FetalPlane_Classifcation-6-V2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2863
• Accuracy: 0.9091
• Precision Macro: 0.8809
• Recall Macro: 0.9155
• F1 Macro: 0.8947

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: 0.0001985679720704591
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: cosine
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro
No log	1.0	242	0.5375	0.7947	0.7726	0.8119	0.7775
No log	2.0	484	0.4322	0.8327	0.7936	0.8493	0.8152
0.6099	3.0	726	0.4742	0.7887	0.7740	0.8388	0.7832
0.6099	4.0	968	0.3931	0.8347	0.7950	0.8570	0.8171
0.5537	5.0	1210	0.4078	0.8331	0.8100	0.8541	0.8225
0.5537	6.0	1452	0.3703	0.8578	0.8245	0.8673	0.8415
0.4986	7.0	1694	0.3924	0.8392	0.8110	0.8637	0.8258
0.4986	8.0	1936	0.3603	0.8663	0.8332	0.8801	0.8517
0.4486	9.0	2178	0.3675	0.8655	0.8290	0.8838	0.8494
0.4486	10.0	2420	0.3744	0.8501	0.8140	0.8796	0.8334
0.4238	11.0	2662	0.3503	0.8630	0.8223	0.8828	0.8442
0.4238	12.0	2904	0.3909	0.8646	0.8303	0.8823	0.8483
0.4094	13.0	3146	0.3271	0.8784	0.8388	0.8929	0.8594
0.4094	14.0	3388	0.3249	0.8812	0.8446	0.8932	0.8639
0.3875	15.0	3630	0.3301	0.8816	0.8554	0.8900	0.8683
0.3875	16.0	3872	0.3408	0.8752	0.8437	0.8943	0.8622
0.3662	17.0	4114	0.3121	0.8812	0.8435	0.8991	0.8648
0.3662	18.0	4356	0.3288	0.8731	0.8346	0.8942	0.8560
0.3491	19.0	4598	0.3577	0.8703	0.8336	0.8823	0.8513
0.3491	20.0	4840	0.3348	0.8808	0.8429	0.8946	0.8624
0.3476	21.0	5082	0.3089	0.8909	0.8650	0.8909	0.8757
0.3476	22.0	5324	0.3295	0.8990	0.8787	0.8913	0.8839
0.3226	23.0	5566	0.2954	0.8897	0.8551	0.9003	0.8734
0.3226	24.0	5808	0.3097	0.8949	0.8710	0.8896	0.8793
0.3208	25.0	6050	0.2922	0.8986	0.8699	0.8998	0.8830
0.3208	26.0	6292	0.3117	0.8933	0.8685	0.8934	0.8796
0.3070	27.0	6534	0.3157	0.8893	0.8563	0.9013	0.8746
0.3070	28.0	6776	0.2962	0.8893	0.8551	0.9030	0.8746
0.2950	29.0	7018	0.3138	0.8869	0.8535	0.9068	0.8725
0.2950	30.0	7260	0.3003	0.8990	0.8688	0.9043	0.8834
0.2953	31.0	7502	0.3105	0.8873	0.8539	0.9016	0.8724
0.2953	32.0	7744	0.2776	0.8994	0.8670	0.9095	0.8847
0.2953	33.0	7986	0.2980	0.9006	0.8760	0.8993	0.8866
0.2886	34.0	8228	0.2808	0.8982	0.8663	0.9077	0.8838
0.2886	35.0	8470	0.2964	0.9002	0.8742	0.8985	0.8844
0.2721	36.0	8712	0.2870	0.9034	0.8759	0.9108	0.8910
0.2721	37.0	8954	0.2939	0.8990	0.8744	0.9035	0.8872
0.2607	38.0	9196	0.2882	0.9014	0.8750	0.9107	0.8901
0.2607	39.0	9438	0.3070	0.9018	0.8762	0.9021	0.8879
0.2688	40.0	9680	0.2801	0.9030	0.8773	0.9090	0.8909
0.2688	41.0	9922	0.2842	0.9010	0.8746	0.9048	0.8878
0.2509	42.0	10164	0.2892	0.8881	0.8568	0.9019	0.8751
0.2509	43.0	10406	0.2836	0.9030	0.8782	0.9057	0.8901
0.2446	44.0	10648	0.2812	0.9046	0.8750	0.9077	0.8886
0.2446	45.0	10890	0.2873	0.8901	0.8568	0.9056	0.8764
0.2456	46.0	11132	0.2863	0.9091	0.8809	0.9155	0.8947
0.2456	47.0	11374	0.3109	0.9002	0.8742	0.9035	0.8865
0.2279	48.0	11616	0.2938	0.9046	0.8783	0.9062	0.8902
0.2279	49.0	11858	0.3084	0.9030	0.8759	0.8990	0.8864
0.2317	50.0	12100	0.2954	0.9010	0.8703	0.9093	0.8866
0.2317	51.0	12342	0.2771	0.9046	0.8750	0.9101	0.8904
0.2184	52.0	12584	0.2850	0.9038	0.8767	0.9035	0.8887
0.2184	53.0	12826	0.2792	0.9051	0.8772	0.9071	0.8904
0.2100	54.0	13068	0.2892	0.9038	0.8754	0.9042	0.8883
0.2100	55.0	13310	0.2730	0.9055	0.8741	0.9125	0.8905
0.2124	56.0	13552	0.3055	0.9030	0.8789	0.9013	0.8890

Framework versions

• Transformers 5.17.0
• Pytorch 2.11.0+cu128
• Datasets 5.0.1
• Tokenizers 0.23.1