scenario-KD-PO-MSV-EN-CL-D2_data-en-massive_all_1_166

This model is a fine-tuned version of haryoaw/scenario-MDBT-TCR_data-cl-massive_all_1_1 on the massive dataset. It achieves the following results on the evaluation set:

• Loss: 11.8204
• Accuracy: 0.4326
• F1: 0.4237

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: 66
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.28	100	12.2965	0.1486	0.0360
No log	0.56	200	11.1275	0.2774	0.1914
No log	0.83	300	9.7545	0.3648	0.2826
No log	1.11	400	10.2753	0.3682	0.2990
7.0012	1.39	500	9.6889	0.4057	0.3417
7.0012	1.67	600	10.7424	0.3645	0.3219
7.0012	1.94	700	9.8959	0.3842	0.3576
7.0012	2.22	800	10.1838	0.3977	0.3483
7.0012	2.5	900	9.9592	0.4001	0.3505
2.7375	2.78	1000	10.7031	0.3922	0.3589
2.7375	3.06	1100	10.7897	0.4041	0.3652
2.7375	3.33	1200	11.5339	0.3874	0.3567
2.7375	3.61	1300	11.7589	0.3932	0.3627
2.7375	3.89	1400	11.0529	0.4019	0.3780
1.7199	4.17	1500	10.9248	0.4050	0.3705
1.7199	4.44	1600	10.7844	0.4199	0.3904
1.7199	4.72	1700	12.4353	0.3804	0.3622
1.7199	5.0	1800	11.5307	0.4034	0.3608
1.7199	5.28	1900	13.0146	0.3763	0.3616
1.212	5.56	2000	11.9432	0.4010	0.3776
1.212	5.83	2100	11.7054	0.4123	0.3901
1.212	6.11	2200	12.1679	0.4034	0.3850
1.212	6.39	2300	12.1939	0.4017	0.3773
1.212	6.67	2400	12.1474	0.3949	0.3746
0.964	6.94	2500	14.2131	0.3647	0.3610
0.964	7.22	2600	12.9514	0.3853	0.3658
0.964	7.5	2700	13.2805	0.3796	0.3803
0.964	7.78	2800	13.4053	0.3774	0.3665
0.964	8.06	2900	12.0507	0.4094	0.3925
0.7594	8.33	3000	13.1875	0.3887	0.3803
0.7594	8.61	3100	11.5916	0.4226	0.3927
0.7594	8.89	3200	12.2110	0.4116	0.3937
0.7594	9.17	3300	11.7099	0.4220	0.3951
0.7594	9.44	3400	11.7371	0.4216	0.3980
0.6613	9.72	3500	11.4222	0.4280	0.3983
0.6613	10.0	3600	11.2944	0.4180	0.3822
0.6613	10.28	3700	11.7026	0.4208	0.4010
0.6613	10.56	3800	11.9702	0.4151	0.3976
0.6613	10.83	3900	12.2859	0.4077	0.3880
0.5936	11.11	4000	11.9756	0.4186	0.4037
0.5936	11.39	4100	11.6912	0.4249	0.4040
0.5936	11.67	4200	12.2043	0.4112	0.3992
0.5936	11.94	4300	12.7177	0.4039	0.3876
0.5936	12.22	4400	12.1222	0.4109	0.3961
0.5421	12.5	4500	11.8550	0.4206	0.4051
0.5421	12.78	4600	11.1417	0.4309	0.4040
0.5421	13.06	4700	12.3230	0.4044	0.3924
0.5421	13.33	4800	12.4166	0.4129	0.3943
0.5421	13.61	4900	12.9257	0.4036	0.3978
0.4976	13.89	5000	12.6773	0.4075	0.3968
0.4976	14.17	5100	12.6017	0.4085	0.4016
0.4976	14.44	5200	12.4200	0.4099	0.4000
0.4976	14.72	5300	12.0356	0.4200	0.4092
0.4976	15.0	5400	12.1298	0.4167	0.4095
0.4667	15.28	5500	11.5741	0.4278	0.4119
0.4667	15.56	5600	12.9944	0.4014	0.3989
0.4667	15.83	5700	11.7750	0.4280	0.4077
0.4667	16.11	5800	12.1071	0.4184	0.4099
0.4667	16.39	5900	11.9469	0.4237	0.4116
0.4461	16.67	6000	12.2299	0.4186	0.4107
0.4461	16.94	6100	12.0657	0.4255	0.4227
0.4461	17.22	6200	12.2029	0.4170	0.4106
0.4461	17.5	6300	12.0376	0.4221	0.4129
0.4461	17.78	6400	12.6844	0.4120	0.4075
0.4236	18.06	6500	12.5797	0.4098	0.3997
0.4236	18.33	6600	12.1018	0.4227	0.4031
0.4236	18.61	6700	12.9408	0.4040	0.4027
0.4236	18.89	6800	11.1018	0.4453	0.4273
0.4236	19.17	6900	12.1794	0.4204	0.4111
0.4093	19.44	7000	11.5382	0.4392	0.4240
0.4093	19.72	7100	11.9820	0.4252	0.4146
0.4093	20.0	7200	12.4871	0.4209	0.4123
0.4093	20.28	7300	12.0761	0.4247	0.4128
0.4093	20.56	7400	11.4837	0.4323	0.4223
0.392	20.83	7500	11.7228	0.4342	0.4222
0.392	21.11	7600	11.3621	0.4385	0.4246
0.392	21.39	7700	12.3272	0.4202	0.4135
0.392	21.67	7800	12.4137	0.4155	0.4057
0.392	21.94	7900	11.6855	0.4316	0.4180
0.3807	22.22	8000	12.1200	0.4246	0.4167
0.3807	22.5	8100	11.9334	0.4258	0.4165
0.3807	22.78	8200	11.8472	0.4297	0.4160
0.3807	23.06	8300	11.6667	0.4345	0.4178
0.3807	23.33	8400	11.9132	0.4274	0.4166
0.3743	23.61	8500	11.3469	0.4442	0.4287
0.3743	23.89	8600	11.6652	0.4331	0.4168
0.3743	24.17	8700	11.6161	0.4324	0.4180
0.3743	24.44	8800	12.0137	0.4285	0.4181
0.3743	24.72	8900	11.7983	0.4372	0.4251
0.3636	25.0	9000	11.3561	0.4424	0.4284
0.3636	25.28	9100	11.8990	0.4314	0.4176
0.3636	25.56	9200	11.9253	0.4298	0.4193
0.3636	25.83	9300	11.4007	0.4448	0.4257
0.3636	26.11	9400	11.4205	0.4419	0.4288
0.3547	26.39	9500	11.8100	0.4335	0.4227
0.3547	26.67	9600	11.8240	0.4320	0.4195
0.3547	26.94	9700	11.9142	0.4320	0.4245
0.3547	27.22	9800	11.6972	0.4372	0.4264
0.3547	27.5	9900	11.8331	0.4324	0.4225
0.3523	27.78	10000	11.8018	0.4351	0.4254
0.3523	28.06	10100	11.9065	0.4312	0.4233
0.3523	28.33	10200	11.6883	0.4376	0.4262
0.3523	28.61	10300	12.0389	0.4294	0.4216
0.3523	28.89	10400	11.8371	0.4308	0.4213
0.347	29.17	10500	11.8024	0.4337	0.4219
0.347	29.44	10600	11.8341	0.4331	0.4225
0.347	29.72	10700	11.8533	0.4324	0.4210
0.347	30.0	10800	11.8204	0.4326	0.4237

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3