README.md

metadata

license: mit
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: model_from_berturk_Feb_5_TrainTestSplit
    results: []

model_from_berturk_Feb_5_TrainTestSplit

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

• Loss: 0.3125
• Precision: 0.9120
• Recall: 0.9126
• F1: 0.9123
• Accuracy: 0.9376

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	185	0.2333	0.9065	0.9066	0.9066	0.9343
No log	2.0	370	0.2115	0.9122	0.9143	0.9133	0.9389
0.3861	3.0	555	0.2049	0.9185	0.9175	0.9180	0.9423
0.3861	4.0	740	0.2073	0.9183	0.9185	0.9184	0.9420
0.3861	5.0	925	0.2174	0.9150	0.9155	0.9153	0.9397
0.1487	6.0	1110	0.2227	0.9177	0.9185	0.9181	0.9415
0.1487	7.0	1295	0.2399	0.9149	0.9160	0.9155	0.9396
0.1487	8.0	1480	0.2504	0.9158	0.9163	0.9160	0.9400
0.0942	9.0	1665	0.2692	0.9141	0.9152	0.9146	0.9392
0.0942	10.0	1850	0.2782	0.9130	0.9153	0.9141	0.9388
0.0589	11.0	2035	0.2908	0.9131	0.9144	0.9138	0.9388
0.0589	12.0	2220	0.2940	0.9121	0.9136	0.9128	0.9377
0.0589	13.0	2405	0.3068	0.9117	0.9130	0.9123	0.9376
0.0407	14.0	2590	0.3107	0.9132	0.9148	0.9140	0.9387
0.0407	15.0	2775	0.3125	0.9120	0.9126	0.9123	0.9376

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2