0.9659	0.6399
0.8183	138.77	451	0.9662	0.6434
0.8183	140.0	455	0.9674	0.6434
0.8183	140.92	458	0.9694	0.6399
0.8183	141.85	461	0.9716	0.6469
0.8183	142.77	464	0.9739	0.6434
0.8183	144.0	468	0.9712	0.6469
0.8183	144.92	471	0.9670	0.6434
0.8183	145.85	474	0.9637	0.6434
0.8183	146.77	477	0.9625	0.6469
0.8183	148.0	481	0.9634	0.6469
0.8183	148.92	484	0.9659	0.6469
0.8183	149.85	487	0.9663	0.6469
0.8183	150.77	490	0.9649	0.6503
0.8183	152.0	494	0.9655	0.6503
0.8183	152.92	497	0.9648	0.6503
0.7321	153.85	500	0.9638	0.6503
0.7321	154.77	503	0.9631	0.6503
0.7321	156.0	507	0.9647	0.6503
0.7321	156.92	510	0.9653	0.6503
0.7321	157.85	513	0.9662	0.6503
0.7321	158.77	516	0.9679	0.6503
0.7321	160.0	520	0.9675	0.6503
0.7321	160.92	523	0.9664	0.6503
0.7321	161.85	526	0.9655	0.6503
0.7321	162.77	529	0.9642	0.6503
0.7321	164.0	533	0.9635	0.6503
0.7321	164.92	536	0.9633	0.6503
0.7321	165.85	539	0.9645	0.6503
0.7321	166.77	542	0.9649	0.6503
0.7321	168.0	546	0.9651	0.6503
0.7321	168.92	549	0.9657	0.6503
0.7321	169.85	552	0.9663	0.6538
0.7321	170.77	555	0.9653	0.6503
0.7321	172.0	559	0.9638	0.6503
0.7321	172.92	562	0.9616	0.6503
0.7321	173.85	565	0.9601	0.6503
0.7321	174.77	568	0.9610	0.6538
0.7321	176.0	572	0.9630	0.6503
0.7321	176.92	575	0.9633	0.6503
0.7321	177.85	578	0.9646	0.6503
0.7321	178.77	581	0.9655	0.6503
0.7321	180.0	585	0.9673	0.6503
0.7321	180.92	588	0.9680	0.6503
0.7321	181.85	591	0.9687	0.6503
0.7321	182.77	594	0.9692	0.6503
0.7321	184.0	598	0.9684	0.6503
0.6941	184.92	601	0.9677	0.6503
0.6941	185.85	604	0.9674	0.6503
0.6941	186.77	607	0.9671	0.6503
0.6941	188.0	611	0.9670	0.6503
0.6941	188.92	614	0.9662	0.6503
0.6941	189.85	617	0.9653	0.6503
0.6941	190.77	620	0.9645	0.6503
0.6941	192.0	624	0.9648	0.6503
0.6941	192.92	627	0.9652	0.6503
0.6941	193.85	630	0.9663	0.6503
0.6941	194.77	633	0.9662	0.6503
0.6941	196.0	637	0.9665	0.6503
0.6941	196.92	640	0.9668	0.6503
0.6941	197.85	643	0.9669	0.6469
0.6941	198.77	646	0.9674	0.6434
0.6941	200.0	650	0.9669	0.6469
0.6941	200.92	653	0.9672	0.6469
0.6941	201.85	656	0.9671	0.6469
0.6941	202.77	659	0.9673	0.6503
0.6941	204.0	663	0.9666	0.6503
0.6941	204.92	666	0.9660	0.6503
0.6941	205.85	669	0.9656	0.6503
0.6941	206.77	672	0.9651	0.6503
0.6941	208.0	676	0.9661	0.6503
0.6941	208.92	679	0.9667	0.6503
0.6941	209.85	682	0.9668	0.6503
0.6941	210.77	685	0.9669	0.6503
0.6941	212.0	689	0.9665	0.6503
0.6941	212.92	692	0.9665	0.6503
0.6941	213.85	695	0.9664	0.6503
0.6941	214.77	698	0.9663	0.6503
0.6696	216.0	702	0.9666	0.6503
0.6696	216.92	705	0.9667	0.6503
0.6696	217.85	708	0.9665	0.6503
0.6696	218.77	711	0.9663	0.6503
0.6696	220.0	715	0.9661	0.6503
0.6696	220.92	718	0.9661	0.6503
0.6696	221.85	721	0.9662	0.6503
0.6696	222.77	724	0.9664	0.6503
0.6696	224.0	728	0.9664	0.6503
0.6696	224.92	731	0.9664	0.6503
0.6696	225.85	734	0.9666	0.6503
0.6696	226.77	737	0.9666	0.6503
0.6696	228.0	741	0.9665	0.6503
0.6696	228.92	744	0.9666	0.6503
0.6696	229.85	747	0.9666	0.6503
0.6696	230.77	750	0.9666	0.6503

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2