scenario-KD-PO-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual_all66

This model is a fine-tuned version of haryoaw/scenario-MDBT-TCR_data-cardiffnlp_tweet_sentiment_multilingual_all on the tweet_sentiment_multilingual dataset. It achieves the following results on the evaluation set:

• Loss: 3.1262
• Accuracy: 0.6046
• F1: 0.6045

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: 32
• eval_batch_size: 32
• seed: 66
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
4.3043	1.09	500	3.2421	0.5683	0.5687
3.0954	2.17	1000	3.1223	0.5702	0.5722
2.2948	3.26	1500	3.3254	0.5741	0.5690
1.7221	4.35	2000	3.3734	0.5910	0.5892
1.3735	5.43	2500	3.3678	0.5683	0.5700
1.1599	6.52	3000	3.4642	0.5872	0.5861
1.0074	7.61	3500	3.3355	0.5795	0.5815
0.8728	8.7	4000	3.2965	0.5856	0.5882
0.8096	9.78	4500	3.4304	0.5799	0.5793
0.7405	10.87	5000	3.3412	0.5856	0.5849
0.6749	11.96	5500	3.2449	0.5829	0.5852
0.6256	13.04	6000	3.0573	0.5992	0.5989
0.591	14.13	6500	3.2229	0.5860	0.5861
0.5531	15.22	7000	3.2349	0.5903	0.5921
0.531	16.3	7500	3.2751	0.5968	0.5966
0.5156	17.39	8000	3.2324	0.5945	0.5960
0.482	18.48	8500	3.3781	0.6007	0.6005
0.4724	19.57	9000	3.1535	0.5883	0.5894
0.4587	20.65	9500	3.3708	0.5910	0.5903
0.4372	21.74	10000	3.1164	0.5995	0.6006
0.4281	22.83	10500	3.0913	0.5907	0.5917
0.4151	23.91	11000	3.0730	0.6053	0.6050
0.407	25.0	11500	3.1531	0.5984	0.5988
0.3966	26.09	12000	3.1692	0.6053	0.6044
0.3936	27.17	12500	3.3192	0.6026	0.6019
0.3848	28.26	13000	3.1133	0.6080	0.6085
0.3754	29.35	13500	3.1272	0.5965	0.5970
0.3734	30.43	14000	3.2430	0.6026	0.6026
0.3685	31.52	14500	3.0758	0.6107	0.6112
0.3617	32.61	15000	3.1241	0.5965	0.5975
0.3592	33.7	15500	3.2830	0.5926	0.5929
0.3545	34.78	16000	3.1901	0.6038	0.6035
0.3486	35.87	16500	3.2132	0.5980	0.5985
0.3502	36.96	17000	3.1464	0.5999	0.6003
0.3458	38.04	17500	3.2477	0.6080	0.6074
0.341	39.13	18000	3.2192	0.5918	0.5926
0.3413	40.22	18500	3.1910	0.6111	0.6114
0.3355	41.3	19000	3.1159	0.6065	0.6072
0.3381	42.39	19500	3.1962	0.6007	0.6008
0.3308	43.48	20000	3.1165	0.6061	0.6064
0.3326	44.57	20500	3.2242	0.6076	0.6072
0.3331	45.65	21000	3.2347	0.6046	0.6046
0.3319	46.74	21500	3.2428	0.5941	0.5941
0.3282	47.83	22000	3.1621	0.6053	0.6058
0.3275	48.91	22500	3.2598	0.6042	0.6044
0.3291	50.0	23000	3.1262	0.6046	0.6045

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3