bart-large-cnn-finetuned-prompt_generation

This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.6474
• Actual score: 0.8766
• Predction score: 0.3367
• Score difference: 0.5399

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: 3e-07
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Actual score	Predction score	Score difference
No log	1.0	15	3.6226	0.8766	-0.4072	1.2838
No log	2.0	30	3.5120	0.8766	-0.2477	1.1243
No log	3.0	45	3.3572	0.8766	-0.3233	1.1999
No log	4.0	60	3.2592	0.8766	-0.0494	0.9260
No log	5.0	75	3.1430	0.8766	-0.3234	1.2000
No log	6.0	90	3.0581	0.8766	-0.4732	1.3498
No log	7.0	105	2.9988	0.8766	-0.5715	1.4481
No log	8.0	120	2.9564	0.8766	-0.6699	1.5465
No log	9.0	135	2.9242	0.8766	-0.5505	1.4271
No log	10.0	150	2.8969	0.8766	-0.4393	1.3159
No log	11.0	165	2.8729	0.8766	-0.4882	1.3648
No log	12.0	180	2.8503	0.8766	-0.6554	1.5320
No log	13.0	195	2.8308	0.8766	-0.7288	1.6054
No log	14.0	210	2.8128	0.8766	-0.7016	1.5783
No log	15.0	225	2.7972	0.8766	-0.7900	1.6666
No log	16.0	240	2.7832	0.8766	-0.6285	1.5052
No log	17.0	255	2.7708	0.8766	-0.5613	1.4379
No log	18.0	270	2.7591	0.8766	-0.6125	1.4891
No log	19.0	285	2.7481	0.8766	-0.5101	1.3868
No log	20.0	300	2.7390	0.8766	-0.4879	1.3646
No log	21.0	315	2.7307	0.8766	-0.4345	1.3112
No log	22.0	330	2.7229	0.8766	-0.3278	1.2044
No log	23.0	345	2.7156	0.8766	-0.3324	1.2090
No log	24.0	360	2.7084	0.8766	-0.2899	1.1665
No log	25.0	375	2.7019	0.8766	-0.1728	1.0494
No log	26.0	390	2.6965	0.8766	-0.2785	1.1552
No log	27.0	405	2.6918	0.8766	-0.1926	1.0692
No log	28.0	420	2.6872	0.8766	-0.1204	0.9970
No log	29.0	435	2.6832	0.8766	-0.0040	0.8806
No log	30.0	450	2.6791	0.8766	-0.0742	0.9508
No log	31.0	465	2.6751	0.8766	0.0669	0.8097
No log	32.0	480	2.6719	0.8766	-0.0049	0.8815
No log	33.0	495	2.6690	0.8766	-0.0196	0.8962
2.6809	34.0	510	2.6663	0.8766	0.0692	0.8074
2.6809	35.0	525	2.6636	0.8766	0.0843	0.7923
2.6809	36.0	540	2.6615	0.8766	-0.0330	0.9096
2.6809	37.0	555	2.6594	0.8766	-0.0065	0.8831
2.6809	38.0	570	2.6575	0.8766	0.2102	0.6664
2.6809	39.0	585	2.6559	0.8766	0.3005	0.5761
2.6809	40.0	600	2.6541	0.8766	0.3360	0.5406
2.6809	41.0	615	2.6528	0.8766	0.2456	0.6310
2.6809	42.0	630	2.6517	0.8766	0.3399	0.5367
2.6809	43.0	645	2.6509	0.8766	0.4224	0.4542
2.6809	44.0	660	2.6499	0.8766	0.4277	0.4490
2.6809	45.0	675	2.6492	0.8766	0.2815	0.5951
2.6809	46.0	690	2.6485	0.8766	0.3053	0.5714
2.6809	47.0	705	2.6481	0.8766	0.2149	0.6618
2.6809	48.0	720	2.6478	0.8766	0.2285	0.6481
2.6809	49.0	735	2.6475	0.8766	0.2546	0.6220
2.6809	50.0	750	2.6474	0.8766	0.3367	0.5399

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1