metadata

license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
model-index:
  - name: distilbert-base-uncased-three_v2
    results: []

distilbert-base-uncased-three_v2

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

Loss: 0.0646
Overall Precision: 0.8692
Overall Recall: 0.9049
Overall F1: 0.8867
Overall Accuracy: 0.9812
Accountname F1: 0.9773
Accountnumber F1: 0.9883
Age F1: 0.8924
Amount F1: 0.9622
Bic F1: 0.8699
Bitcoinaddress F1: 0.9496
Buildingnumber F1: 0.9553
City F1: 0.9484
Companyname F1: 0.7802
County F1: 0.9546
Creditcardcvv F1: 0.8307
Creditcardissuer F1: 0.9822
Creditcardnumber F1: 0.8662
Currency F1: 0.7573
Currencycode F1: 0.7819
Currencyname F1: 0.0943
Currencysymbol F1: 0.8871
Date F1: 0.8276
Dob F1: 0.8951
Email F1: 0.9543
Ethereumaddress F1: 1.0
Eyecolor F1: 0.9344
Firstname F1: 0.8670
Gender F1: 0.9632
Height F1: 0.9716
Iban F1: 0.9468
Ip F1: 0.8493
Ipv4 F1: 0.8556
Ipv6 F1: 0.6738
Jobarea F1: 0.8987
Jobtitle F1: 0.9577
Jobtype F1: 0.9303
Lastname F1: 0.8520
Litecoinaddress F1: 0.8554
Mac F1: 0.9928
Maskednumber F1: 0.8369
Middlename F1: 0.7513
Nearbygpscoordinate F1: 0.9269
Ordinaldirection F1: 0.9817
Password F1: 0.8792
Phoneimei F1: 0.9887
Phonenumber F1: 0.9230
Pin F1: 0.8550
Prefix F1: 0.9199
Secondaryaddress F1: 0.9740
Sex F1: 0.9667
Ssn F1: 0.9150
State F1: 0.9738
Street F1: 0.8667
Time F1: 0.9069
Url F1: 0.9878
Useragent F1: 0.9943
Username F1: 0.8805
Vehiclevin F1: 0.9620
Vehiclevrm F1: 0.9821
Zipcode F1: 0.9423

