relevance-analysis

This model is a fine-tuned version of mdhugol/indonesia-bert-sentiment-classification on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5308
• Accuracy: 0.8230

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: 1e-08
• train_batch_size: 8
• eval_batch_size: 8
• seed: 41
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 200
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.4407	1.7986	500	3.4936	0.2
3.0684	3.5971	1000	3.1423	0.2162
2.7449	5.3957	1500	2.7929	0.2243
2.4066	7.1942	2000	2.4499	0.2459
2.0853	8.9928	2500	2.1284	0.2838
1.8021	10.7914	3000	1.8369	0.3514
1.546	12.5899	3500	1.5842	0.4176
1.3521	14.3885	4000	1.3852	0.4811
1.1889	16.1871	4500	1.2250	0.5392
1.0671	17.9856	5000	1.1067	0.5824
0.9838	19.7842	5500	1.0203	0.6365
0.8825	21.5827	6000	0.9541	0.6757
0.846	23.3813	6500	0.9053	0.7027
0.8263	25.1799	7000	0.8669	0.7284
0.7888	26.9784	7500	0.8360	0.75
0.7516	28.7770	8000	0.8106	0.7608
0.7417	30.5755	8500	0.7894	0.7703
0.7277	32.3741	9000	0.7702	0.7797
0.7217	34.1727	9500	0.7532	0.7878
0.694	35.9712	10000	0.7382	0.7932
0.673	37.7698	10500	0.7249	0.7973
0.6955	39.5683	11000	0.7124	0.7986
0.6544	41.3669	11500	0.7013	0.8027
0.6548	43.1655	12000	0.6906	0.8068
0.6355	44.9640	12500	0.6811	0.8108
0.6386	46.7626	13000	0.6720	0.8122
0.627	48.5612	13500	0.6637	0.8122
0.6199	50.3597	14000	0.6559	0.8122
0.6291	52.1583	14500	0.6487	0.8122
0.5938	53.9568	15000	0.6422	0.8122
0.603	55.7554	15500	0.6358	0.8135
0.594	57.5540	16000	0.6298	0.8162
0.5931	59.3525	16500	0.6243	0.8176
0.5865	61.1511	17000	0.6189	0.8176
0.5797	62.9496	17500	0.6143	0.8176
0.5764	64.7482	18000	0.6099	0.8176
0.5856	66.5468	18500	0.6054	0.8176
0.5513	68.3453	19000	0.6019	0.8176
0.5823	70.1439	19500	0.5978	0.8176
0.5588	71.9424	20000	0.5945	0.8176
0.5571	73.7410	20500	0.5913	0.8189
0.5722	75.5396	21000	0.5882	0.8189
0.5507	77.3381	21500	0.5853	0.8189
0.5524	79.1367	22000	0.5827	0.8189
0.5487	80.9353	22500	0.5800	0.8189
0.545	82.7338	23000	0.5776	0.8189
0.5465	84.5324	23500	0.5754	0.8189
0.5645	86.3309	24000	0.5730	0.8189
0.5195	88.1295	24500	0.5712	0.8189
0.5405	89.9281	25000	0.5692	0.8189
0.5331	91.7266	25500	0.5674	0.8189
0.5384	93.5252	26000	0.5657	0.8189
0.5407	95.3237	26500	0.5639	0.8189
0.5368	97.1223	27000	0.5623	0.8189
0.5254	98.9209	27500	0.5607	0.8189
0.5327	100.7194	28000	0.5592	0.8189
0.5309	102.5180	28500	0.5578	0.8203
0.5309	104.3165	29000	0.5564	0.8203
0.516	106.1151	29500	0.5550	0.8203
0.5325	107.9137	30000	0.5537	0.8203
0.5239	109.7122	30500	0.5525	0.8203
0.5122	111.5108	31000	0.5514	0.8203
0.5309	113.3094	31500	0.5502	0.8203
0.5185	115.1079	32000	0.5491	0.8203
0.5209	116.9065	32500	0.5480	0.8203
0.5101	118.7050	33000	0.5470	0.8203
0.5063	120.5036	33500	0.5461	0.8203
0.5241	122.3022	34000	0.5452	0.8203
0.5056	124.1007	34500	0.5444	0.8203
0.5122	125.8993	35000	0.5435	0.8203
0.5065	127.6978	35500	0.5428	0.8203
0.5142	129.4964	36000	0.5419	0.8189
0.5115	131.2950	36500	0.5412	0.8189
0.5158	133.0935	37000	0.5405	0.8189
0.5076	134.8921	37500	0.5398	0.8203
0.5079	136.6906	38000	0.5393	0.8203
0.5014	138.4892	38500	0.5386	0.8203
0.511	140.2878	39000	0.5380	0.8203
0.5035	142.0863	39500	0.5375	0.8203
0.5008	143.8849	40000	0.5370	0.8203
0.5052	145.6835	40500	0.5365	0.8216
0.508	147.4820	41000	0.5360	0.8216
0.4975	149.2806	41500	0.5356	0.8216
0.5102	151.0791	42000	0.5351	0.8216
0.5092	152.8777	42500	0.5346	0.8216
0.4985	154.6763	43000	0.5343	0.8216
0.4997	156.4748	43500	0.5339	0.8216
0.5058	158.2734	44000	0.5336	0.8216
0.4978	160.0719	44500	0.5333	0.8230
0.4983	161.8705	45000	0.5330	0.8230
0.5097	163.6691	45500	0.5327	0.8230
0.4966	165.4676	46000	0.5325	0.8230
0.5003	167.2662	46500	0.5322	0.8230
0.4967	169.0647	47000	0.5320	0.8230
0.497	170.8633	47500	0.5318	0.8230
0.5108	172.6619	48000	0.5316	0.8230
0.4891	174.4604	48500	0.5315	0.8230
0.4967	176.2590	49000	0.5314	0.8230
0.5029	178.0576	49500	0.5313	0.8230
0.5023	179.8561	50000	0.5312	0.8230
0.5007	181.6547	50500	0.5311	0.8230
0.4966	183.4532	51000	0.5310	0.8230
0.4921	185.2518	51500	0.5310	0.8230
0.4966	187.0504	52000	0.5309	0.8230
0.5056	188.8489	52500	0.5309	0.8230
0.4888	190.6475	53000	0.5309	0.8230
0.4995	192.4460	53500	0.5309	0.8230
0.4976	194.2446	54000	0.5309	0.8230
0.4942	196.0432	54500	0.5308	0.8230
0.4993	197.8417	55000	0.5308	0.8230
0.4988	199.6403	55500	0.5308	0.8230

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.1