metadata

library_name: transformers
license: mit
base_model: microsoft/Multilingual-MiniLM-L12-H384
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: m-minilm-l12-h384-data-augumented-dra-tam-mal-aw-classification-finetune
    results: []

m-minilm-l12-h384-data-augumented-dra-tam-mal-aw-classification-finetune

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.6411
Accuracy: 0.7702
F1: 0.8164

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: 0.0001
train_batch_size: 128
eval_batch_size: 128
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.6713	0.2222	20	0.6683	0.5998	0.7499
0.6494	0.4444	40	0.6559	0.6019	0.6459
0.6431	0.6667	60	0.6389	0.6508	0.7346
0.6079	0.8889	80	0.6318	0.6720	0.7166
0.5667	1.1111	100	0.5755	0.7021	0.7401
0.5353	1.3333	120	0.5437	0.7213	0.7927
0.5345	1.5556	140	0.5306	0.7482	0.7964
0.5178	1.7778	160	0.5366	0.7184	0.8031
0.4952	2.0	180	0.5046	0.7543	0.8050
0.4183	2.2222	200	0.5798	0.7278	0.7466
0.4257	2.4444	220	0.5373	0.7673	0.8075
0.3932	2.6667	240	0.5214	0.7665	0.8093
0.3914	2.8889	260	0.5125	0.7616	0.8133
0.3447	3.1111	280	0.5534	0.7653	0.8076
0.3122	3.3333	300	0.5874	0.7543	0.7901
0.3116	3.5556	320	0.5594	0.7649	0.8003
0.326	3.7778	340	0.5446	0.7661	0.8158
0.2979	4.0	360	0.5750	0.7681	0.8145
0.2457	4.2222	380	0.6121	0.7677	0.8140
0.2383	4.4444	400	0.5861	0.7689	0.8118
0.2396	4.6667	420	0.6161	0.7734	0.8156
0.2311	4.8889	440	0.5909	0.7751	0.8121
0.2139	5.1111	460	0.6411	0.7702	0.8164
0.2038	5.3333	480	0.6462	0.7718	0.8154
0.1884	5.5556	500	0.6443	0.7645	0.8043
0.1889	5.7778	520	0.6588	0.7665	0.8064
0.2081	6.0	540	0.6581	0.7665	0.8054

Framework versions

Transformers 4.45.2
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.20.3