SegFormer_b3_mappillary_

This model is a fine-tuned version of nvidia/segformer-b3-finetuned-cityscapes-1024-1024 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9471
• Mean Iou: 0.6940
• Mean Accuracy: 0.8072
• Overall Accuracy: 0.9431
• Accuracy Construction--barrier--fence: 0.7139
• Accuracy Construction--barrier--guard-rail: 0.7940
• Accuracy Construction--barrier--other-barrier: 0.7037
• Accuracy Construction--barrier--wall: 0.7066
• Accuracy Construction--flat--road: 0.9635
• Accuracy Construction--flat--service-lane: 0.6054
• Accuracy Construction--flat--sidewalk: 0.8864
• Accuracy Construction--structure--building: 0.9471
• Accuracy Human--person: 0.8464
• Accuracy Human--rider--bicyclist: 0.7717
• Accuracy Marking--crosswalk-zebra: 0.8305
• Accuracy Marking--general: 0.7034
• Accuracy Nature--sky: 0.9905
• Accuracy Nature--terrain: 0.8364
• Accuracy Nature--vegetation: 0.9463
• Accuracy Object--support--pole: 0.5949
• Accuracy Object--support--traffic-sign-frame: 0.6872
• Accuracy Object--traffic-light: 0.7613
• Accuracy Object--traffic-sign--front: 0.8302
• Accuracy Object--vehicle--bicycle: 0.7734
• Accuracy Object--vehicle--bus: 0.8816
• Accuracy Object--vehicle--car: 0.9634
• Accuracy Object--vehicle--truck: 0.8290
• Iou Construction--barrier--fence: 0.5854
• Iou Construction--barrier--guard-rail: 0.6329
• Iou Construction--barrier--other-barrier: 0.5635
• Iou Construction--barrier--wall: 0.5578
• Iou Construction--flat--road: 0.9223
• Iou Construction--flat--service-lane: 0.4671
• Iou Construction--flat--sidewalk: 0.7992
• Iou Construction--structure--building: 0.8928
• Iou Human--person: 0.6907
• Iou Human--rider--bicyclist: 0.5950
• Iou Marking--crosswalk-zebra: 0.7205
• Iou Marking--general: 0.5949
• Iou Nature--sky: 0.9814
• Iou Nature--terrain: 0.6899
• Iou Nature--vegetation: 0.8954
• Iou Object--support--pole: 0.4690
• Iou Object--support--traffic-sign-frame: 0.5685
• Iou Object--traffic-light: 0.5949
• Iou Object--traffic-sign--front: 0.7276
• Iou Object--vehicle--bicycle: 0.5790
• Iou Object--vehicle--bus: 0.8005
• Iou Object--vehicle--car: 0.9072
• Iou Object--vehicle--truck: 0.7273

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: 6e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy Construction--barrier--fence	Accuracy Construction--barrier--guard-rail	Accuracy Construction--barrier--other-barrier	Accuracy Construction--barrier--wall	Accuracy Construction--flat--road	Accuracy Construction--flat--service-lane	Accuracy Construction--flat--sidewalk	Accuracy