scenario-KD-PO-MSV-D2_data-AmazonScience_massive_all_1_166sss

This model is a fine-tuned version of haryoaw/scenario-MDBT-TCR_data-AmazonScience_massive_all_1_1 on the massive dataset. It achieves the following results on the evaluation set:

• Loss: 1.6978
• Accuracy: 0.8427
• F1: 0.8234

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
5.0835	0.2672	5000	4.9255	0.5565	0.4181
3.7578	0.5344	10000	3.6958	0.6645	0.5564
3.0382	0.8017	15000	3.1257	0.7175	0.6427
2.5632	1.0689	20000	2.7946	0.7482	0.6849
2.3804	1.3361	25000	2.5602	0.7687	0.7030
2.2046	1.6033	30000	2.4410	0.7784	0.7154
2.0993	1.8706	35000	2.3297	0.7855	0.7332
1.822	2.1378	40000	2.2600	0.7953	0.7521
1.6997	2.4050	45000	2.2103	0.7996	0.7579
1.6912	2.6722	50000	2.1752	0.8015	0.7650
1.6155	2.9394	55000	2.1519	0.8039	0.7582
1.4369	3.2067	60000	2.1148	0.8090	0.7690
1.3745	3.4739	65000	2.0965	0.8107	0.7741
1.3834	3.7411	70000	2.0734	0.8132	0.7784
1.3345	4.0083	75000	2.0164	0.8155	0.7840
1.2115	4.2756	80000	2.0363	0.8154	0.7821
1.179	4.5428	85000	2.0027	0.8204	0.7867
1.1673	4.8100	90000	2.0222	0.8202	0.7880
1.0291	5.0772	95000	2.0115	0.8203	0.7892
1.0227	5.3444	100000	2.0039	0.8203	0.7891
1.0341	5.6117	105000	1.9760	0.8219	0.7888
1.0222	5.8789	110000	1.9320	0.8253	0.7943
0.9092	6.1461	115000	1.9920	0.8234	0.7902
0.914	6.4133	120000	1.9570	0.8265	0.7994
0.9529	6.6806	125000	1.9528	0.8256	0.7974
0.938	6.9478	130000	1.9069	0.8287	0.7987
0.827	7.2150	135000	1.9705	0.8251	0.7972
0.8051	7.4822	140000	1.9544	0.8287	0.8022
0.8191	7.7495	145000	1.9406	0.8263	0.8013
0.7775	8.0167	150000	1.9386	0.8285	0.8024
0.7684	8.2839	155000	1.9467	0.8256	0.7969
0.7416	8.5511	160000	1.9151	0.8314	0.8076
0.7478	8.8183	165000	1.9375	0.8280	0.8036
0.6885	9.0856	170000	1.8948	0.8311	0.8060
0.6906	9.3528	175000	1.9180	0.8299	0.8050
0.7034	9.6200	180000	1.8992	0.8317	0.8082
0.6998	9.8872	185000	1.9034	0.8317	0.8094
0.6234	10.1545	190000	1.9070	0.8340	0.8128
0.6493	10.4217	195000	1.9198	0.8289	0.8062
0.6292	10.6889	200000	1.8976	0.8307	0.8112
0.6412	10.9561	205000	1.8986	0.8310	0.8066
0.5954	11.2233	210000	1.9010	0.8326	0.8131
0.6134	11.4906	215000	1.8906	0.8317	0.8091
0.6156	11.7578	220000	1.8856	0.8324	0.8088
0.5623	12.0250	225000	1.8711	0.8339	0.8137
0.5677	12.2922	230000	1.8724	0.8350	0.8119
0.5585	12.5595	235000	1.8606	0.8355	0.8143
0.5789	12.8267	240000	1.8650	0.8343	0.8125
0.5306	13.0939	245000	1.8626	0.8340	0.8097
0.5309	13.3611	250000	1.8673	0.8354	0.8162
0.5363	13.6283	255000	1.8262	0.8379	0.8169
0.5468	13.8956	260000	1.8386	0.8367	0.8158
0.5079	14.1628	265000	1.8266	0.8373	0.8162
