sentiment_pc_combinedBase

This model is a fine-tuned version of ahmedrachid/FinancialBERT-Sentiment-Analysis on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5153
• Accuracy: 0.8683
• F1: 0.8376

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
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.1739	50	0.5234	0.8096	0.7723
No log	0.3478	100	0.4390	0.8457	0.8151
No log	0.5217	150	0.4168	0.8491	0.8137
No log	0.6957	200	0.4252	0.8522	0.8150
No log	0.8696	250	0.3931	0.8561	0.8196
No log	1.0435	300	0.4409	0.8409	0.8118
No log	1.2174	350	0.4108	0.8657	0.8271
No log	1.3913	400	0.4382	0.8613	0.8292
No log	1.5652	450	0.4147	0.8622	0.8287
0.415	1.7391	500	0.4069	0.8652	0.8331
0.415	1.9130	550	0.4170	0.8591	0.8275
0.415	2.0870	600	0.4533	0.8626	0.8296
0.415	2.2609	650	0.4613	0.87	0.8401
0.415	2.4348	700	0.4531	0.8770	0.8447
0.415	2.6087	750	0.4534	0.8583	0.8277
0.415	2.7826	800	0.4756	0.8570	0.8274
0.415	2.9565	850	0.4482	0.8683	0.8391
0.415	3.1304	900	0.4858	0.8665	0.8350
0.415	3.3043	950	0.4873	0.8639	0.8341
0.1812	3.4783	1000	0.5153	0.8683	0.8376
0.1812	3.6522	1050	0.5345	0.8578	0.8281
0.1812	3.8261	1100	0.5372	0.8609	0.8331
0.1812	4.0	1150	0.5172	0.8670	0.8379
0.1812	4.1739	1200	0.5643	0.8643	0.8342
0.1812	4.3478	1250	0.5783	0.8622	0.8326
0.1812	4.5217	1300	0.5909	0.8565	0.8273

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1