Construction--structure--building	Accuracy Human--person	Accuracy Human--rider--bicyclist	Accuracy Marking--crosswalk-zebra	Accuracy Marking--general	Accuracy Nature--sky	Accuracy Nature--terrain	Accuracy Nature--vegetation	Accuracy Object--support--pole	Accuracy Object--support--traffic-sign-frame	Accuracy Object--traffic-light	Accuracy Object--traffic-sign--front	Accuracy Object--vehicle--bicycle	Accuracy Object--vehicle--bus	Accuracy Object--vehicle--car	Accuracy Object--vehicle--truck	Iou Construction--barrier--fence	Iou Construction--barrier--guard-rail	Iou Construction--barrier--other-barrier	Iou Construction--barrier--wall	Iou Construction--flat--road	Iou Construction--flat--service-lane	Iou Construction--flat--sidewalk	Iou Construction--structure--building	Iou Human--person	Iou Human--rider--bicyclist	Iou Marking--crosswalk-zebra	Iou Marking--general	Iou Nature--sky	Iou Nature--terrain	Iou Nature--vegetation	Iou Object--support--pole	Iou Object--support--traffic-sign-frame	Iou Object--traffic-light	Iou Object--traffic-sign--front	Iou Object--vehicle--bicycle	Iou Object--vehicle--bus	Iou Object--vehicle--car	Iou Object--vehicle--truck	Validation Loss	Mean Accuracy	Mean Iou	Overall Accuracy
1.2232	0.4444	1000	0.5702	0.0	0.0076	0.1947	0.9719	0.0	0.7853	0.9291	0.0010	0.0	0.0	0.0565	0.9883	0.7214	0.9384	0.0000	0.0	0.0	0.0080	0.0	0.0	0.9465	0.0000	0.3826	0.0	0.0076	0.1690	0.8322	0.0	0.6317	0.8093	0.0010	0.0	0.0	0.0513	0.9556	0.5364	0.8328	0.0000	0.0	0.0	0.0080	0.0	0.0	0.7836	0.0000	1.2801	0.3095	0.2609	0.8827
1.1887	0.8889	2000	0.5990	0.6808	0.5331	0.5714	0.9134	0.0	0.9065	0.9353	0.7385	0.0	0.6071	0.5674	0.9882	0.7938	0.9389	0.4174	0.0	0.4767	0.7433	0.0	0.6268	0.9588	0.7212	0.4805	0.5040	0.4172	0.4380	0.8668	0.0	0.6597	0.8553	0.5604	0.0	0.4833	0.4250	0.9728	0.6336	0.8685	0.3218	0.0	0.3922	0.5945	0.0	0.5726	0.8500	0.5743	1.0995	0.5964	0.4987	0.9140
1.12	1.3333	3000	0.6494	0.6527	0.6627	0.6206	0.9596	0.0414	0.8344	0.9347	0.8106	0.0002	0.6065	0.6003	0.9890	0.7771	0.9443	0.4668	0.0002	0.6540	0.7695	0.0241	0.8612	0.9461	0.7723	0.5067	0.5496	0.4857	0.4887	0.8946	0.0409	0.7378	0.8653	0.5809	0.0002	0.4956	0.4698	0.9762	0.6538	0.8781	0.3705	0.0002	0.4815	0.6305	0.0241	0.7106	0.8770	0.6324	1.0300	0.6338	0.5370	0.9257
1.0221	1.7778	4000	0.6228	0.8566	0.6871	0.6574	0.9564	0.5549	0.8285	0.9382	0.8244	0.0000	0.7269	0.5655	0.9900	0.8062	0.9339	0.5275	0.3594	0.6964	0.7914	0.6096	0.8429	0.9542	0.7699	0.5058	0.5160	0.5043	0.4915	0.8995	0.3755	0.7321	0.8703	0.6045	0.0000	0.5934	0.4876	0.9777	0.6576	0.8812	0.3961	0.3432	0.5040	0.6565	0.4414	0.6979	0.8799	0.6538	1.0093	0.7174	0.5943	0.9286
