learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 16
seed: 42
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
num_epochs: 30

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 Per Class	Mar 100 Per Class
No log	1.0	20	2.0915	0.0081	0.0263	0.0046	-1.0	0.0	0.0147	0.1116	0.2395	0.4535	-1.0	0.0	0.5132	0.0081	0.4535
No log	2.0	40	1.0759	0.0323	0.0709	0.026	-1.0	0.0427	0.0465	0.286	0.5302	0.7326	-1.0	0.5	0.7632	0.0323	0.7326
No log	3.0	60	0.8394	0.1187	0.1754	0.1552	-1.0	0.3952	0.0873	0.5116	0.686	0.8186	-1.0	0.84	0.8158	0.1187	0.8186
No log	4.0	80	0.8340	0.2123	0.3611	0.2429	-1.0	0.3145	0.1944	0.4907	0.6767	0.8023	-1.0	0.74	0.8105	0.2123	0.8023
No log	5.0	100	0.8458	0.3631	0.5562	0.401	-1.0	0.4248	0.3693	0.5395	0.6488	0.8209	-1.0	0.7	0.8368	0.3631	0.8209
No log	6.0	120	0.8974	0.5083	0.7635	0.5792	-1.0	0.5446	0.5075	0.5628	0.7256	0.793	-1.0	0.74	0.8	0.5083	0.793
No log	7.0	140	0.7325	0.5733	0.8059	0.6774	-1.0	0.6707	0.5778	0.6395	0.7791	0.8488	-1.0	0.8	0.8553	0.5733	0.8488
No log	8.0	160	0.7845	0.533	0.8377	0.6465	-1.0	0.6044	0.5473	0.586	0.7651	0.8395	-1.0	0.88	0.8342	0.533	0.8395
No log	9.0	180	0.6819	0.6434	0.9182	0.768	-1.0	0.6145	0.6583	0.6233	0.814	0.8674	-1.0	0.74	0.8842	0.6434	0.8674
No log	10.0	200	0.6374	0.6437	0.8979	0.7412	-1.0	0.6576	0.652	0.6674	0.8093	0.8791	-1.0	0.78	0.8921	0.6437	0.8791
No log	11.0	220	0.6894	0.6307	0.9119	0.7639	-1.0	0.6304	0.638	0.6442	0.7907	0.8605	-1.0	0.78	0.8711	0.6307	0.8605
No log	12.0	240	0.6690	0.6607	0.9295	0.7626	-1.0	0.6879	0.6676	0.6698	0.8186	0.8628	-1.0	0.78	0.8737	0.6607	0.8628
No log	13.0	260	0.7227	0.6474	0.9269	0.8003	-1.0	0.7434	0.6429	0.6372	0.807	0.8651	-1.0	0.82	0.8711	0.6474	0.8651
No log	14.0	280	0.6454	0.6822	0.9356	0.8332	-1.0	0.7348	0.6815	0.6698	0.814	0.8628	-1.0	0.82	0.8684	0.6822	0.8628
No log	15.0	300	0.6530	0.6647	0.929	0.81	-1.0	0.6964	0.6692	0.6558	0.8047	0.8605	-1.0	0.82	0.8658	0.6647	0.8605
No log	16.0	320	0.6420	0.684	0.9337	0.8505	-1.0	0.6801	0.6834	0.6628	0.8163	0.8674	-1.0	0.8	0.8763	0.684	0.8674
No log	17.0	340	0.6138	0.6939	0.9386	0.8481	-1.0	0.7017	0.7017	0.6488	0.8256	0.8628	-1.0	0.76	0.8763	0.6939	0.8628
No log	18.0	360	0.6283	0.6955	0.9437	0.8341	-1.0	0.6738	0.709	0.6372	0.8326	0.8767	-1.0	0.76	0.8921	0.6955	0.8767
No log	19.0	380	0.6072	0.6929	0.9363	0.8163	-1.0	0.7504	0.6923	0.6674	0.8209	0.886	-1.0	0.86	0.8895	0.6929	0.886
No log	20.0	400	0.6230	0.6937	0.9443	0.8531	-1.0	0.7029	0.7032	0.6698	0.8372	0.8651	-1.0	0.8	0.8737	0.6937	0.8651
No log	21.0	420	0.6391	0.7019	0.95	0.8383	-1.0	0.7245	0.7081	0.6744	0.8349	0.8605	-1.0	0.8	0.8684	0.7019	0.8605
No log	22.0	440	0.6134	0.7091	0.9467	0.8429	-1.0	0.7441	0.7102	0.6837	0.8372	0.8814	-1.0	0.82	0.8895	0.7091	0.8814
No log	23.0	460	0.6170	0.6998	0.9467	0.8443	-1.0	0.6879	0.706	0.6744	0.8256	0.8814	-1.0	0.82	0.8895	0.6998	0.8814
No log	24.0	480	0.6257	0.701	0.9524	0.8573	-1.0	0.7236	0.7059	0.6791	0.8279	0.8698	-1.0	0.78	0.8816	0.701	0.8698
0.7187	25.0	500	0.6078	0.6936	0.9469	0.8465	-1.0	0.692	0.7054	0.6721	0.8349	0.8674	-1.0	0.76	0.8816	0.6936	0.8674
0.7187	26.0	520	0.6108	0.6939	0.9532	0.8567	-1.0	0.663	0.705	0.6767	0.8302	0.8674	-1.0	0.76	0.8816	0.6939	0.8674
0.7187	27.0	540	0.6056	0.6962	0.9546	0.8568	-1.0	0.6702	0.708	0.6767	0.8349	0.8744	-1.0	0.78	0.8868	0.6962	0.8744
0.7187	28.0	560	0.6028	0.6966	0.9556	0.8564	-1.0	0.6705	0.7084	0.6767	0.8372	0.8744	-1.0	0.78	0.8868	0.6966	0.8744
0.7187	29.0	580	0.6028	0.6949	0.9542	0.8545	-1.0	0.6705	0.706	0.6767	0.8349	0.8767	-1.0	0.78	0.8895	0.6949	0.8767
0.7187	30.0	600	0.6027	0.6948	0.9542	0.8543	-1.0	0.6705	0.7059	0.6767	0.8349	0.8767	-1.0	0.78	0.8895	0.6948	0.8767

Framework versions

Transformers 4.57.6
Pytorch 2.11.0+cu130
Datasets 2.14.3
Tokenizers 0.22.2

javillo-ur/yolo_finetuned_raccoon

yolo_finetuned_raccoon

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

javillo-ur/
yolo_finetuned_raccoon