Anwaarma/Arabert-twitter-sentiment

Testmeee

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02-twitter on the ajgt_twitter_ar dataset. It achieves the following results on the evaluation set:

• Loss: 0.3318
• Accuracy: 0.95

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: 1e-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	Accuracy
0.4258	1.11	50	0.2875	0.88
0.1848	2.22	100	0.2274	0.93
0.1117	3.33	150	0.2743	0.9
0.0598	4.44	200	0.2292	0.94
0.0354	5.56	250	0.2714	0.95
0.0319	6.67	300	0.2870	0.95
0.0195	7.78	350	0.3029	0.95
0.0136	8.89	400	0.3097	0.95
0.0078	10.0	450	0.3197	0.95
0.0088	11.11	500	0.3227	0.95
0.0057	12.22	550	0.3263	0.95
0.0058	13.33	600	0.3283	0.95
0.0051	14.44	650	0.3318	0.95

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.14.1