1.0284	2.2222	5000	0.6842	0.7088	0.6491	0.6110	0.9499	0.3896	0.9066	0.9419	0.8207	0.3589	0.7697	0.6303	0.9876	0.8394	0.9360	0.5362	0.4824	0.7015	0.8113	0.7283	0.8290	0.9455	0.7542	0.5253	0.5870	0.5136	0.4817	0.9053	0.3123	0.7648	0.8702	0.6265	0.3358	0.6093	0.5193	0.9782	0.6693	0.8823	0.4089	0.4306	0.5184	0.6559	0.5020	0.7316	0.8863	0.6639	0.9947	0.7379	0.6252	0.9313
1.0125	2.6667	6000	0.6698	0.8612	0.6979	0.5951	0.9543	0.7659	0.8466	0.9480	0.8134	0.6030	0.7552	0.6331	0.9907	0.7757	0.9392	0.5164	0.5922	0.7079	0.7869	0.7361	0.8216	0.9564	0.7390	0.5314	0.5682	0.5170	0.5032	0.9060	0.4156	0.7689	0.8739	0.6418	0.4794	0.6298	0.5257	0.9786	0.6632	0.8837	0.4118	0.4875	0.5293	0.6700	0.5361	0.7336	0.8850	0.6647	0.9839	0.7698	0.6437	0.9330
1.0475	3.1111	7000	0.6851	0.8080	0.6232	0.6705	0.9491	0.5703	0.8856	0.9434	0.8270	0.6655	0.7459	0.6440	0.9900	0.8396	0.9418	0.5247	0.5958	0.7055	0.7797	0.7327	0.8714	0.9528	0.7800	0.5436	0.6190	0.5340	0.5274	0.9047	0.4117	0.7595	0.8768	0.6479	0.4965	0.6424	0.5233	0.9793	0.6715	0.8858	0.4194	0.4950	0.5389	0.6802	0.5389	0.7657	0.8908	0.6714	0.9755	0.7709	0.6532	0.9341
1.2139	3.5556	8000	0.6550	0.8046	0.6946	0.6737	0.9579	0.6496	0.8669	0.9432	0.8090	0.6377	0.7658	0.6484	0.9902	0.7722	0.9485	0.5491	0.6013	0.6810	0.7868	0.7504	0.8784	0.9529	0.7980	0.5439	0.6155	0.5385	0.5375	0.9114	0.4499	0.7722	0.8766	0.6564	0.5043	0.6365	0.5389	0.9797	0.6720	0.8879	0.4249	0.5045	0.5471	0.6883	0.5450	0.7749	0.8922	0.6816	0.9700	0.7746	0.6600	0.9361
0.9667	4.0	9000	0.6269	0.7822	0.6690	0.7303	0.9449	0.6441	0.9071	0.9422	0.8301	0.6951	0.7563	0.6526	0.9888	0.8000	0.9521	0.5469	0.5728	0.6941	0.7925	0.7470	0.8842	0.9616	0.8004	0.5315	0.6148	0.5304	0.5512	0.9059	0.4298	0.7538	0.8811	0.6575	0.5099	0.6570	0.5477	0.9798	0.6840	0.8885	0.4329	0.5032	0.5549	0.6858	0.5426	0.7918	0.8910	0.6854	0.9717	0.7792	0.6613	0.9356
0.9185	4.4444	10000	0.6886	0.7736	0.6861	0.6848	0.9642	0.6010	0.8124	0.9358	0.8273	0.6631	0.7653	0.6549	0.9889	0.8662	0.9455	0.5549	0.6414	0.7289	0.8127	0.7651	0.8360	0.9525	0.7956	0.5515	0.6024	0.5216	0.5326	0.9076	0.4402	0.7398	0.8806	0.6629	0.5227	0.6618	0.5553	0.9799	0.6831	0.8869	0.4373	0.5192	0.5496	0.6861	0.5643	0.7686	0.8936	0.6718	0.9696	0.7802	0.6617	0.9355
0.9251	4.8889	11000	0.7057	0.7517	0.7043	0.6223	0.9539	0.6552	0.8834	0.9429	0.8267	0.7147	0.8372	0.6626	0.9894	0.7855	0.9511	0.5368	0.6170	0.7379	0.7972	0.7274	0.8983	0.9570	0.8072	0.5431	0.6008	0.5544	0.5358	0.9137	0.4574	0.7904	0.8787	0.6589	0.5273	0.6437	0.5545	0.9802	0.6792	0.8900	0.4332	0.5236	0.5506	0.6992	0.5731	0.7735	0.8926	0.6796	0.9623	0.7855	0.6667	0.9375
