wavlm-base_2

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

• Loss: 0.0244
• Accuracy: 0.9966

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 2
• seed: 0
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4872	0.25	100	0.2180	0.8974
0.1571	0.5	200	0.2582	0.9334
0.0644	0.76	300	0.0244	0.9966
0.0553	1.01	400	0.1156	0.9928
0.1108	1.26	500	0.1576	0.9898
0.0849	1.51	600	0.0871	0.9947
0.0635	1.76	700	0.1088	0.9939
0.0504	2.02	800	0.4074	0.9790
0.1075	2.27	900	0.2955	0.9814
0.2387	2.52	1000	0.0651	0.9956
0.3052	2.77	1100	0.2379	0.8974
0.3336	3.02	1200	0.3527	0.8974
0.3322	3.28	1300	0.3307	0.8974
0.3201	3.53	1400	0.3405	0.8974
0.3406	3.78	1500	0.3335	0.8974
0.3475	4.03	1600	0.3341	0.8974
0.3312	4.28	1700	0.3361	0.8974
0.3367	4.54	1800	0.3310	0.8974
0.3284	4.79	1900	0.3339	0.8974
0.3267	5.04	2000	0.3350	0.8974
0.338	5.29	2100	0.3308	0.8974
0.3277	5.55	2200	0.3309	0.8974
0.3294	5.8	2300	0.3313	0.8974
0.3315	6.05	2400	0.3360	0.8974
0.3397	6.3	2500	0.3307	0.8974
0.3318	6.55	2600	0.3359	0.8974
0.3312	6.81	2700	0.3308	0.8974
0.3155	7.06	2800	0.3317	0.8974
0.3304	7.31	2900	0.3362	0.8974
0.338	7.56	3000	0.3342	0.8974
0.3241	7.81	3100	0.3310	0.8974
0.3325	8.07	3200	0.3326	0.8974
0.3202	8.32	3300	0.3345	0.8974
0.3315	8.57	3400	0.3335	0.8974
0.3288	8.82	3500	0.3312	0.8974
0.3371	9.07	3600	0.3401	0.8974
0.3409	9.33	3700	0.3330	0.8974
0.3236	9.58	3800	0.3330	0.8974
0.3224	9.83	3900	0.3321	0.8974
0.3439	10.08	4000	0.3326	0.8974
0.3382	10.33	4100	0.3310	0.8974
0.3307	10.59	4200	0.3382	0.8974
0.3231	10.84	4300	0.3325	0.8974
0.3095	11.09	4400	0.3348	0.8974
0.3442	11.34	4500	0.3327	0.8974
0.3269	11.59	4600	0.3326	0.8974
0.3323	11.85	4700	0.3308	0.8974
0.3313	12.1	4800	0.3308	0.8974
0.3283	12.35	4900	0.3314	0.8974
0.3331	12.6	5000	0.3307	0.8974
0.3317	12.85	5100	0.3344	0.8974
0.3283	13.11	5200	0.3320	0.8974
0.3263	13.36	5300	0.3311	0.8974
0.3421	13.61	5400	0.3307	0.8974
0.3164	13.86	5500	0.3318	0.8974
0.3315	14.11	5600	0.3335	0.8974
0.3415	14.37	5700	0.3315	0.8974
0.3325	14.62	5800	0.3307	0.8974
0.3264	14.87	5900	0.3330	0.8974
0.3223	15.12	6000	0.3307	0.8974
0.3289	15.37	6100	0.3329	0.8974
0.3353	15.63	6200	0.3311	0.8974
0.3246	15.88	6300	0.3311	0.8974
0.3425	16.13	6400	0.3307	0.8974
0.331	16.38	6500	0.3307	0.8974
0.3293	16.64	6600	0.3353	0.8974
0.3249	16.89	6700	0.3339	0.8974
0.3214	17.14	6800	0.3338	0.8974
0.3259	17.39	6900	0.3327	0.8974
0.3408	17.64	7000	0.3318	0.8974
0.3258	17.9	7100	0.3318	0.8974
0.3299	18.15	7200	0.3308	0.8974
0.327	18.4	7300	0.3371	0.8974
0.3317	18.65	7400	0.3308	0.8974
0.3291	18.9	7500	0.3310	0.8974
0.3263	19.16	7600	0.3325	0.8974
0.3223	19.41	7700	0.3346	0.8974
0.3403	19.66	7800	0.3316	0.8974
0.3265	19.91	7900	0.3309	0.8974
0.33	20.16	8000	0.3318	0.8974
0.3488	20.42	8100	0.3313	0.8974
0.3293	20.67	8200	0.3335	0.8974
0.3095	20.92	8300	0.3356	0.8974
0.3366	21.17	8400	0.3332	0.8974
0.317	21.42	8500	0.3338	0.8974
0.3299	21.68	8600	0.3308	0.8974
0.3434	21.93	8700	0.3310	0.8974
0.3208	22.18	8800	0.3309	0.8974
0.3351	22.43	8900	0.3324	0.8974
0.3301	22.68	9000	0.3308	0.8974
0.3196	22.94	9100	0.3330	0.8974
0.3339	23.19	9200	0.3333	0.8974
0.3249	23.44	9300	0.3308	0.8974
0.3247	23.69	9400	0.3338	0.8974
0.3369	23.94	9500	0.3313	0.8974
0.3291	24.2	9600	0.3320	0.8974
0.3307	24.45	9700	0.3309	0.8974
0.3328	24.7	9800	0.3307	0.8974
0.3277	24.95	9900	0.3342	0.8974
0.3278	25.2	10000	0.3310	0.8974