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: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.2
num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Overall Precision	Overall Recall	Overall F1	Overall Accuracy	Accountname F1	Accountnumber F1	Age F1	Amount F1	Bic F1	Bitcoinaddress F1	Buildingnumber F1	City F1	Companyname F1	County F1	Creditcardcvv F1	Creditcardissuer F1	Creditcardnumber F1	Currency F1	Currencycode F1	Currencyname F1	Currencysymbol F1	Date F1	Dob F1	Email F1	Ethereumaddress F1	Eyecolor F1	Firstname F1	Gender F1	Height F1	Iban F1	Ip F1	Ipv4 F1	Ipv6 F1	Jobarea F1	Jobtitle F1	Jobtype F1	Lastname F1	Litecoinaddress F1	Mac F1	Maskednumber F1	Middlename F1	Nearbygpscoordinate F1	Ordinaldirection F1	Password F1	Phoneimei F1	Phonenumber F1	Pin F1	Prefix F1	Secondaryaddress F1	Sex F1	Ssn F1	State F1	Street F1	Time F1	Url F1	Useragent F1	Username F1	Vehiclevin F1	Vehiclevrm F1	Zipcode F1
0.2314	1.0	1299	0.1322	0.7381	0.8044	0.7698	0.9699	0.8367	0.8496	0.6806	0.5746	0.3430	0.8041	0.8363	0.8420	0.6640	0.3719	0.0357	0.4493	0.4554	0.5	0.0	0.0	0.4768	0.7367	0.8515	0.9459	0.9908	0.1847	0.75	0.6004	0.6974	0.7952	0.8253	0.8246	0.7153	0.1512	0.7939	0.4468	0.7328	0.0	0.9026	0.4479	0.1935	0.8676	0.4604	0.7485	0.9364	0.8741	0.0060	0.8869	0.8963	0.8999	0.8115	0.8037	0.7648	0.8413	0.9516	0.9388	0.8054	0.3968	0.4866	0.8015
0.0799	2.0	2598	0.0761	0.8402	0.8811	0.8602	0.9782	0.9717	0.9659	0.8657	0.9269	0.8091	0.9468	0.9322	0.9342	0.7616	0.9097	0.7692	0.9821	0.7663	0.7208	0.5127	0.0248	0.8171	0.7995	0.8872	0.9503	0.9985	0.8933	0.8138	0.9242	0.9632	0.9326	0.8285	0.8610	0.5590	0.8205	0.9382	0.9207	0.8141	0.8254	0.9928	0.7820	0.5016	0.8957	0.9611	0.8398	0.9838	0.9222	0.7815	0.9156	0.9650	0.9534	0.9125	0.9593	0.8373	0.8884	0.9804	0.9806	0.8505	0.9056	0.9592	0.9209
0.0604	3.0	3897	0.0681	0.8591	0.8945	0.8765	0.9799	0.9829	0.9882	0.8940	0.9528	0.8575	0.9588	0.9468	0.9462	0.7795	0.9411	0.8156	0.9821	0.8064	0.7556	0.7036	0.0317	0.8614	0.8159	0.8861	0.9547	0.9969	0.9311	0.8487	0.9505	0.9718	0.9388	0.8284	0.7476	0.7306	0.8966	0.9562	0.9323	0.8372	0.8718	0.9928	0.8047	0.6657	0.9103	0.9769	0.8639	0.9937	0.9262	0.8207	0.9170	0.9737	0.9663	0.9201	0.9705	0.8333	0.9061	0.9915	0.9909	0.8685	0.9808	0.9671	0.9289
0.0469	4.0	5196	0.0646	0.8692	0.9049	0.8867	0.9812	0.9773	0.9883	0.8924	0.9622	0.8699	0.9496	0.9553	0.9484	0.7802	0.9546	0.8307	0.9822	0.8662	0.7573	0.7819	0.0943	0.8871	0.8276	0.8951	0.9543	1.0	0.9344	0.8670	0.9632	0.9716	0.9468	0.8493	0.8556	0.6738	0.8987	0.9577	0.9303	0.8520	0.8554	0.9928	0.8369	0.7513	0.9269	0.9817	0.8792	0.9887	0.9230	0.8550	0.9199	0.9740	0.9667	0.9150	0.9738	0.8667	0.9069	0.9878	0.9943	0.8805	0.9620	0.9821	0.9423
0.0365	5.0	6495	0.0688	0.8757	0.9066	0.8909	0.9815	0.9877	0.9891	0.8980	0.9720	0.8845	0.9588	0.9591	0.9511	0.8042	0.9555	0.8268	0.9870	0.8498	0.7164	0.7954	0.2376	0.9093	0.8287	0.8881	0.9543	1.0	0.9392	0.8648	0.9776	0.9771	0.9657	0.8602	0.8201	0.7170	0.9104	0.9594	0.9336	0.8596	0.8698	0.9928	0.8396	0.7732	0.9266	0.9837	0.8613	0.9799	0.9370	0.8524	0.9232	0.9724	0.9713	0.9317	0.9748	0.8684	0.9145	0.9897	0.9909	0.8822	0.9620	0.9594	0.9490
0.0277	6.0	7794	0.0766	0.8745	0.9120	0.8929	0.9813	0.9877	0.9900	0.8945	0.9746	0.8930	0.9434	0.9551	0.9471	0.7986	0.96	0.8303	0.9853	0.8607	0.7025	0.8011	0.2532	0.9124	0.8337	0.8961	0.9518	0.9985	0.9396	0.8713	0.9748	0.9801	0.9543	0.8564	0.8474	0.7036	0.9193	0.9567	0.9320	0.8612	0.8450	0.9952	0.8282	0.7761	0.9323	0.9839	0.8869	0.9925	0.9387	0.8360	0.9268	0.9723	0.9702	0.9372	0.9712	0.8759	0.9190	0.9859	0.9909	0.8835	0.9511	0.9697	0.9430
0.0225	7.0	9093	0.0850	0.8795	0.9113	0.8951	0.9819	0.9896	0.9936	0.8952	0.9746	0.8916	0.9545	0.9595	0.9505	0.8040	0.9581	0.8362	0.9919	0.8677	0.7402	0.7932	0.2246	0.9097	0.8339	0.8964	0.9503	1.0	0.9396	0.8748	0.9796	0.9829	0.9604	0.8682	0.8297	0.7449	0.9147	0.9594	0.9281	0.8581	0.8466	0.9952	0.8331	0.7916	0.9373	0.9861	0.8877	0.9925	0.9406	0.8546	0.9289	0.9733	0.9718	0.9330	0.9756	0.8799	0.9217	0.9868	0.9932	0.8802	0.9593	0.9622	0.9459

Framework versions

Transformers 4.40.1
Pytorch 2.2.2+cu121
Datasets 2.19.0
Tokenizers 0.19.1