metadata

license: cc-by-4.0
base_model: Helsinki-NLP/opus-mt-tc-big-en-ar
tags:
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: opus-mt-tc-big-en-ar-finetuned-franco-to-arabic-3
    results: []

opus-mt-tc-big-en-ar-finetuned-franco-to-arabic-3

This model is a fine-tuned version of Helsinki-NLP/opus-mt-tc-big-en-ar on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.8103
Bleu: 8.7335
Gen Len: 23.16

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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	14	8.4807	0.0	403.72
No log	2.0	28	7.7371	0.0	189.68
No log	3.0	42	7.2123	0.1961	80.52
No log	4.0	56	6.8014	0.1764	100.08
No log	5.0	70	6.4285	0.2171	54.96
No log	6.0	84	6.1286	0.209	88.92
No log	7.0	98	5.8323	0.2334	26.0
No log	8.0	112	5.6044	0.2352	25.6
No log	9.0	126	5.3873	0.2435	25.84
No log	10.0	140	5.2068	0.2459	24.48
No log	11.0	154	5.0464	0.6902	23.68
No log	12.0	168	4.8825	0.7391	24.4
No log	13.0	182	4.7366	1.185	22.68
No log	14.0	196	4.5743	1.3994	23.6
No log	15.0	210	4.4676	1.9653	23.84
No log	16.0	224	4.3406	1.9566	23.88
No log	17.0	238	4.2305	2.1215	22.88
No log	18.0	252	4.1240	4.4593	22.28
No log	19.0	266	4.0583	3.2999	22.52
No log	20.0	280	3.9704	7.1094	23.12
No log	21.0	294	3.9151	7.3081	23.12
No log	22.0	308	3.8737	7.2132	23.72
No log	23.0	322	3.8393	9.0724	22.92
No log	24.0	336	3.8173	8.8528	23.0
No log	25.0	350	3.8103	8.7335	23.16

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu118
Datasets 2.14.4
Tokenizers 0.13.3