0.9888	5.3333	12000	0.6904	0.7658	0.6800	0.7087	0.9531	0.4207	0.9055	0.9461	0.8465	0.6549	0.7846	0.6661	0.9901	0.8429	0.9417	0.5456	0.6098	0.7133	0.7911	0.7692	0.8669	0.9610	0.7646	0.5524	0.6182	0.5451	0.5508	0.9124	0.3442	0.7815	0.8830	0.6455	0.5194	0.6485	0.5570	0.9801	0.6864	0.8893	0.4357	0.5281	0.5588	0.6955	0.5617	0.7826	0.8926	0.6882	0.9605	0.7747	0.6633	0.9375
0.9546	5.7778	13000	0.6704	0.7887	0.7170	0.7026	0.9628	0.4858	0.8695	0.9423	0.8124	0.7615	0.7621	0.6666	0.9916	0.8676	0.9344	0.5581	0.6431	0.7165	0.8075	0.7896	0.9154	0.9551	0.8263	0.5433	0.6183	0.5450	0.5311	0.9132	0.3969	0.7741	0.8832	0.6712	0.5271	0.6757	0.5596	0.9801	0.6811	0.8886	0.4390	0.5401	0.5654	0.7025	0.5450	0.7482	0.8970	0.6980	0.9588	0.7890	0.6662	0.9375
0.9385	6.2222	14000	0.6859	0.7862	0.6720	0.6695	0.9601	0.6037	0.8797	0.9448	0.8479	0.7002	0.8024	0.6754	0.9889	0.8351	0.9472	0.5586	0.6714	0.7535	0.8095	0.7670	0.9010	0.9607	0.8368	0.5527	0.6162	0.5557	0.5480	0.9159	0.4532	0.7871	0.8851	0.6626	0.5317	0.6850	0.5667	0.9804	0.6934	0.8916	0.4434	0.5473	0.5574	0.7054	0.5678	0.7668	0.8971	0.6960	0.9513	0.7938	0.6742	0.9393
0.9649	6.6667	15000	0.6868	0.7777	0.7190	0.7330	0.9664	0.6360	0.8593	0.9432	0.8330	0.7332	0.8220	0.6249	0.9897	0.8720	0.9404	0.5629	0.6166	0.7347	0.8088	0.7473	0.8147	0.9568	0.8332	0.5564	0.6200	0.5713	0.5632	0.9161	0.4644	0.7815	0.8859	0.6710	0.5559	0.6696	0.5501	0.9806	0.6810	0.8908	0.4462	0.5236	0.5690	0.7051	0.5538	0.7603	0.8991	0.7125	0.9559	0.7918	0.6751	0.9392
0.9837	7.1111	16000	0.6970	0.8135	0.6830	0.6925	0.9508	0.5759	0.9156	0.9450	0.8385	0.7269	0.7993	0.6869	0.9875	0.8002	0.9509	0.5846	0.6613	0.7484	0.8044	0.7864	0.8693	0.9609	0.8070	0.5560	0.6131	0.5649	0.5464	0.9146	0.4438	0.7789	0.8850	0.6657	0.5365	0.6918	0.5729	0.9798	0.6902	0.8924	0.4488	0.5208	0.5689	0.7026	0.5609	0.7937	0.8993	0.7045	0.9550	0.7950	0.6753	0.9389
0.9885	7.5556	17000	0.7252	0.8036	0.6680	0.6714	0.9597	0.5902	0.8797	0.9472	0.8421	0.6778	0.8127	0.6709	0.9905	0.8355	0.9443	0.5382	0.7031	0.7404	0.8000	0.7346	0.8996	0.9620	0.8160	0.5601	0.6416	0.5558	0.5551	0.9155	0.4445	0.7852	0.8867	0.6692	0.5455	0.6706	0.5626	0.9808	0.6954	0.8930	0.4421	0.5448	0.5695	0.7092	0.5917	0.8030	0.9001	0.7195	0.9514	0.7919	0.6801	0.9399
0.9472	8.0	18000	0.7201	0.7481	0.6961	0.6713	0.9581	0.7132	0.8887	0.9477	0.8458	0.7235	0.7983	0.6833	0.9912	0.8176	0.9451	0.5440	0.6924	0.7267	0.8083	0.7538	0.8793	0.9614	0.8478	0.5686	0.6204	0.5513	0.5547	0.9178	0.4753	0.7893	0.8851	0.6653	0.5531	0.6900	0.5722	0.9807	0.6971	0.8937	0.4429	0.5504	0.5774	0.7131	0.5683	0.7907	0.9012	0.7287	0.9464	0.7983	0.6821	0.9405
