README.md

metadata

license: apache-2.0
base_model: DmitryPogrebnoy/MedRuRobertaLarge
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: MedRuRobertaLarge_neg
    results: []

MedRuRobertaLarge_neg

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

• Loss: 0.4946
• Precision: 0.5932
• Recall: 0.5804
• F1: 0.5868
• Accuracy: 0.9015

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	50	0.6600	0.0	0.0	0.0	0.7759
No log	2.0	100	0.5893	0.0	0.0	0.0	0.7826
No log	3.0	150	0.4690	0.0265	0.0173	0.0210	0.8164
No log	4.0	200	0.4564	0.0979	0.1252	0.1099	0.8204
No log	5.0	250	0.3628	0.1881	0.2852	0.2266	0.8538
No log	6.0	300	0.3105	0.3469	0.3622	0.3544	0.8901
No log	7.0	350	0.3382	0.4084	0.3738	0.3903	0.8909
No log	8.0	400	0.2926	0.4774	0.4682	0.4728	0.9020
No log	9.0	450	0.2955	0.4630	0.4817	0.4721	0.9046
0.3854	10.0	500	0.3161	0.5367	0.4933	0.5141	0.9080
0.3854	11.0	550	0.3103	0.4612	0.6069	0.5241	0.9018
0.3854	12.0	600	0.3020	0.5614	0.6166	0.5877	0.9136
0.3854	13.0	650	0.3738	0.5625	0.5896	0.5757	0.9157
0.3854	14.0	700	0.3322	0.4834	0.5877	0.5304	0.9031
0.3854	15.0	750	0.3619	0.4855	0.5472	0.5145	0.9083
0.3854	16.0	800	0.3597	0.4815	0.6763	0.5625	0.9018
0.3854	17.0	850	0.4065	0.5488	0.6281	0.5858	0.9145
0.3854	18.0	900	0.4491	0.6047	0.6513	0.6271	0.9184
0.3854	19.0	950	0.4184	0.4972	0.6898	0.5779	0.8986
0.0771	20.0	1000	0.3366	0.5929	0.6089	0.6008	0.9238
0.0771	21.0	1050	0.6161	0.6834	0.5241	0.5932	0.9009
0.0771	22.0	1100	0.6387	0.3497	0.6859	0.4632	0.8377
0.0771	23.0	1150	0.3559	0.6004	0.6224	0.6112	0.9198
0.0771	24.0	1200	0.4161	0.5926	0.6166	0.6043	0.9203
0.0771	25.0	1250	0.4341	0.6365	0.6108	0.6234	0.9199
0.0771	26.0	1300	0.3910	0.6184	0.5838	0.6006	0.9153
0.0771	27.0	1350	0.3706	0.5396	0.6435	0.5870	0.9209
0.0771	28.0	1400	0.5059	0.5833	0.6069	0.5949	0.9157

Framework versions

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