resnet-Mistral-SYDNEY-without-captioning

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

• Loss: 1.0132
• Accuracy: 67.45
• Bleu-1: 0.7790
• Bleu-2: 0.7040
• Bleu-3: 0.6455
• Bleu-4: 0.5961
• Meteor: 0.7679
• Rouge-l: 0.7291
• Cider: 2.3684

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 50
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 128
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Bleu-1	Bleu-2	Bleu-3	Bleu-4	Meteor	Rouge-l	Cider
No log	1.0	44	3.9895	54.51	0.1725	0.0594	0.0180	0.0057	0.0872	0.1810	0.0242
No log	2.0	88	3.5357	43.72	0.2157	0.1198	0.0669	0.0297	0.1452	0.2122	0.0773
No log	3.0	132	1.4116	63.8	0.6962	0.5891	0.4974	0.4249	0.5715	0.6016	1.6083
No log	4.0	176	0.9785	67.01	0.7589	0.6755	0.5956	0.5217	0.6724	0.6717	2.1196
No log	5.0	220	0.7982	64.82	0.7997	0.7121	0.6334	0.5636	0.7225	0.7043	2.2921
No log	6.0	264	0.7939	67.92	0.7636	0.6681	0.5894	0.5193	0.6753	0.6591	2.1788
No log	7.0	308	0.8154	69.21	0.7660	0.6779	0.5971	0.5174	0.7131	0.6848	2.2232
No log	8.0	352	0.7648	67.95	0.7386	0.6424	0.5657	0.4995	0.7157	0.6624	2.2066
No log	9.0	396	0.8123	67.53	0.7811	0.6958	0.6228	0.5579	0.7215	0.6920	2.3312
No log	10.0	440	0.8544	66.57	0.7746	0.6866	0.6065	0.5347	0.7226	0.7032	2.3121
No log	11.0	484	0.7928	66.27	0.7799	0.7148	0.6558	0.6010	0.7517	0.7235	2.5344
No log	12.0	528	0.8591	67.01	0.7561	0.6671	0.5968	0.5347	0.7405	0.7035	2.2104
No log	13.0	572	0.8828	66.66	0.7565	0.6645	0.5783	0.4983	0.6868	0.6634	2.0617
No log	14.0	616	0.8539	68.02	0.7872	0.6992	0.6225	0.5499	0.7018	0.6916	2.2318
No log	15.0	660	0.9093	67.96	0.8291	0.7658	0.7025	0.6428	0.7814	0.7529	2.5036
No log	16.0	704	0.9806	68.8	0.8115	0.7334	0.6616	0.5931	0.7485	0.7278	2.4942
No log	17.0	748	0.9556	67.99	0.8059	0.7356	0.6765	0.6207	0.7552	0.7006	2.4525
No log	18.0	792	0.9694	67.28	0.7731	0.6824	0.5985	0.5244	0.7041	0.7011	2.2946
No log	19.0	836	0.8887	67.15	0.8254	0.7551	0.6834	0.6137	0.7537	0.7274	2.4964
No log	20.0	880	0.9509	66.89	0.7772	0.6978	0.6302	0.5678	0.7239	0.6890	2.2622
No log	21.0	924	1.0408	67.02	0.7959	0.7184	0.6400	0.5680	0.7255	0.6901	2.3351
No log	22.0	968	1.0163	68.2	0.8005	0.7407	0.6871	0.6373	0.7718	0.7341	2.8635
No log	23.0	1012	1.0269	66.95	0.7923	0.7205	0.6564	0.5981	0.7272	0.7003	2.6112
0.8197	24.0	1056	1.0350	68.25	0.8021	0.7365	0.6741	0.6172	0.7532	0.7336	2.5735
0.8197	25.0	1100	1.0132	67.45	0.7790	0.7040	0.6455	0.5961	0.7679	0.7291	2.3684

Framework versions

• Transformers 5.12.1
• Pytorch 2.12.1+cu130
• Datasets 5.0.0
• Tokenizers 0.22.2