README.md

metadata

license: mit
base_model: camembert-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: VogagenRelation
    results: []

VogagenRelation

This model is a fine-tuned version of camembert-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6927
• Accuracy: 0.5203

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-08
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.21	100	0.6928	0.5133
No log	0.42	200	0.6928	0.5148
No log	0.62	300	0.6928	0.5156
No log	0.83	400	0.6928	0.5156
0.6925	1.04	500	0.6928	0.5148
0.6925	1.25	600	0.6928	0.5156
0.6925	1.46	700	0.6928	0.5156
0.6925	1.66	800	0.6928	0.5172
0.6925	1.87	900	0.6928	0.5164
0.6929	2.08	1000	0.6928	0.5156
0.6929	2.29	1100	0.6928	0.5172
0.6929	2.49	1200	0.6928	0.5172
0.6929	2.7	1300	0.6928	0.5164
0.6929	2.91	1400	0.6928	0.5164
0.6928	3.12	1500	0.6928	0.5172
0.6928	3.33	1600	0.6928	0.5195
0.6928	3.53	1700	0.6928	0.5195
0.6928	3.74	1800	0.6928	0.5211
0.6928	3.95	1900	0.6928	0.5179
0.6927	4.16	2000	0.6928	0.5187
0.6927	4.37	2100	0.6928	0.5179
0.6927	4.57	2200	0.6927	0.5179
0.6927	4.78	2300	0.6927	0.5187
0.6927	4.99	2400	0.6927	0.5187
0.6929	5.2	2500	0.6928	0.5179
0.6929	5.41	2600	0.6928	0.5187
0.6929	5.61	2700	0.6928	0.5211
0.6929	5.82	2800	0.6927	0.5187
0.6929	6.03	2900	0.6927	0.5187
0.692	6.24	3000	0.6927	0.5195
0.692	6.44	3100	0.6927	0.5203
0.692	6.65	3200	0.6927	0.5203
0.692	6.86	3300	0.6927	0.5203
0.692	7.07	3400	0.6927	0.5203
0.6933	7.28	3500	0.6927	0.5203
0.6933	7.48	3600	0.6927	0.5203
0.6933	7.69	3700	0.6927	0.5203
0.6933	7.9	3800	0.6927	0.5203

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.14.1