histv4_ftis_pretrain_smlm

This model is a fine-tuned version of Arthur-Tsai/histv4_pretrain_smlm on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8107
• Accuracy: 0.9403
• Macro F1: 0.8477

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 4
• seed: 42
• optimizer: Use 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: 6731
• training_steps: 134625

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
41.5551	0.0010	134	23.5263	0.0544	0.0272
16.2558	1.0010	268	8.7211	0.1802	0.0537
6.9342	2.0010	402	6.4634	0.5184	0.1220
5.9178	3.0010	536	5.1432	0.5568	0.1404
5.201	4.0010	670	4.8583	0.5807	0.1479
4.7006	5.0010	804	4.0314	0.5984	0.1548
4.2736	6.0010	938	3.4220	0.6089	0.1647
3.8421	7.0009	1072	3.0184	0.6234	0.1776
3.2177	8.0009	1206	2.7418	0.6212	0.1735
3.0887	9.0009	1340	2.4012	0.6326	0.1792
2.7333	10.0009	1474	2.3004	0.6488	0.1914
2.5787	11.0009	1608	2.2017	0.6553	0.2067
2.4664	12.0009	1742	2.0509	0.6689	0.2198
2.3341	13.0009	1876	1.9938	0.6728	0.2479
2.131	14.0009	2010	2.0652	0.6776	0.2446
2.0956	15.0009	2144	1.8677	0.6889	0.2751
2.2169	16.0009	2278	1.8121	0.6926	0.2824
1.9588	17.0009	2412	1.8118	0.7018	0.3110
1.9699	18.0009	2546	1.7990	0.6966	0.3193
1.8866	19.0009	2680	1.6326	0.7145	0.3412
1.6753	20.0008	2814	1.6780	0.7134	0.3435
1.7359	21.0008	2948	1.6077	0.7326	0.3548
1.5679	22.0008	3082	1.3690	0.7725	0.4050
1.4809	23.0008	3216	1.4555	0.7674	0.4124
1.3325	24.0008	3350	1.3400	0.7898	0.4263
1.3643	25.0008	3484	1.3297	0.7869	0.4563
1.274	26.0008	3618	1.2339	0.8029	0.4740
1.2455	27.0008	3752	1.1572	0.8076	0.4825
1.1939	28.0008	3886	1.2200	0.8075	0.4868
1.0667	29.0008	4020	1.1408	0.8158	0.5129
1.0309	30.0008	4154	1.0759	0.8242	0.5365
0.9944	31.0008	4288	1.1034	0.8225	0.5291
0.9707	32.0008	4422	1.0481	0.8276	0.5551
0.9262	33.0008	4556	1.0645	0.8302	0.5516
0.9374	34.0007	4690	0.9864	0.8401	0.5692
0.8312	35.0007	4824	1.0473	0.8426	0.5848
0.8721	36.0007	4958	0.9661	0.8523	0.5934
0.7767	37.0007	5092	0.9365	0.8457	0.5993
0.7556	38.0007	5226	0.9342	0.8506	0.6013
0.7268	39.0007	5360	0.9538	0.8563	0.5919
0.6446	40.0007	5494	0.8707	0.8673	0.6246
0.6231	41.0007	5628	0.8759	0.8631	0.6210
0.5897	42.0007	5762	0.9076	0.8694	0.6320
0.5714	43.0007	5896	0.9501	0.8567	0.6286
0.598	44.0007	6030	1.0102	0.8491	0.6329
0.5199	45.0007	6164	0.8747	0.8757	0.6491
0.5363	46.0007	6298	0.8333	0.8725	0.6515
0.5191	47.0006	6432	0.9014	0.8717	0.6571
0.4919	48.0006	6566	0.9997	0.8581	0.6336
0.4623	49.0006	6700	0.8609	0.8793	0.6605
0.4495	50.0006	6834	0.8148	0.8808	0.6749
0.4388	51.0006	6968	0.9222	0.8722	0.6622
0.4272	52.0006	7102	0.9723	0.8688	0.6609
0.3909	53.0006	7236	0.7780	0.8886	0.6889
0.3846	54.0006	7370	0.8528	0.8855	0.6834
