metadata

library_name: transformers
license: mit
base_model: intfloat/e5-small
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: intfloat-e5-small-arabic-fp16-allagree
    results: []

intfloat-e5-small-arabic-fp16-allagree

This model is a fine-tuned version of intfloat/e5-small on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5430
Accuracy: 0.7845
Precision: 0.7994
Recall: 0.7845
F1: 0.7886

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: 64
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 128
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.3
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.0961	0.7463	50	1.0715	0.3853	0.7241	0.3853	0.2241
1.0181	1.4925	100	0.9009	0.6875	0.7524	0.6875	0.6122
0.8328	2.2388	150	0.7633	0.7052	0.7815	0.7052	0.6315
0.7492	2.9851	200	0.6860	0.7295	0.6952	0.7295	0.6858
0.6931	3.7313	250	0.6982	0.7369	0.7275	0.7369	0.7209
0.6623	4.4776	300	0.6326	0.7705	0.7544	0.7705	0.7485
0.6107	5.2239	350	0.6350	0.7556	0.7696	0.7556	0.7530
0.5789	5.9701	400	0.5892	0.7649	0.7924	0.7649	0.7713
0.5593	6.7164	450	0.5449	0.7985	0.7946	0.7985	0.7963
0.5343	7.4627	500	0.5486	0.7845	0.8008	0.7845	0.7897
0.5177	8.2090	550	0.5373	0.8013	0.8038	0.8013	0.8016
0.5201	8.9552	600	0.5370	0.7882	0.7996	0.7882	0.7915
0.4962	9.7015	650	0.5357	0.7882	0.7994	0.7882	0.7917

Framework versions

Transformers 4.48.2
Pytorch 2.6.0+cu124
Datasets 3.3.1
Tokenizers 0.21.0