resnet-50-finetuned-pokemon-finetuned-pokemon

This model is a fine-tuned version of TeeA/resnet-50-finetuned-pokemon on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 14.1746
• Accuracy: 0.0849

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: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1894	0.99	38	9.2115	0.0137
1.1389	1.99	76	9.2521	0.0129
1.0432	2.98	114	9.4765	0.0144
1.0625	4.0	153	9.7668	0.0137
1.0805	4.99	191	10.2526	0.0137
1.0353	5.99	229	10.3238	0.0129
0.9747	6.98	267	10.5779	0.0165
0.9708	8.0	306	10.7458	0.0180
0.8886	8.99	344	11.0072	0.0194
0.8408	9.99	382	11.3171	0.0223
0.802	10.98	420	11.5545	0.0245
0.7903	12.0	459	11.7722	0.0288
0.7553	12.99	497	11.9834	0.0353
0.7413	13.99	535	11.9815	0.0446
0.6272	14.98	573	12.0871	0.0496
0.6944	16.0	612	12.3713	0.0590
0.6322	16.99	650	12.6826	0.0554
0.6131	17.99	688	12.4819	0.0612
0.5916	18.98	726	12.6246	0.0647
0.5094	20.0	765	12.6641	0.0669
0.5201	20.99	803	12.8861	0.0662
0.4731	21.99	841	12.7431	0.0655
0.5132	22.98	879	12.7786	0.0705
0.5036	24.0	918	12.9990	0.0727
0.4863	24.99	956	13.0419	0.0727
0.4852	25.99	994	13.0573	0.0734
0.4983	26.98	1032	13.1310	0.0719
0.459	28.0	1071	13.0688	0.0748
0.4556	28.99	1109	13.4128	0.0748
0.4729	29.99	1147	13.3530	0.0741
0.4659	30.98	1185	13.2308	0.0763
0.4337	32.0	1224	13.3264	0.0748
0.456	32.99	1262	13.3506	0.0741
0.4423	33.99	1300	13.3607	0.0784
0.4037	34.98	1338	13.2521	0.0734
0.3891	36.0	1377	13.3702	0.0777
0.3992	36.99	1415	13.4762	0.0777
0.4014	37.99	1453	13.5382	0.0791
0.3549	38.98	1491	13.5550	0.0791
0.4048	40.0	1530	13.6406	0.0799
0.3711	40.99	1568	13.5120	0.0777
0.3834	41.99	1606	13.9230	0.0799
0.3475	42.98	1644	13.8602	0.0791
0.3465	44.0	1683	13.6931	0.0806
0.3682	44.99	1721	13.7774	0.0784
0.3613	45.99	1759	14.0235	0.0791
0.368	46.98	1797	13.9289	0.0813
0.3961	48.0	1836	14.2549	0.0806
0.365	48.99	1874	14.1114	0.0813
0.3259	49.99	1912	13.9710	0.0806
0.2998	50.98	1950	14.0288	0.0806
0.3203	52.0	1989	13.9398	0.0813
0.3104	52.99	2027	14.0255	0.0820
0.3232	53.99	2065	13.9355	0.0827
0.3521	54.98	2103	13.8627	0.0806
0.3322	56.0	2142	14.0179	0.0806
0.3129	56.99	2180	13.9640	0.0820
0.3159	57.99	2218	14.1997	0.0799
0.3118	58.98	2256	14.1639	0.0820
0.3196	60.0	2295	14.0334	0.0806
0.301	60.99	2333	13.9954	0.0820
0.3142	61.99	2371	14.1432	0.0799
0.3192	62.98	2409	14.0269	0.0784
0.3342	64.0	2448	14.0450	0.0806
0.3045	64.99	2486	14.1746	0.0849
0.2991	65.99	2524	14.3192	0.0806
0.3228	66.98	2562	14.1782	0.0784
0.2711	68.0	2601	14.4261	0.0849
0.2473	68.99	2639	14.2303	0.0827
0.3287	69.99	2677	14.2750	0.0827
0.2673	70.98	2715	14.2303	0.0820
0.2843	72.0	2754	14.4086	0.0806
0.3099	72.99	2792	14.5184	0.0827
0.3102	73.99	2830	14.2768	0.0835
0.2911	74.98	2868	14.1010	0.0835
0.2927	76.0	2907	14.4618	0.0813
0.2967	76.99	2945	14.3581	0.0820
0.2446	77.99	2983	14.4562	0.0835
0.3035	78.98	3021	14.2681	0.0835
0.2989	80.0	3060	14.2768	0.0827
0.2486	80.99	3098	14.4242	0.0820
0.2622	81.99	3136	14.3810	0.0835
0.2892	82.98	3174	14.4637	0.0827
0.2668	84.0	3213	14.4597	0.0835
0.2527	84.99	3251	14.3098	0.0820
0.2636	85.99	3289	14.3741	0.0835
0.247	86.98	3327	14.5369	0.0842
0.2693	88.0	3366	14.4039	0.0835
0.2692	88.99	3404	14.6161	0.0835
0.28	89.99	3442	14.5244	0.0835
0.2535	90.98	3480	14.4062	0.0842
0.2887	92.0	3519	14.4113	0.0806
0.257	92.99	3557	14.3442	0.0842
0.2627	93.99	3595	14.4693	0.0835
0.2804	94.98	3633	14.3223	0.0835
0.2529	96.0	3672	14.3844	0.0835
0.2327	96.99	3710	14.4284	0.0835
0.2643	97.99	3748	14.5567	0.0835
0.284	98.98	3786	14.6738	0.0813
0.2503	99.35	3800	14.5363	0.0842

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0