mobilenet_v2-activity-recognition

This model is a fine-tuned version of google/mobilenet_v2_1.0_224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0450
• Accuracy: 0.6718

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.8136	0.1778	10	2.7919	0.0733
2.8041	0.3556	20	2.7240	0.1043
2.6841	0.5333	30	2.6304	0.1421
2.5799	0.7111	40	2.4856	0.2497
2.4537	0.8889	50	2.3143	0.3431
2.2593	1.0667	60	2.1425	0.4005
2.0671	1.2444	70	1.9995	0.4360
1.8958	1.4222	80	1.8545	0.4683
1.7891	1.6	90	1.7437	0.4939
1.6659	1.7778	100	1.6373	0.5317
1.6006	1.9556	110	1.5372	0.5568
1.4752	2.1333	120	1.4766	0.5705
1.3654	2.3111	130	1.4303	0.5862
1.3452	2.4889	140	1.3513	0.6048
1.3134	2.6667	150	1.3941	0.5663
1.2905	2.8444	160	1.2859	0.6159
1.2201	3.0222	170	1.2661	0.6174
1.1225	3.2	180	1.2662	0.6181
1.0991	3.3778	190	1.1911	0.6392
1.1171	3.5556	200	1.2437	0.6142
1.0643	3.7333	210	1.1952	0.6318
1.1095	3.9111	220	1.1333	0.6519
1.0284	4.0889	230	1.1642	0.6362
0.9896	4.2667	240	1.1140	0.6519
0.9507	4.4444	250	1.0811	0.6672
0.9437	4.6222	260	1.0729	0.6652
0.9522	4.8	270	1.0724	0.6650
0.953	4.9778	280	1.0645	0.6713
0.8857	5.1556	290	1.1049	0.6508
0.907	5.3333	300	1.0808	0.6580
0.8723	5.5111	310	1.0437	0.6766
0.824	5.6889	320	1.0227	0.6801
0.846	5.8667	330	1.0186	0.6746
0.845	6.0444	340	1.0166	0.6805
0.8015	6.2222	350	1.0379	0.6720
0.8798	6.4	360	0.9889	0.6879
0.8076	6.5778	370	1.0059	0.6829
0.8105	6.7556	380	1.0098	0.6783
0.7414	6.9333	390	0.9801	0.6859
0.7869	7.1111	400	0.9624	0.6993
0.7728	7.2889	410	1.0938	0.6547
0.7762	7.4667	420	0.9867	0.6825
0.7769	7.6444	430	1.0512	0.6670
0.7563	7.8222	440	1.0346	0.6770
0.762	8.0	450	1.0647	0.6597
0.726	8.1778	460	1.0134	0.6812
0.7515	8.3556	470	0.9921	0.6787
0.7034	8.5333	480	1.0043	0.6833
0.7426	8.7111	490	0.9721	0.6936
0.7225	8.8889	500	1.0450	0.6718
0.7372	9.0667	510	0.9957	0.6812
0.7238	9.2444	520	0.9928	0.6894
0.7824	9.4222	530	1.0413	0.6753
0.7218	9.6	540	0.9717	0.6877
0.6976	9.7778	550	0.9839	0.6859
0.7288	9.9556	560	1.0229	0.6728

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1