scenario-NON-KD-PO-COPY-D2_data-AmazonScience_massive_all_1_155

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: 2.1172
• Accuracy: 0.8547
• F1: 0.8307

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
0.5096	0.2672	5000	0.6672	0.8301	0.7901
0.4124	0.5344	10000	0.6564	0.8398	0.8058
0.3501	0.8017	15000	0.6697	0.8466	0.8172
0.232	1.0689	20000	0.7193	0.8480	0.8171
0.2262	1.3361	25000	0.7387	0.8489	0.8243
0.2146	1.6033	30000	0.7377	0.8480	0.8243
0.2007	1.8706	35000	0.7711	0.8463	0.8186
0.1472	2.1378	40000	0.8809	0.8492	0.8218
0.1509	2.4050	45000	0.8559	0.8495	0.8242
0.1513	2.6722	50000	0.8919	0.8481	0.8226
0.1428	2.9394	55000	0.8636	0.8503	0.8272
0.1154	3.2067	60000	0.9868	0.8512	0.8312
0.1148	3.4739	65000	0.9541	0.8502	0.8253
0.1152	3.7411	70000	0.9729	0.8503	0.8305
0.1031	4.0083	75000	1.0581	0.8459	0.8260
0.0942	4.2756	80000	1.0732	0.8487	0.8248
0.0961	4.5428	85000	1.0528	0.8509	0.8284
0.099	4.8100	90000	1.0421	0.8486	0.8250
0.0761	5.0772	95000	1.1302	0.8480	0.8221
0.0822	5.3444	100000	1.1138	0.8492	0.8261
0.0833	5.6117	105000	1.1368	0.8478	0.8243
0.0768	5.8789	110000	1.1240	0.8503	0.8256
0.0593	6.1461	115000	1.2497	0.8491	0.8277
0.0687	6.4133	120000	1.2261	0.8484	0.8252
0.0738	6.6806	125000	1.1406	0.8482	0.8257
0.0651	6.9478	130000	1.2154	0.8488	0.8232
0.0604	7.2150	135000	1.3001	0.8459	0.8245
0.0646	7.4822	140000	1.2426	0.8488	0.8223
0.0653	7.7495	145000	1.2843	0.8499	0.8283
0.0433	8.0167	150000	1.3361	0.8500	0.8261
0.0497	8.2839	155000	1.2612	0.8492	0.8241
0.0517	8.5511	160000	1.3544	0.8474	0.8270
0.0624	8.8183	165000	1.2964	0.8489	0.8275
0.0381	9.0856	170000	1.3753	0.8486	0.8242
0.0405	9.3528	175000	1.4366	0.8484	0.8219
0.0488	9.6200	180000	1.4429	0.8458	0.8235
0.0495	9.8872	185000	1.3962	0.8496	0.8273
0.0363	10.1545	190000	1.5269	0.8483	0.8247
0.0423	10.4217	195000	1.4415	0.8489	0.8236
0.0448	10.6889	200000	1.3974	0.8482	0.8218
0.0373	10.9561	205000	1.4432	0.8465	0.8245
0.0294	11.2233	210000	1.5381	0.8493	0.8234
0.0336	11.4906	215000	1.5496	0.8490	0.8218
0.041	11.7578	220000	1.4867	0.8468	0.8240
0.028	12.0250	225000	1.5215	0.8498	0.8270
0.0287	12.2922	230000	1.5916	0.8495	0.8256
0.0314	12.5595	235000	1.6239	0.8500	0.8291
0.0347	12.8267	240000	1.5390	0.8464	0.8243
0.0247	13.0939	245000	1.5492	0.8502	0.8271
0.0281	13.3611	250000	1.5729	0.8465	0.8262
0.025	13.6283	255000	1.6933	0.8470	0.8249
0.0325	13.8956	260000	1.5823	0.8511	0.8289
0.0268	14.1628	265000	1.6099	0.8476	0.8213
0.0339	14.4300	270000	1.6234	0.8514	0.8306
0.0285	14.6972	275000	1.5725	0.8524	0.8288