1.0383	8.4444	19000	0.6841	0.8030	0.6607	0.7211	0.9608	0.6133	0.8750	0.9476	0.8187	0.7862	0.7857	0.7044	0.9904	0.7855	0.9469	0.5681	0.6974	0.7499	0.8211	0.7416	0.9011	0.9617	0.7960	0.5668	0.6247	0.5538	0.5585	0.9165	0.4327	0.7842	0.8875	0.6750	0.5290	0.6976	0.5747	0.9808	0.6858	0.8934	0.4497	0.5503	0.5707	0.7053	0.5493	0.7834	0.9005	0.7010	0.9520	0.7965	0.6770	0.9402
0.9236	8.8889	20000	0.7167	0.7557	0.6830	0.6632	0.9529	0.6264	0.9067	0.9491	0.8344	0.7219	0.8216	0.6903	0.9915	0.8470	0.9414	0.5647	0.6572	0.7443	0.8192	0.7185	0.9042	0.9572	0.8550	0.5760	0.6254	0.5504	0.5277	0.9154	0.4656	0.7797	0.8856	0.6756	0.5507	0.6972	0.5781	0.9809	0.6988	0.8934	0.4510	0.5599	0.5776	0.7189	0.5749	0.7896	0.9062	0.7376	0.9508	0.7966	0.6833	0.9401
0.9449	9.3333	21000	0.6725	0.7971	0.6865	0.7227	0.9620	0.5470	0.8828	0.9469	0.8447	0.7145	0.7529	0.6901	0.9904	0.8329	0.9436	0.5848	0.6544	0.7513	0.8164	0.7588	0.8998	0.9587	0.8154	0.5635	0.6228	0.5656	0.5530	0.9177	0.4237	0.7880	0.8874	0.6751	0.5370	0.6693	0.5631	0.9809	0.6869	0.8930	0.4575	0.5440	0.5806	0.7129	0.5681	0.7772	0.9014	0.7005	0.9518	0.7925	0.6769	0.9402
1.0282	9.7778	22000	0.7153	0.7991	0.6694	0.7252	0.9588	0.6269	0.9026	0.9493	0.8327	0.6922	0.8130	0.6820	0.9896	0.8370	0.9424	0.5814	0.6956	0.7443	0.8242	0.7874	0.8868	0.9597	0.8655	0.5787	0.6230	0.5442	0.5660	0.9197	0.4716	0.7932	0.8884	0.6794	0.5422	0.7089	0.5799	0.9809	0.6948	0.8937	0.4572	0.5618	0.5797	0.7185	0.5705	0.7931	0.9042	0.7298	0.9437	0.8035	0.6861	0.9415
0.9765	10.2222	23000	0.6973	0.7662	0.7285	0.7140	0.9569	0.7153	0.8956	0.9462	0.8375	0.7362	0.8066	0.7039	0.9911	0.8316	0.9443	0.5579	0.6472	0.7364	0.8269	0.7831	0.8961	0.9585	0.8547	0.5693	0.6107	0.5397	0.5661	0.9183	0.4959	0.7929	0.8888	0.6711	0.5112	0.7028	0.5814	0.9810	0.6986	0.8938	0.4502	0.5546	0.5844	0.7148	0.5392	0.7921	0.9046	0.7281	0.9486	0.8057	0.6822	0.9412
0.9217	10.6667	24000	0.7099	0.8119	0.6995	0.7175	0.9558	0.6655	0.8974	0.9473	0.8464	0.7354	0.8212	0.6935	0.9899	0.8197	0.9482	0.5786	0.6549	0.7626	0.8175	0.7396	0.8998	0.9538	0.8605	0.5764	0.6236	0.5480	0.5643	0.9178	0.4929	0.7869	0.8894	0.6780	0.5618	0.7014	0.5822	0.9811	0.6938	0.8947	0.4580	0.5510	0.5796	0.7129	0.5901	0.8015	0.9062	0.7248	0.9453	0.8055	0.6877	0.9413
0.9323	11.1111	25000	0.6859	0.7958	0.6713	0.6833	0.9660	0.5001	0.8722	0.9513	0.8517	0.7259	0.8034	0.6833	0.9899	0.8629	0.9440	0.5823	0.6580	0.7518	0.8151	0.7806	0.8573	0.9601	0.8014	0.5699	0.6324	0.5636	0.5513	0.9192	0.4098	0.7874	0.8881	0.6784	0.5608	0.7110	0.5797	0.9811	0.6961	0.8942	0.4607	0.5398	0.5843	0.7207	0.5785	0.7805	0.9006	0.7011	0.9506	0.7910	0.6821	0.9414
