Sentiment140_DistilBERT_5E

This model is a fine-tuned version of distilbert-base-uncased on the sentiment140 dataset. It achieves the following results on the evaluation set:

Loss: 0.4897
Accuracy: 0.8333

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-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6784	0.08	50	0.6516	0.6933
0.6301	0.16	100	0.5384	0.7533
0.5438	0.24	150	0.4559	0.8
0.4625	0.32	200	0.4287	0.8133
0.4528	0.4	250	0.4056	0.8267
0.4609	0.48	300	0.3883	0.8333
0.4705	0.56	350	0.3886	0.8067
0.4539	0.64	400	0.3967	0.82
0.4483	0.72	450	0.3758	0.82
0.4699	0.8	500	0.4003	0.8133
0.467	0.88	550	0.4021	0.8267
0.454	0.96	600	0.3735	0.8333
0.4227	1.04	650	0.3840	0.8267
0.3584	1.12	700	0.3775	0.8333
0.3618	1.2	750	0.4026	0.8267
0.3634	1.28	800	0.3891	0.8133
0.3751	1.36	850	0.3895	0.8267
0.3484	1.44	900	0.3919	0.8267
0.3764	1.52	950	0.3770	0.84
0.3488	1.6	1000	0.4028	0.82
0.3665	1.68	1050	0.3779	0.8333
0.3925	1.76	1100	0.3726	0.84
0.3624	1.84	1150	0.3655	0.84
0.3876	1.92	1200	0.3648	0.8133
0.3935	2.0	1250	0.3633	0.8467
0.2944	2.08	1300	0.3808	0.8333
0.2957	2.16	1350	0.3836	0.8333
0.266	2.24	1400	0.3940	0.8267
0.2747	2.32	1450	0.3952	0.84
0.314	2.4	1500	0.4060	0.8133
0.3419	2.48	1550	0.4025	0.8133
0.2782	2.56	1600	0.4218	0.82
0.3218	2.64	1650	0.4039	0.8333
0.2863	2.72	1700	0.4130	0.8267
0.3336	2.8	1750	0.4026	0.8133
0.3224	2.88	1800	0.3910	0.8267
0.2709	2.96	1850	0.3979	0.84
0.2701	3.04	1900	0.4127	0.8333
0.2782	3.12	1950	0.4335	0.82
0.2425	3.2	2000	0.4229	0.8333
0.2457	3.28	2050	0.4168	0.8333
0.217	3.36	2100	0.4264	0.8267
0.2522	3.44	2150	0.4250	0.8333
0.2402	3.52	2200	0.4371	0.8333
0.2465	3.6	2250	0.4429	0.8333
0.2427	3.68	2300	0.4435	0.8333
0.2408	3.76	2350	0.4500	0.84
0.1976	3.84	2400	0.4536	0.8333
0.23	3.92	2450	0.4645	0.8333
0.2449	4.0	2500	0.4557	0.8467
0.1933	4.08	2550	0.4672	0.84
0.213	4.16	2600	0.4717	0.84
0.1772	4.24	2650	0.4843	0.8267
0.1917	4.32	2700	0.4690	0.8467
0.2094	4.4	2750	0.4728	0.8467
0.1903	4.48	2800	0.4755	0.8467
0.2541	4.56	2850	0.4791	0.84
0.1805	4.64	2900	0.4877	0.84
0.2183	4.72	2950	0.4940	0.8267
0.2257	4.8	3000	0.4905	0.8333
0.2496	4.88	3050	0.4883	0.84
0.1846	4.96	3100	0.4897	0.8333

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu113
Datasets 2.6.1
Tokenizers 0.13.1

Etelis/Sentiment140_DistilBERT_5E

Sentiment140_DistilBERT_5E

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model