scenario-KD-PR-MSV-EN-CL-D2_data-en-massive_all_1_144

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: 3.1600
• Accuracy: 0.4441
• F1: 0.4308

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: 44
• 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	3.7790	0.2429	0.1196
No log	0.56	200	3.4643	0.3507	0.2479
No log	0.83	300	3.4223	0.3611	0.2794
No log	1.11	400	3.1902	0.4263	0.3452
2.4418	1.39	500	3.2027	0.4200	0.3472
2.4418	1.67	600	3.5415	0.3617	0.3319
2.4418	1.94	700	3.2162	0.4149	0.3680
2.4418	2.22	800	3.1688	0.4370	0.3957
2.4418	2.5	900	3.3381	0.3933	0.3575
1.4539	2.78	1000	3.1812	0.4304	0.3918
1.4539	3.06	1100	3.4003	0.3960	0.3753
1.4539	3.33	1200	3.4464	0.3814	0.3694
1.4539	3.61	1300	3.2492	0.4227	0.3837
1.4539	3.89	1400	3.4571	0.3881	0.3827
1.2201	4.17	1500	3.4012	0.4094	0.3895
1.2201	4.44	1600	3.3870	0.4077	0.3824
1.2201	4.72	1700	3.5253	0.3935	0.3770
1.2201	5.0	1800	3.4169	0.3977	0.3823
1.2201	5.28	1900	3.4388	0.3916	0.3815
1.0765	5.56	2000	3.3133	0.4234	0.3943
1.0765	5.83	2100	3.2300	0.4373	0.4103
1.0765	6.11	2200	3.2652	0.4225	0.3992
1.0765	6.39	2300	3.3747	0.4110	0.4035
1.0765	6.67	2400	3.7005	0.3663	0.3695
1.0152	6.94	2500	3.4336	0.4022	0.3895
1.0152	7.22	2600	3.3431	0.4204	0.3932
1.0152	7.5	2700	3.3754	0.4078	0.4005
1.0152	7.78	2800	3.3738	0.4162	0.4053
1.0152	8.06	2900	3.7654	0.3669	0.3875
0.9578	8.33	3000	3.4896	0.4073	0.4109
0.9578	8.61	3100	3.6077	0.3918	0.3920
0.9578	8.89	3200	3.2703	0.4377	0.4127
0.9578	9.17	3300	3.3504	0.4135	0.3898
0.9578	9.44	3400	3.3873	0.4152	0.4068
0.9331	9.72	3500	3.4685	0.4081	0.4060
0.9331	10.0	3600	3.1091	0.4561	0.4230
0.9331	10.28	3700	3.3033	0.4234	0.3854
0.9331	10.56	3800	3.5147	0.3876	0.3824
0.9331	10.83	3900	3.4273	0.4152	0.4093
0.9075	11.11	4000	3.2401	0.4372	0.4178
0.9075	11.39	4100	3.4298	0.4138	0.3970
0.9075	11.67	4200	3.2139	0.4418	0.4194
0.9075	11.94	4300	3.3277	0.4243	0.4046
0.9075	12.22	4400	3.3797	0.4164	0.3987
0.8887	12.5	4500	3.3331	0.4188	0.4077
0.8887	12.78	4600	3.2767	0.4292	0.4096
0.8887	13.06	4700	3.3673	0.4233	0.4130
0.8887	13.33	4800	3.3226	0.4161	0.4109
0.8887	13.61	4900	3.3967	0.4083	0.4066
0.8783	13.89	5000	3.3190	0.4196	0.4068
0.8783	14.17	5100	3.4477	0.4006	0.3977
0.8783	14.44	5200	3.2397	0.4366	0.4165
0.8783	14.72	5300	3.2908	0.4250	0.4123
0.8783	15.0	5400	3.3391	0.4159	0.3990
0.8692	15.28	5500	3.5906	0.3802	0.3686
0.8692	15.56	5600	3.3987	0.4087	0.4060
0.8692	15.83	5700	3.2266	0.4383	0.4252
0.8692	16.11	5800	3.3234	0.4192	0.4106
0.8692	16.39	5900	3.2569	0.4254	0.4140
0.8624	16.67	6000	3.2896	0.4222	0.4126
0.8624	16.94	6100	3.3086	0.4211	0.4108
0.8624	17.22	6200	3.3834	0.4123	0.4098
0.8624	17.5	6300	3.0719	0.4570	0.4288
0.8624	17.78	6400	3.2763	0.4259	0.4080
0.856	18.06	6500	3.2298	0.4363	0.4148
0.856	18.33	6600	3.2321	0.4329	0.4140
0.856	18.61	6700	3.2208	0.4371	0.4205
0.856	18.89	6800	3.2499	0.4308	0.4057
0.856	19.17	6900	3.3323	0.4188	0.4131
0.8496	19.44	7000	3.2094	0.4348	0.4212
0.8496	19.72	7100	3.3067	0.4230	0.4173
0.8496	20.0	7200	3.2897	0.4256	0.4083
0.8496	20.28	7300	3.2663	0.4211	0.4042
0.8496	20.56	7400	3.3320	0.4195	0.4090
0.8436	20.83	7500	3.2299	0.4317	0.4144
0.8436	21.11	7600	3.2754	0.4247	0.4121
0.8436	21.39	7700	3.3103	0.4246	0.4267
0.8436	21.67	7800	3.3754	0.4113	0.4081
0.8436	21.94	7900	3.3186	0.4270	0.4273
0.8427	22.22	8000	3.2873	0.4313	0.4212
0.8427	22.5	8100	3.3444	0.4208	0.4162
0.8427	22.78	8200	3.2735	0.4307	0.4194
0.8427	23.06	8300	3.1492	0.4495	0.4267
0.8427	23.33	8400	3.2441	0.4311	0.4202
0.8371	23.61	8500	3.2327	0.4361	0.4236
0.8371	23.89	8600	3.2446	0.4305	0.4226
0.8371	24.17	8700	3.2584	0.4266	0.4210
0.8371	24.44	8800	3.1232	0.4490	0.4311
0.8371	24.72	8900	3.2589	0.4319	0.4228
0.8353	25.0	9000	3.2467	0.4333	0.4207
0.8353	25.28	9100	3.1507	0.4435	0.4270
0.8353	25.56	9200	3.2932	0.4264	0.4241
0.8353	25.83	9300	3.2214	0.4350	0.4247
0.8353	26.11	9400	3.2290	0.4284	0.4214
0.834	26.39	9500	3.1636	0.4452	0.4323
0.834	26.67	9600	3.3134	0.4231	0.4206
0.834	26.94	9700	3.1800	0.4432	0.4269
0.834	27.22	9800	3.2237	0.4354	0.4293
0.834	27.5	9900	3.1967	0.4379	0.4287
0.8309	27.78	10000	3.2085	0.4346	0.4259
0.8309	28.06	10100	3.2214	0.4360	0.4282
0.8309	28.33	10200	3.1793	0.4440	0.4326
0.8309	28.61	10300	3.1608	0.4444	0.4331
0.8309	28.89	10400	3.1759	0.4413	0.4288
0.8312	29.17	10500	3.1611	0.4452	0.4338
0.8312	29.44	10600	3.1461	0.4472	0.4327
0.8312	29.72	10700	3.1562	0.4436	0.4287
0.8312	30.0	10800	3.1600	0.4441	0.4308

Framework versions

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