britannica-illustration-detector-smoke

This model is a fine-tuned version of ustc-community/dfine-small-coco on the davanstrien/britannica-illustration-labels-coco dataset. It achieves the following results on the evaluation set:

Loss: 1.7220
Map: 0.7702
Map 50: 0.9069
Map 75: 0.828
Map Small: 0.4046
Map Medium: 0.6092
Map Large: 0.807
Mar 1: 0.1333
Mar 10: 0.5722
Mar 100: 0.853
Mar Small: 0.4667
Mar Medium: 0.7401
Mar Large: 0.883
Map Illustration: 0.7702
Mar 100 Illustration: 0.853

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: 8
eval_batch_size: 8
seed: 42
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
num_epochs: 30.0

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Illustration	Mar 100 Illustration
No log	1.0	63	8.5381	0.2898	0.3759	0.3083	0.0	0.122	0.3339	0.0834	0.3654	0.6762	0.0	0.3857	0.7498	0.2898	0.6762
No log	2.0	126	4.0089	0.6347	0.7777	0.6919	0.002	0.3863	0.6992	0.1227	0.5438	0.8009	0.0444	0.5369	0.8694	0.6347	0.8009
No log	3.0	189	2.9852	0.7099	0.8585	0.7667	0.1496	0.5173	0.7603	0.1284	0.5648	0.8281	0.1778	0.6516	0.8753	0.7099	0.8281
No log	4.0	252	2.5071	0.74	0.8825	0.7998	0.2527	0.5553	0.7866	0.1311	0.5695	0.8388	0.3667	0.6742	0.8815	0.74	0.8388
No log	5.0	315	2.2600	0.7436	0.8899	0.8046	0.2361	0.5774	0.7856	0.1313	0.5687	0.8434	0.3222	0.7037	0.8809	0.7436	0.8434
No log	6.0	378	2.0769	0.7529	0.8992	0.8139	0.2633	0.597	0.7925	0.1302	0.5686	0.8537	0.3333	0.729	0.8877	0.7529	0.8537
No log	7.0	441	1.9959	0.7556	0.9008	0.8187	0.3664	0.5957	0.7948	0.1308	0.569	0.8521	0.4333	0.7226	0.8863	0.7556	0.8521
16.6582	8.0	504	1.9436	0.7554	0.9005	0.8209	0.3439	0.5972	0.7937	0.1308	0.5712	0.8557	0.4333	0.7369	0.8873	0.7554	0.8557
16.6582	9.0	567	1.9043	0.7665	0.9077	0.8303	0.3847	0.6042	0.8043	0.1326	0.5748	0.8565	0.5222	0.7355	0.8879	0.7665	0.8565
16.6582	10.0	630	1.8505	0.77	0.9106	0.8328	0.3676	0.6147	0.8073	0.1314	0.5748	0.8587	0.4333	0.741	0.8902	0.77	0.8587
16.6582	11.0	693	1.8438	0.7601	0.8952	0.8166	0.4224	0.5963	0.7985	0.1322	0.5666	0.8539	0.4889	0.7304	0.8861	0.7601	0.8539
16.6582	12.0	756	1.7951	0.7666	0.9069	0.8275	0.3875	0.6137	0.8027	0.1337	0.5714	0.8577	0.4556	0.7442	0.888	0.7666	0.8577
16.6582	13.0	819	1.8245	0.7633	0.9031	0.8228	0.4024	0.6035	0.8007	0.1324	0.5675	0.8554	0.4778	0.7369	0.8866	0.7633	0.8554
16.6582	14.0	882	1.7966	0.7649	0.9033	0.8239	0.3922	0.6045	0.803	0.1341	0.5688	0.8593	0.4667	0.7364	0.8916	0.7649	0.8593
16.6582	15.0	945	1.7978	0.7628	0.9026	0.8185	0.4056	0.6056	0.8002	0.1321	0.5702	0.858	0.4778	0.741	0.8888	0.7628	0.858
9.2678	16.0	1008	1.7664	0.765	0.8989	0.8212	0.4027	0.5903	0.8052	0.1339	0.5728	0.8581	0.4667	0.7332	0.8909	0.765	0.8581
9.2678	17.0	1071	1.7589	0.7656	0.9041	0.8275	0.3899	0.6029	0.8029	0.133	0.5698	0.8562	0.4667	0.7424	0.8864	0.7656	0.8562
9.2678	18.0	1134	1.7783	0.7583	0.896	0.8137	0.4044	0.5913	0.7981	0.1321	0.569	0.8523	0.4667	0.7276	0.885	0.7583	0.8523
9.2678	19.0	1197	1.7526	0.7656	0.9054	0.8228	0.3983	0.6047	0.8034	0.1321	0.5709	0.8579	0.4778	0.7355	0.89	0.7656	0.8579
9.2678	20.0	1260	1.7508	0.7646	0.9018	0.8209	0.4031	0.5991	0.8025	0.1307	0.571	0.8515	0.4667	0.735	0.8823	0.7646	0.8515
9.2678	21.0	1323	1.7394	0.7701	0.9045	0.825	0.3899	0.6077	0.8081	0.1323	0.5722	0.8539	0.4667	0.735	0.8853	0.7701	0.8539
9.2678	22.0	1386	1.7394	0.7655	0.9029	0.8197	0.3924	0.5963	0.8049	0.1337	0.5672	0.851	0.4667	0.7286	0.8832	0.7655	0.851
9.2678	23.0	1449	1.7332	0.7638	0.9	0.8194	0.3941	0.5944	0.8039	0.1321	0.5712	0.8498	0.4667	0.724	0.8828	0.7638	0.8498
8.6225	24.0	1512	1.7530	0.763	0.9009	0.8209	0.3899	0.6027	0.8005	0.1328	0.5702	0.8509	0.4667	0.7359	0.8814	0.763	0.8509
8.6225	25.0	1575	1.7355	0.7638	0.8993	0.8208	0.3872	0.5984	0.8029	0.1342	0.5689	0.8503	0.4556	0.7263	0.8829	0.7638	0.8503
8.6225	26.0	1638	1.7490	0.7668	0.9027	0.8224	0.399	0.6074	0.8039	0.1346	0.57	0.8523	0.4667	0.741	0.8819	0.7668	0.8523
8.6225	27.0	1701	1.7417	0.7662	0.9042	0.8201	0.4046	0.6022	0.8036	0.1339	0.5704	0.8536	0.4667	0.741	0.8835	0.7662	0.8536
8.6225	28.0	1764	1.7246	0.769	0.9059	0.8227	0.4046	0.606	0.8064	0.1332	0.5713	0.8539	0.4667	0.7369	0.8849	0.769	0.8539
8.6225	29.0	1827	1.7233	0.7703	0.907	0.828	0.4046	0.6083	0.8074	0.1333	0.5723	0.8527	0.4667	0.7396	0.8827	0.7703	0.8527
8.6225	30.0	1890	1.7302	0.7669	0.9048	0.8237	0.4046	0.6041	0.8039	0.1339	0.5714	0.8534	0.4667	0.7378	0.884	0.7669	0.8534

Framework versions

Transformers 5.15.1
Pytorch 2.13.0+cu130
Datasets 5.0.1
Tokenizers 0.22.2

davanstrien/britannica-illustration-detector-smoke

britannica-illustration-detector-smoke

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种