pp_distilbert_ft_emotions

This model is a fine-tuned version of distilbert-base-uncased on the emotion dataset. It achieves the following results on the evaluation set:

• Loss: 0.1493
• Accuracy: 0.9275

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: 5e-05
• train_batch_size: 80
• eval_batch_size: 80
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.25	50	0.7329	0.758
No log	0.5	100	0.2915	0.9195
No log	0.75	150	0.2150	0.927
No log	1.0	200	0.1780	0.9285
No log	1.25	250	0.1777	0.9295
No log	1.5	300	0.1547	0.937
No log	1.75	350	0.1467	0.935
No log	2.0	400	0.1446	0.937
No log	2.25	450	0.1482	0.934
0.3073	2.5	500	0.1335	0.9385
0.3073	2.75	550	0.1344	0.9415
0.3073	3.0	600	0.1229	0.9425
0.3073	3.25	650	0.1381	0.939
0.3073	3.5	700	0.1292	0.941
0.3073	3.75	750	0.1278	0.944
0.3073	4.0	800	0.1258	0.944

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2