0.5073	14.4300	270000	1.8412	0.8371	0.8143
0.5244	14.6972	275000	1.8119	0.8399	0.8194
0.5193	14.9645	280000	1.8262	0.8369	0.8163
0.4852	15.2317	285000	1.8196	0.8376	0.8174
0.4986	15.4989	290000	1.8244	0.8356	0.8158
0.5004	15.7661	295000	1.8278	0.8376	0.8181
0.4746	16.0333	300000	1.8058	0.8390	0.8168
0.481	16.3006	305000	1.7852	0.8391	0.8208
0.4754	16.5678	310000	1.8060	0.8380	0.8166
0.4755	16.8350	315000	1.7905	0.8385	0.8173
0.4551	17.1022	320000	1.7950	0.8389	0.8202
0.4656	17.3695	325000	1.7796	0.8392	0.8188
0.4666	17.6367	330000	1.7877	0.8389	0.8179
0.4558	17.9039	335000	1.7922	0.8390	0.8200
0.4497	18.1711	340000	1.7725	0.8397	0.8190
0.4379	18.4384	345000	1.7586	0.8398	0.8205
0.4502	18.7056	350000	1.7776	0.8394	0.8188
0.4463	18.9728	355000	1.7701	0.8387	0.8176
0.4385	19.2400	360000	1.7699	0.8389	0.8192
0.4242	19.5072	365000	1.7744	0.8394	0.8190
0.4406	19.7745	370000	1.7983	0.8366	0.8186
0.4154	20.0417	375000	1.7586	0.8397	0.8205
0.4122	20.3089	380000	1.7634	0.8412	0.8231
0.4298	20.5761	385000	1.7475	0.8410	0.8209
0.43	20.8434	390000	1.7540	0.8393	0.8187
0.4081	21.1106	395000	1.7563	0.8402	0.8207
0.4184	21.3778	400000	1.7535	0.8413	0.8238
0.4098	21.6450	405000	1.7541	0.8395	0.8212
0.416	21.9122	410000	1.7410	0.8409	0.8218
0.4027	22.1795	415000	1.7348	0.8405	0.8209
0.4096	22.4467	420000	1.7360	0.8425	0.8226
0.4063	22.7139	425000	1.7433	0.8397	0.8202
0.4029	22.9811	430000	1.7388	0.8409	0.8231
0.3906	23.2484	435000	1.7412	0.8414	0.8233
0.3993	23.5156	440000	1.7291	0.8411	0.8198
0.4034	23.7828	445000	1.7241	0.8416	0.8223
0.3841	24.0500	450000	1.7305	0.8425	0.8230
0.391	24.3172	455000	1.7231	0.8413	0.8214
0.3963	24.5845	460000	1.7288	0.8410	0.8205
0.3943	24.8517	465000	1.7274	0.8419	0.8239
0.3915	25.1189	470000	1.7197	0.8417	0.8232
0.3742	25.3861	475000	1.7133	0.8419	0.8222
0.3839	25.6534	480000	1.7144	0.8419	0.8229
0.3845	25.9206	485000	1.7102	0.8422	0.8237
0.3731	26.1878	490000	1.7191	0.8415	0.8230
0.3818	26.4550	495000	1.7046	0.8422	0.8232
0.3753	26.7222	500000	1.7109	0.8414	0.8231
0.3682	26.9895	505000	1.6991	0.8422	0.8238
0.3715	27.2567	510000	1.7091	0.8420	0.8221
0.3809	27.5239	515000	1.7134	0.8410	0.8204
0.3702	27.7911	520000	1.7026	0.8425	0.8238
0.3672	28.0584	525000	1.6927	0.8431	0.8236
0.3653	28.3256	530000	1.7012	0.8426	0.8232
0.3727	28.5928	535000	1.7002	0.8424	0.8218
0.3745	28.8600	540000	1.7044	0.8416	0.8229
0.3654	29.1273	545000	1.7039	0.8422	0.8233
0.3633	29.3945	550000	1.7015	0.8425	0.8234
0.3561	29.6617	555000	1.6956	0.8429	0.8236
0.3734	29.9289	560000	1.6978	0.8427	0.8234

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1