Job_compatibility_model

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

• Loss: 0.6238
• Accuracy: 0.8598

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: 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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	32	0.6922	0.5
No log	2.0	64	0.6509	0.6238
No log	3.0	96	0.4218	0.8411
No log	4.0	128	0.3622	0.8481
No log	5.0	160	0.3383	0.8645
No log	6.0	192	0.3626	0.8528
No log	7.0	224	0.3939	0.8621
No log	8.0	256	0.4223	0.8715
No log	9.0	288	0.4271	0.8692
No log	10.0	320	0.4869	0.8621
No log	11.0	352	0.5057	0.8645
No log	12.0	384	0.5702	0.8528
No log	13.0	416	0.5277	0.8692
No log	14.0	448	0.5228	0.8785
No log	15.0	480	0.5332	0.8762
0.2235	16.0	512	0.5859	0.8715
0.2235	17.0	544	0.5938	0.8762
0.2235	18.0	576	0.6005	0.8715
0.2235	19.0	608	0.5941	0.8715
0.2235	20.0	640	0.6115	0.8762
0.2235	21.0	672	0.6098	0.8715
0.2235	22.0	704	0.6091	0.8715
0.2235	23.0	736	0.6223	0.8621
0.2235	24.0	768	0.6309	0.8598
0.2235	25.0	800	0.6238	0.8598

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.2