0.929	11.5556	26000	0.6956	0.8104	0.6802	0.7273	0.9547	0.6745	0.9086	0.9495	0.8465	0.7555	0.7899	0.6825	0.9898	0.8266	0.9445	0.5925	0.6659	0.7600	0.8262	0.7654	0.8893	0.9595	0.8236	0.5714	0.6223	0.5615	0.5525	0.9167	0.4970	0.7839	0.8894	0.6790	0.5851	0.6976	0.5759	0.9812	0.6937	0.8947	0.4602	0.5346	0.5815	0.7146	0.5840	0.7920	0.9037	0.7258	0.9525	0.8052	0.6869	0.9409
0.8921	12.0	27000	0.6719	0.8224	0.6834	0.7274	0.9639	0.5837	0.8808	0.9451	0.8577	0.6925	0.8134	0.6868	0.9903	0.8302	0.9503	0.5825	0.6433	0.7607	0.8280	0.7629	0.8677	0.9618	0.8272	0.5720	0.6331	0.5666	0.5600	0.9210	0.4573	0.7944	0.8900	0.6774	0.5534	0.7080	0.5820	0.9812	0.6917	0.8943	0.4581	0.5484	0.5837	0.7179	0.5890	0.7936	0.9036	0.7309	0.9438	0.7971	0.6873	0.9421
0.9551	12.4444	28000	0.7332	0.7665	0.7067	0.7068	0.9581	0.6366	0.8887	0.9475	0.8418	0.7535	0.8265	0.7210	0.9890	0.8284	0.9471	0.5917	0.6982	0.7549	0.8249	0.7723	0.8759	0.9596	0.8520	0.5727	0.6215	0.5521	0.5651	0.9198	0.4867	0.8036	0.8905	0.6794	0.5470	0.6854	0.5916	0.9811	0.7008	0.8955	0.4620	0.5434	0.5861	0.7145	0.5758	0.8071	0.9071	0.7292	0.9458	0.8079	0.6877	0.9420
0.8853	12.8889	29000	0.7067	0.7827	0.6972	0.7191	0.9596	0.6150	0.8954	0.9461	0.8428	0.7492	0.8193	0.6857	0.9897	0.8314	0.9486	0.5897	0.7024	0.7694	0.8239	0.7638	0.8787	0.9582	0.8555	0.5838	0.6320	0.5621	0.5740	0.9178	0.4580	0.7907	0.8909	0.6854	0.5749	0.7054	0.5832	0.9811	0.6979	0.8950	0.4631	0.5541	0.5799	0.7168	0.5798	0.7993	0.9070	0.7305	0.9464	0.8057	0.6897	0.9420
1.0445	13.3333	30000	0.6939	0.8091	0.6937	0.6978	0.9626	0.6119	0.8830	0.9513	0.8379	0.6934	0.8343	0.7040	0.9896	0.8217	0.9455	0.5910	0.7049	0.7586	0.8229	0.7597	0.8936	0.9607	0.8294	0.5777	0.6280	0.5516	0.5586	0.9208	0.4762	0.7940	0.8893	0.6850	0.5532	0.7093	0.5892	0.9812	0.6906	0.8951	0.4627	0.5545	0.5872	0.7234	0.5837	0.7906	0.9076	0.7394	0.9473	0.8022	0.6891	0.9422
0.946	13.7778	31000	0.6916	0.8010	0.6912	0.7125	0.9641	0.5945	0.8880	0.9499	0.8527	0.7083	0.8163	0.6996	0.9914	0.8220	0.9449	0.5781	0.6487	0.7520	0.8224	0.7735	0.8895	0.9617	0.8119	0.5815	0.6206	0.5618	0.5571	0.9218	0.4617	0.7986	0.8899	0.6784	0.5543	0.7120	0.5907	0.9813	0.6972	0.8955	0.4601	0.5516	0.5895	0.7197	0.5796	0.7783	0.9062	0.7193	0.9451	0.7985	0.6872	0.9426
0.9696	14.2222	32000	0.6905	0.7843	0.7100	0.6553	0.9608	0.6718	0.9059	0.9505	0.8451	0.7576	0.8399	0.6803	0.9898	0.7672	0.9526	0.5770	0.6876	0.7555	0.8224	0.7752	0.8676	0.9621	0.8332	0.5796	0.6307	0.5556	0.5338	0.9224	0.5101	0.8054	0.8884	0.6852	0.5862	0.7097	0.5871	0.9812	0.6704	0.8936	0.4614	0.5547	0.5850	0.7187	0.5806	0.8023	0.9059	0.7305	0.9502	0.8018	0.6904	0.9423