0.0309	14.9645	280000	1.6213	0.8498	0.8251
0.0274	15.2317	285000	1.6650	0.8481	0.8235
0.02	15.4989	290000	1.7295	0.8488	0.8261
0.0299	15.7661	295000	1.6868	0.8513	0.8277
0.0196	16.0333	300000	1.7188	0.8492	0.8234
0.0221	16.3006	305000	1.7492	0.8511	0.8265
0.0273	16.5678	310000	1.6775	0.8505	0.8267
0.0212	16.8350	315000	1.7104	0.8492	0.8253
0.0188	17.1022	320000	1.6627	0.8502	0.8271
0.0225	17.3695	325000	1.7774	0.8504	0.8256
0.0204	17.6367	330000	1.7537	0.8489	0.8240
0.0261	17.9039	335000	1.6829	0.8516	0.8273
0.0153	18.1711	340000	1.8292	0.8532	0.8295
0.0226	18.4384	345000	1.7632	0.8511	0.8273
0.0213	18.7056	350000	1.7662	0.8515	0.8252
0.0162	18.9728	355000	1.7639	0.8519	0.8280
0.0155	19.2400	360000	1.8033	0.8521	0.8262
0.0151	19.5072	365000	1.8838	0.8505	0.8241
0.0151	19.7745	370000	1.8777	0.8532	0.8268
0.0138	20.0417	375000	1.8770	0.8522	0.8284
0.0152	20.3089	380000	1.8785	0.8525	0.8275
0.0136	20.5761	385000	1.8761	0.8530	0.8310
0.0181	20.8434	390000	1.8843	0.8526	0.8288
0.0113	21.1106	395000	1.8946	0.8540	0.8306
0.0144	21.3778	400000	1.8532	0.8530	0.8318
0.0149	21.6450	405000	1.9300	0.8519	0.8296
0.0165	21.9122	410000	1.9072	0.8532	0.8316
0.0128	22.1795	415000	1.9919	0.8516	0.8270
0.0071	22.4467	420000	1.9807	0.8543	0.8313
0.0099	22.7139	425000	1.9709	0.8527	0.8277
0.011	22.9811	430000	1.9782	0.8521	0.8279
0.0102	23.2484	435000	2.0483	0.8515	0.8268
0.013	23.5156	440000	2.0266	0.8527	0.8290
0.0145	23.7828	445000	1.9633	0.8533	0.8303
0.0091	24.0500	450000	2.0645	0.8514	0.8274
0.0128	24.3172	455000	2.0243	0.8541	0.8314
0.0101	24.5845	460000	2.0680	0.8518	0.8281
0.0087	24.8517	465000	2.0453	0.8526	0.8290
0.0087	25.1189	470000	2.0808	0.8529	0.8294
0.0079	25.3861	475000	2.1050	0.8522	0.8285
0.0101	25.6534	480000	2.0773	0.8518	0.8275
0.0094	25.9206	485000	2.0602	0.8533	0.8305
0.0097	26.1878	490000	2.0994	0.8519	0.8267
0.01	26.4550	495000	2.0608	0.8549	0.8309
0.0107	26.7222	500000	2.0748	0.8539	0.8300
0.0088	26.9895	505000	2.1072	0.8539	0.8305
0.0057	27.2567	510000	2.1086	0.8548	0.8318
0.0096	27.5239	515000	2.1032	0.8539	0.8297
0.0082	27.7911	520000	2.1119	0.8538	0.8294
0.0081	28.0584	525000	2.1045	0.8543	0.8296
0.0079	28.3256	530000	2.1268	0.8539	0.8296
0.0089	28.5928	535000	2.1128	0.8547	0.8303
0.0079	28.8600	540000	2.1003	0.8540	0.8294
0.0056	29.1273	545000	2.1130	0.8541	0.8294
0.0082	29.3945	550000	2.1167	0.8544	0.8302
0.0056	29.6617	555000	2.1182	0.8546	0.8306
0.0076	29.9289	560000	2.1172	0.8547	0.8307

Framework versions

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