whatsapp-group-classifierv3

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

• Loss: 0.4325
• Accuracy: 0.8454
• Precision: 0.8680
• Recall: 0.8522
• F1: 0.8589

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: 4e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.9598	1.0	513	0.6553	0.7366	0.7724	0.7280	0.7280
0.6517	2.0	1026	0.5718	0.7751	0.7888	0.7802	0.7817
0.5707	3.0	1539	0.5134	0.8024	0.8214	0.8059	0.8115
0.5246	4.0	2052	0.4874	0.8146	0.8376	0.8151	0.8243
0.4773	5.0	2565	0.4717	0.8215	0.8417	0.8295	0.8344
0.4512	6.0	3078	0.4586	0.8244	0.8465	0.8330	0.8389
0.4496	7.0	3591	0.4534	0.8332	0.8538	0.8380	0.8450
0.4164	8.0	4104	0.4432	0.8366	0.8615	0.8412	0.8501
0.4184	9.0	4617	0.4396	0.8356	0.8601	0.8407	0.8493
0.4075	10.0	5130	0.4346	0.8332	0.8563	0.8418	0.8480
0.3923	11.0	5643	0.4329	0.8395	0.8614	0.8453	0.8519
0.3886	12.0	6156	0.4367	0.8390	0.8623	0.8450	0.8525
0.3792	13.0	6669	0.4248	0.8390	0.8621	0.8431	0.8512
0.3659	14.0	7182	0.4252	0.84	0.8624	0.8483	0.8539
0.3738	15.0	7695	0.4236	0.8376	0.8606	0.8445	0.8515
0.3502	16.0	8208	0.4308	0.8444	0.8651	0.8521	0.8575
0.3574	17.0	8721	0.4292	0.8439	0.8658	0.8504	0.8572
0.3521	18.0	9234	0.4266	0.8449	0.8667	0.8512	0.8581
0.3306	19.0	9747	0.4247	0.8415	0.8666	0.8450	0.8543
0.3581	20.0	10260	0.4316	0.8449	0.8692	0.8531	0.8596
0.3223	21.0	10773	0.4342	0.8483	0.8713	0.8565	0.8624
0.3235	22.0	11286	0.4270	0.8473	0.8712	0.8551	0.8616
0.3281	23.0	11799	0.4263	0.8449	0.8686	0.8496	0.8579
0.313	24.0	12312	0.4319	0.8463	0.8677	0.8538	0.8598
0.32	25.0	12825	0.4305	0.8468	0.8681	0.8539	0.8600
0.3104	26.0	13338	0.4280	0.8488	0.8708	0.8543	0.8616
0.3203	27.0	13851	0.4309	0.8473	0.8698	0.8547	0.8611
0.304	28.0	14364	0.4298	0.8444	0.8671	0.8494	0.8572
0.3057	29.0	14877	0.4358	0.8498	0.8720	0.8575	0.8634
0.3238	30.0	15390	0.4325	0.8454	0.8680	0.8522	0.8589

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0