0.3197	25.46	10100	0.3349	0.8974
0.3273	25.71	10200	0.3321	0.8974
0.3345	25.96	10300	0.3312	0.8974
0.3351	26.21	10400	0.3325	0.8974
0.3144	26.47	10500	0.3346	0.8974
0.3361	26.72	10600	0.3311	0.8974
0.3334	26.97	10700	0.3307	0.8974
0.3287	27.22	10800	0.3373	0.8974
0.3374	27.47	10900	0.3307	0.8974
0.3302	27.73	11000	0.3307	0.8974
0.3245	27.98	11100	0.3315	0.8974
0.3353	28.23	11200	0.3335	0.8974
0.3191	28.48	11300	0.3336	0.8974
0.3226	28.73	11400	0.3308	0.8974
0.3384	28.99	11500	0.3322	0.8974
0.3368	29.24	11600	0.3337	0.8974
0.3224	29.49	11700	0.3332	0.8974
0.3224	29.74	11800	0.3318	0.8974
0.3363	29.99	11900	0.3310	0.8974
0.327	30.25	12000	0.3307	0.8974
0.3291	30.5	12100	0.3307	0.8974
0.3369	30.75	12200	0.3322	0.8974
0.3211	31.0	12300	0.3329	0.8974
0.329	31.25	12400	0.3321	0.8974
0.3206	31.51	12500	0.3309	0.8974
0.3339	31.76	12600	0.3332	0.8974
0.3323	32.01	12700	0.3316	0.8974
0.3273	32.26	12800	0.3323	0.8974
0.3362	32.51	12900	0.3307	0.8974
0.3387	32.77	13000	0.3309	0.8974
0.3173	33.02	13100	0.3311	0.8974
0.3291	33.27	13200	0.3309	0.8974
0.3316	33.52	13300	0.3315	0.8974
0.3366	33.77	13400	0.3332	0.8974
0.3115	34.03	13500	0.3383	0.8974
0.3275	34.28	13600	0.3324	0.8974
0.3373	34.53	13700	0.3315	0.8974
0.3247	34.78	13800	0.3313	0.8974
0.3349	35.03	13900	0.3325	0.8974
0.3223	35.29	14000	0.3312	0.8974
0.3321	35.54	14100	0.3308	0.8974
0.3304	35.79	14200	0.3316	0.8974
0.3262	36.04	14300	0.3320	0.8974
0.3239	36.29	14400	0.3317	0.8974
0.3325	36.55	14500	0.3308	0.8974
0.325	36.8	14600	0.3316	0.8974
0.3416	37.05	14700	0.3311	0.8974
0.3226	37.3	14800	0.3309	0.8974
0.3286	37.56	14900	0.3307	0.8974
0.3284	37.81	15000	0.3312	0.8974
0.3298	38.06	15100	0.3326	0.8974
0.3383	38.31	15200	0.3311	0.8974
0.3418	38.56	15300	0.3308	0.8974
0.3123	38.82	15400	0.3311	0.8974
0.3237	39.07	15500	0.3346	0.8974
0.3261	39.32	15600	0.3325	0.8974
0.3269	39.57	15700	0.3312	0.8974
0.3267	39.82	15800	0.3319	0.8974
0.3381	40.08	15900	0.3327	0.8974
0.3238	40.33	16000	0.3326	0.8974
0.3299	40.58	16100	0.3320	0.8974
0.3385	40.83	16200	0.3309	0.8974
0.3268	41.08	16300	0.3322	0.8974
0.3253	41.34	16400	0.3320	0.8974
0.3261	41.59	16500	0.3314	0.8974
0.3362	41.84	16600	0.3324	0.8974
0.3203	42.09	16700	0.3326	0.8974
0.325	42.34	16800	0.3323	0.8974
0.3172	42.6	16900	0.3326	0.8974
0.3361	42.85	17000	0.3308	0.8974
0.3432	43.1	17100	0.3310	0.8974
0.3396	43.35	17200	0.3313	0.8974
0.3163	43.6	17300	0.3328	0.8974
0.3353	43.86	17400	0.3318	0.8974
0.3299	44.11	17500	0.3317	0.8974
0.3213	44.36	17600	0.3319	0.8974
0.3253	44.61	17700	0.3329	0.8974
0.3391	44.86	17800	0.3322	0.8974
0.3179	45.12	17900	0.3330	0.8974
0.3348	45.37	18000	0.3321	0.8974
0.3116	45.62	18100	0.3326	0.8974
0.3334	45.87	18200	0.3322	0.8974
0.3401	46.12	18300	0.3315	0.8974
0.3381	46.38	18400	0.3311	0.8974
0.3154	46.63	18500	0.3327	0.8974
0.3348	46.88	18600	0.3322	0.8974
0.3285	47.13	18700	0.3325	0.8974
0.3256	47.39	18800	0.3329	0.8974
0.3389	47.64	18900	0.3325	0.8974
0.3288	47.89	19000	0.3327	0.8974
0.3172	48.14	19100	0.3327	0.8974
0.3211	48.39	19200	0.3325	0.8974
0.3348	48.65	19300	0.3325	0.8974
0.3327	48.9	19400	0.3326	0.8974
0.3341	49.15	19500	0.3326	0.8974
0.3344	49.4	19600	0.3325	0.8974
0.3207	49.65	19700	0.3326	0.8974
0.3299	49.91	19800	0.3326	0.8974

Framework versions

• Transformers 4.34.0.dev0
• Pytorch 2.0.0.post302
• Datasets 2.14.5
• Tokenizers 0.13.3