0.8856	14.6667	33000	0.7203	0.7959	0.6819	0.6989	0.9613	0.6177	0.8972	0.9426	0.8508	0.7254	0.8229	0.7096	0.9906	0.8196	0.9501	0.5966	0.6393	0.7720	0.8279	0.7610	0.8689	0.9598	0.8403	0.5813	0.6279	0.5540	0.5573	0.9220	0.4750	0.8023	0.8917	0.6840	0.5445	0.7191	0.5960	0.9813	0.6890	0.8947	0.4660	0.5240	0.5830	0.7174	0.5725	0.7942	0.9061	0.7282	0.9463	0.8022	0.6874	0.9426
0.9123	15.1111	34000	0.7244	0.7802	0.6843	0.6924	0.9604	0.5314	0.9040	0.9458	0.8365	0.7827	0.8232	0.7057	0.9898	0.8120	0.9508	0.5830	0.6900	0.7404	0.8258	0.7759	0.8790	0.9617	0.8280	0.5817	0.6259	0.5544	0.5495	0.9210	0.4291	0.8022	0.8912	0.6924	0.5925	0.7097	0.5922	0.9812	0.6887	0.8946	0.4650	0.5633	0.5945	0.7252	0.5687	0.7975	0.9058	0.7296	0.9481	0.8003	0.6894	0.9425
0.9586	15.5556	35000	0.7137	0.7836	0.6917	0.7091	0.9619	0.6224	0.8781	0.9486	0.8357	0.7273	0.8370	0.6981	0.9898	0.8432	0.9466	0.5884	0.6847	0.7565	0.8245	0.7786	0.8842	0.9611	0.8524	0.5882	0.6294	0.5518	0.5505	0.9204	0.4511	0.7939	0.8908	0.6890	0.5533	0.7182	0.5907	0.9814	0.6929	0.8953	0.4680	0.5768	0.5912	0.7226	0.5717	0.7944	0.9075	0.7366	0.9494	0.8051	0.6898	0.9424
0.9337	16.0	36000	0.7155	0.7862	0.6914	0.7178	0.9557	0.5895	0.9073	0.9460	0.8511	0.7229	0.8386	0.7159	0.9905	0.8339	0.9462	0.5888	0.6935	0.7591	0.8223	0.7813	0.8813	0.9630	0.8236	0.5925	0.6283	0.5521	0.5583	0.9194	0.4651	0.7933	0.8913	0.6831	0.5614	0.7088	0.5931	0.9813	0.6906	0.8948	0.4658	0.5669	0.5917	0.7203	0.5728	0.7947	0.9070	0.7252	0.9495	0.8053	0.6895	0.9422
0.9739	16.4444	37000	0.7139	0.7663	0.6945	0.6975	0.9611	0.5809	0.8826	0.9474	0.8453	0.7471	0.8455	0.7156	0.9902	0.8433	0.9473	0.5967	0.6439	0.7555	0.8226	0.7806	0.8873	0.9631	0.8268	0.5878	0.6361	0.5534	0.5557	0.9209	0.4557	0.7981	0.8918	0.6886	0.5691	0.7105	0.5932	0.9813	0.6899	0.8949	0.4672	0.5338	0.5970	0.7264	0.5797	0.8063	0.9066	0.7326	0.9489	0.8024	0.6903	0.9426
0.9184	16.8889	38000	0.6972	0.7814	0.6872	0.7017	0.9622	0.6181	0.8887	0.9467	0.8445	0.7520	0.8351	0.6967	0.9905	0.8006	0.9520	0.5775	0.6836	0.7536	0.8198	0.7777	0.8606	0.9603	0.8438	0.5810	0.6347	0.5504	0.5524	0.9210	0.4678	0.7927	0.8911	0.6882	0.5697	0.7208	0.5930	0.9814	0.6832	0.8948	0.4638	0.5662	0.5950	0.7199	0.5804	0.7956	0.9074	0.7288	0.9493	0.8014	0.6904	0.9425
