README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - f1
model-index:
  - name: BERTicSENTNEG4
    results: []

BERTicSENTNEG4

This model is a fine-tuned version of Tanor/BERTicSENTNEG4 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0842
• F1: 0.6275

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: 64
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 32

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	53	0.0668	0.0
No log	2.0	106	0.0432	0.6383
No log	3.0	159	0.0400	0.5714
No log	4.0	212	0.0458	0.5957
No log	5.0	265	0.0444	0.5882
No log	6.0	318	0.0556	0.5957
No log	7.0	371	0.0566	0.5714
No log	8.0	424	0.0587	0.5862
No log	9.0	477	0.0566	0.5660
0.0389	10.0	530	0.0693	0.5455
0.0389	11.0	583	0.0612	0.6383
0.0389	12.0	636	0.0596	0.6
0.0389	13.0	689	0.0671	0.6038
0.0389	14.0	742	0.0740	0.5957
0.0389	15.0	795	0.0799	0.5778
0.0389	16.0	848	0.0702	0.5957
0.0389	17.0	901	0.0737	0.6087
0.0389	18.0	954	0.0674	0.5660
0.0053	19.0	1007	0.0725	0.5957
0.0053	20.0	1060	0.0738	0.6
0.0053	21.0	1113	0.0821	0.625
0.0053	22.0	1166	0.0737	0.6
0.0053	23.0	1219	0.0828	0.6122
0.0053	24.0	1272	0.0776	0.6182
0.0053	25.0	1325	0.0792	0.6182
0.0053	26.0	1378	0.0791	0.6275
0.0053	27.0	1431	0.0812	0.6275
0.0053	28.0	1484	0.0819	0.6038
0.0029	29.0	1537	0.0831	0.6275
0.0029	30.0	1590	0.0834	0.6275
0.0029	31.0	1643	0.0837	0.6275
0.0029	32.0	1696	0.0842	0.6275

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1
• Datasets 2.13.1
• Tokenizers 0.13.3