scenario-KD-PR-MSV-EN-CL-D2_data-en-massive_all_1_155

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: 3.1707
• Accuracy: 0.4471
• F1: 0.4385

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: 55
• 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	3.7123	0.2625	0.1148
No log	0.56	200	3.5358	0.3390	0.2394
No log	0.83	300	3.3412	0.3922	0.3123
No log	1.11	400	3.4008	0.3772	0.3110
2.4588	1.39	500	3.2536	0.4111	0.3405
2.4588	1.67	600	3.2840	0.4097	0.3563
2.4588	1.94	700	3.1644	0.4307	0.3710
2.4588	2.22	800	3.1380	0.4364	0.3872
2.4588	2.5	900	3.2618	0.4180	0.3833
1.4472	2.78	1000	3.1488	0.4300	0.3861
1.4472	3.06	1100	3.1174	0.4488	0.4097
1.4472	3.33	1200	3.1442	0.4481	0.4091
1.4472	3.61	1300	3.3004	0.4111	0.3841
1.4472	3.89	1400	3.3097	0.4010	0.3802
1.2115	4.17	1500	3.3768	0.4052	0.3846
1.2115	4.44	1600	3.3732	0.4077	0.3919
1.2115	4.72	1700	3.3767	0.4125	0.3980
1.2115	5.0	1800	3.3720	0.4195	0.3896
1.2115	5.28	1900	3.3548	0.4147	0.3941
1.0796	5.56	2000	3.6313	0.3784	0.3743
1.0796	5.83	2100	3.2951	0.4186	0.3930
1.0796	6.11	2200	3.2913	0.4267	0.3988
1.0796	6.39	2300	3.1985	0.4357	0.4107
1.0796	6.67	2400	3.4566	0.3963	0.3932
1.01	6.94	2500	3.3982	0.4094	0.4000
1.01	7.22	2600	3.2082	0.4343	0.3960
1.01	7.5	2700	3.3417	0.4153	0.4042
1.01	7.78	2800	3.2235	0.4332	0.4025
1.01	8.06	2900	3.2782	0.4264	0.4084
0.9549	8.33	3000	3.3575	0.4120	0.3981
0.9549	8.61	3100	3.2973	0.4231	0.4094
0.9549	8.89	3200	3.3885	0.4086	0.3964
0.9549	9.17	3300	3.3343	0.4185	0.4105
0.9549	9.44	3400	3.3463	0.4177	0.4090
0.9242	9.72	3500	3.2792	0.4287	0.4131
0.9242	10.0	3600	3.3775	0.4110	0.4032
0.9242	10.28	3700	3.3542	0.4210	0.4144
0.9242	10.56	3800	3.2521	0.4345	0.4194
0.9242	10.83	3900	3.3305	0.4234	0.4080
0.9048	11.11	4000	3.5624	0.3953	0.4027
0.9048	11.39	4100	3.5235	0.3909	0.3993
0.9048	11.67	4200	3.3855	0.4153	0.3993
0.9048	11.94	4300	3.4062	0.4026	0.3931
0.9048	12.22	4400	3.3170	0.4241	0.4154
0.8863	12.5	4500	3.2977	0.4287	0.4153
0.8863	12.78	4600	3.5425	0.3886	0.3976
0.8863	13.06	4700	3.4107	0.4081	0.3997
0.8863	13.33	4800	3.4859	0.4068	0.4001
0.8863	13.61	4900	3.4532	0.4104	0.4136
0.8764	13.89	5000	3.4220	0.4105	0.4108
0.8764	14.17	5100	3.4623	0.4121	0.4171
0.8764	14.44	5200	3.4474	0.4026	0.4045
0.8764	14.72	5300	3.3895	0.4177	0.4161
0.8764	15.0	5400	3.2330	0.4416	0.4290
0.8673	15.28	5500	3.3441	0.4180	0.4133
0.8673	15.56	5600	3.3918	0.4172	0.4166
0.8673	15.83	5700	3.2297	0.4393	0.4187
0.8673	16.11	5800	3.2193	0.4418	0.4287
0.8673	16.39	5900	3.4330	0.4124	0.4190
0.8595	16.67	6000	3.2666	0.4351	0.4289
0.8595	16.94	6100	3.1744	0.4529	0.4413
0.8595	17.22	6200	3.4892	0.4036	0.4129
0.8595	17.5	6300	3.3720	0.4189	0.4215
0.8595	17.78	6400	3.3287	0.4213	0.4179
0.8523	18.06	6500	3.4352	0.4089	0.4124
0.8523	18.33	6600	3.2985	0.4255	0.4233
0.8523	18.61	6700	3.2437	0.4355	0.4274
0.8523	18.89	6800	3.3418	0.4200	0.4139
0.8523	19.17	6900	3.2395	0.4346	0.4342
0.8491	19.44	7000	3.2704	0.4283	0.4115
0.8491	19.72	7100	3.2447	0.4378	0.4256
0.8491	20.0	7200	3.2999	0.4281	0.4272
0.8491	20.28	7300	3.2139	0.4346	0.4371
0.8491	20.56	7400	3.3605	0.4190	0.4138
0.8447	20.83	7500	3.3631	0.4216	0.4171
0.8447	21.11	7600	3.2030	0.4422	0.4276
0.8447	21.39	7700	3.3002	0.4257	0.4256
0.8447	21.67	7800	3.3028	0.4275	0.4296
0.8447	21.94	7900	3.2922	0.4281	0.4225
0.8412	22.22	8000	3.1588	0.4464	0.4339
0.8412	22.5	8100	3.2553	0.4367	0.4307
0.8412	22.78	8200	3.1886	0.4433	0.4365
0.8412	23.06	8300	3.3312	0.4245	0.4270
0.8412	23.33	8400	3.2022	0.4447	0.4338
0.8389	23.61	8500	3.3122	0.4214	0.4179
0.8389	23.89	8600	3.0719	0.4621	0.4380
0.8389	24.17	8700	3.2395	0.4386	0.4262
0.8389	24.44	8800	3.2242	0.4364	0.4339
0.8389	24.72	8900	3.3582	0.4201	0.4197
0.8333	25.0	9000	3.1279	0.4537	0.4377
0.8333	25.28	9100	3.1643	0.4458	0.4361
0.8333	25.56	9200	3.1543	0.4503	0.4358
0.8333	25.83	9300	3.2963	0.4251	0.4243
0.8333	26.11	9400	3.0952	0.4567	0.4375
0.8326	26.39	9500	3.2282	0.4385	0.4347
0.8326	26.67	9600	3.1402	0.4512	0.4411
0.8326	26.94	9700	3.2730	0.4321	0.4331
0.8326	27.22	9800	3.2324	0.4393	0.4374
0.8326	27.5	9900	3.2203	0.4418	0.4380
0.8304	27.78	10000	3.1916	0.4444	0.4388
0.8304	28.06	10100	3.2167	0.4395	0.4370
0.8304	28.33	10200	3.1614	0.4477	0.4386
0.8304	28.61	10300	3.1688	0.4483	0.4361
0.8304	28.89	10400	3.1459	0.4511	0.4427
0.8306	29.17	10500	3.1191	0.4546	0.4433
0.8306	29.44	10600	3.1557	0.4493	0.4417
0.8306	29.72	10700	3.1777	0.4456	0.4385
0.8306	30.0	10800	3.1707	0.4471	0.4385

Framework versions

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