0.8846	17.3333	39000	0.7089	0.7910	0.7070	0.7082	0.9598	0.6232	0.8892	0.9475	0.8555	0.7225	0.8497	0.7055	0.9903	0.8387	0.9466	0.5876	0.6796	0.7605	0.8270	0.7748	0.8921	0.9637	0.8201	0.5856	0.6362	0.5568	0.5551	0.9208	0.4680	0.7949	0.8915	0.6833	0.5643	0.7150	0.5958	0.9814	0.6920	0.8955	0.4683	0.5595	0.5946	0.7252	0.5839	0.8086	0.9059	0.7214	0.9489	0.8065	0.6915	0.9426
0.9373	17.7778	40000	0.9482	0.6922	0.8057	0.9429	0.6971	0.7852	0.6909	0.7067	0.9637	0.5863	0.8871	0.9477	0.8389	0.7507	0.8232	0.7083	0.9904	0.8428	0.9464	0.5965	0.6947	0.7636	0.8238	0.7786	0.9076	0.9597	0.8419	0.5824	0.6364	0.5599	0.5511	0.9219	0.4566	0.7969	0.8922	0.6904	0.5642	0.7268	0.5964	0.9814	0.6909	0.8948	0.4692	0.5720	0.5920	0.7254	0.5770	0.7995	0.9087	0.7337
0.894	18.2222	41000	0.9471	0.6933	0.8056	0.9432	0.7137	0.7797	0.7031	0.7001	0.9611	0.5874	0.8973	0.9473	0.8530	0.7323	0.8257	0.7197	0.9902	0.8290	0.9482	0.5972	0.6871	0.7597	0.8267	0.7673	0.8899	0.9604	0.8536	0.5840	0.6338	0.5571	0.5579	0.9222	0.4590	0.8020	0.8922	0.6843	0.5682	0.7181	0.5954	0.9814	0.6969	0.8960	0.4692	0.5645	0.5942	0.7280	0.5832	0.8114	0.9093	0.7372
0.8541	18.6667	42000	0.9476	0.6929	0.8047	0.9431	0.7240	0.7790	0.6978	0.6984	0.9619	0.5759	0.8922	0.9475	0.8447	0.7502	0.8324	0.7193	0.9900	0.8232	0.9486	0.5936	0.6959	0.7596	0.8295	0.7749	0.8816	0.9622	0.8249	0.5861	0.6352	0.5591	0.5595	0.9223	0.4537	0.7996	0.8925	0.6884	0.5759	0.7247	0.5987	0.9814	0.6911	0.8957	0.4690	0.5710	0.5938	0.7259	0.5772	0.8035	0.9074	0.7252
0.9359	19.1111	43000	0.9491	0.6924	0.8057	0.9430	0.7183	0.7779	0.7006	0.7076	0.9626	0.5701	0.8904	0.9480	0.8437	0.7625	0.8214	0.7181	0.9897	0.8361	0.9466	0.5981	0.6876	0.7613	0.8244	0.7775	0.8907	0.9623	0.8368	0.5867	0.6338	0.5611	0.5567	0.9225	0.4491	0.8018	0.8922	0.6906	0.5846	0.7215	0.5958	0.9814	0.6909	0.8955	0.4693	0.5668	0.5932	0.7219	0.5748	0.7984	0.9076	0.7299
0.9389	19.5556	44000	0.9478	0.6932	0.8066	0.9430	0.7033	0.7910	0.7026	0.7066	0.9625	0.6185	0.8918	0.9485	0.8492	0.7514	0.8247	0.7110	0.9905	0.8390	0.9457	0.5982	0.6959	0.7550	0.8254	0.7690	0.8872	0.9619	0.8225	0.5839	0.6314	0.5615	0.5556	0.9225	0.4665	0.8005	0.8925	0.6882	0.5862	0.7236	0.5963	0.9815	0.6892	0.8952	0.4693	0.5713	0.5943	0.7262	0.5817	0.7969	0.9075	0.7222
0.8656	20.0	45000	0.9471	0.6940	0.8072	0.9431	0.7139	0.7940	0.7037	0.7066	0.9635	0.6054	0.8864	0.9471	0.8464	0.7717	0.8305	0.7034	0.9905	0.8364	0.9463	0.5949	0.6872	0.7613	0.8302	0.7734	0.8816	0.9634	0.8290	0.5854	0.6329	0.5635	0.5578	0.9223	0.4671	0.7992	0.8928	0.6907	0.5950	0.7205	0.5949	0.9814	0.6899	0.8954	0.4690	0.5685	0.5949	0.7276	0.5790	0.8005	0.9072	0.7273

Framework versions

• Transformers 4.48.3
• Pytorch 2.1.2+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0