results

This model is a fine-tuned version of savasy/bert-base-turkish-sentiment-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2027
• Accuracy: 0.9416
• Precision: 0.9417
• Recall: 0.9416
• F1: 0.9416

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.9703	0.0628	100	0.8177	0.7173	0.7228	0.7173	0.7080
0.6545	0.1255	200	0.4711	0.9024	0.9032	0.9024	0.9024
0.3702	0.1883	300	0.2601	0.9174	0.9173	0.9174	0.9166
0.2437	0.2511	400	0.2307	0.9238	0.9233	0.9238	0.9232
0.2308	0.3139	500	0.2263	0.9259	0.9256	0.9259	0.9254
0.2103	0.3766	600	0.2381	0.9247	0.9243	0.9247	0.9241
0.2046	0.4394	700	0.2187	0.9328	0.9331	0.9328	0.9326
0.2151	0.5022	800	0.1948	0.9341	0.9341	0.9341	0.9339
0.2116	0.5650	900	0.2083	0.9329	0.9325	0.9329	0.9326
0.2155	0.6277	1000	0.2264	0.9325	0.9325	0.9325	0.9322
0.1934	0.6905	1100	0.2007	0.9348	0.9353	0.9348	0.9346
0.1897	0.7533	1200	0.2052	0.9331	0.9337	0.9331	0.9328
0.1914	0.8161	1300	0.1959	0.9364	0.9361	0.9364	0.9362
0.1672	0.8788	1400	0.2224	0.9346	0.9347	0.9346	0.9343
0.2093	0.9416	1500	0.2093	0.9381	0.9379	0.9381	0.9380
0.2001	1.0044	1600	0.1998	0.9369	0.9366	0.9369	0.9367
0.1638	1.0672	1700	0.2174	0.9369	0.9367	0.9369	0.9367
0.1504	1.1299	1800	0.2146	0.9384	0.9382	0.9384	0.9383
0.1823	1.1927	1900	0.2088	0.9385	0.9384	0.9385	0.9384
0.1808	1.2555	2000	0.2018	0.9382	0.9382	0.9382	0.9381
0.1601	1.3183	2100	0.1942	0.9381	0.9389	0.9381	0.9379
0.1791	1.3810	2200	0.2190	0.9278	0.9315	0.9278	0.9274
0.1657	1.4438	2300	0.1945	0.9402	0.9406	0.9402	0.9401
0.1449	1.5066	2400	0.2027	0.9416	0.9417	0.9416	0.9416
0.1693	1.5694	2500	0.1992	0.9396	0.9399	0.9396	0.9394
0.1539	1.6321	2600	0.2085	0.9377	0.9386	0.9377	0.9375
0.1573	1.6949	2700	0.2101	0.9386	0.9382	0.9386	0.9384
0.1401	1.7577	2800	0.2087	0.9384	0.9392	0.9384	0.9383
0.1478	1.8205	2900	0.2056	0.9409	0.9410	0.9409	0.9408

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0