0.3653	55.0006	7504	0.7721	0.8926	0.6973
0.3885	56.0006	7638	0.7939	0.8934	0.6943
0.3468	57.0006	7772	0.7859	0.8946	0.7011
0.3341	58.0006	7906	0.7899	0.8885	0.6989
0.3416	59.0006	8040	0.7807	0.8947	0.7031
0.279	60.0005	8174	0.7811	0.8943	0.7118
0.301	61.0005	8308	0.8019	0.8912	0.6952
0.3008	62.0005	8442	0.7605	0.8967	0.7143
0.292	63.0005	8576	0.7776	0.8977	0.7150
0.285	64.0005	8710	0.8281	0.8943	0.7087
0.2631	65.0005	8844	0.8047	0.9035	0.7263
0.2593	66.0005	8978	0.7932	0.8993	0.7246
0.242	67.0005	9112	0.7740	0.9039	0.7298
0.2665	68.0005	9246	0.7261	0.9027	0.7325
0.2534	69.0005	9380	0.8071	0.9040	0.7329
0.2205	70.0005	9514	0.7483	0.9097	0.7411
0.2538	71.0005	9648	0.7166	0.9111	0.7388
0.223	72.0005	9782	0.7492	0.9100	0.7455
0.2131	73.0005	9916	0.7118	0.9126	0.7507
0.1985	74.0004	10050	0.6889	0.9079	0.7453
0.1967	75.0004	10184	0.7733	0.9042	0.7486
0.1911	76.0004	10318	0.7425	0.9132	0.7515
0.1917	77.0004	10452	0.7944	0.9094	0.7509
0.1907	78.0004	10586	0.7017	0.9141	0.7560
0.1806	79.0004	10720	0.7798	0.9142	0.7561
0.1897	80.0004	10854	0.7465	0.9163	0.7587
0.1771	81.0004	10988	0.7284	0.9149	0.7596
0.1654	82.0004	11122	0.8105	0.9079	0.7535
0.1562	83.0004	11256	0.7314	0.9132	0.7605
0.1809	84.0004	11390	0.6930	0.9175	0.7625
0.1709	85.0004	11524	0.7884	0.9150	0.7654
0.158	86.0004	11658	0.7529	0.9140	0.7635
0.1593	87.0003	11792	0.7558	0.9208	0.7668
0.1542	88.0003	11926	0.7057	0.9208	0.7722
0.144	89.0003	12060	0.7488	0.9183	0.7713
0.1559	90.0003	12194	0.7894	0.9160	0.7744
0.1395	91.0003	12328	0.8427	0.9142	0.7752
0.1354	92.0003	12462	0.7397	0.9158	0.7780
0.1485	93.0003	12596	0.7778	0.9196	0.7739
0.145	94.0003	12730	0.7901	0.9200	0.7789
0.135	95.0003	12864	0.7491	0.9237	0.7833
0.1434	96.0003	12998	0.7421	0.9176	0.7794
0.1408	97.0003	13132	0.8034	0.9204	0.7826
0.1188	98.0003	13266	0.8200	0.9187	0.7861
0.1323	99.0003	13400	0.7874	0.9185	0.7848
0.116	100.0003	13534	0.7543	0.9221	0.7865
0.1263	101.0002	13668	0.7470	0.9250	0.7888
0.1074	102.0002	13802	0.7529	0.9219	0.7870
0.1155	103.0002	13936	0.7924	0.9241	0.7924
0.124	104.0002	14070	0.7581	0.9235	0.7877
0.1146	105.0002	14204	0.7882	0.9232	0.7955
0.1155	106.0002	14338	0.8141	0.9262	0.7934
0.1187	107.0002	14472	0.6847	0.9253	0.7929
0.1113	108.0002	14606	0.7904	0.9231	0.7911
0.1079	109.0002	14740	0.7682	0.9234	0.7915
0.1114	110.0002	14874	0.7867	0.9220	0.7914
0.1134	111.0002	15008	0.7289	0.9269	0.7949
0.1099	112.0002	15142	0.7459	0.9249	0.7970
0.108	113.0002	15276	0.7593	0.9250	0.8014
0.0994	114.0001	15410	0.7299	0.9224	0.7995
0.1045	115.0001	15544	0.7006	0.9271	0.8025
0.1009	116.0001	15678	0.7219	0.9271	0.7998
0.1005	117.0001	15812	0.7966	0.9275	0.8023
0.0968	118.0001	15946	0.8041	0.9241	0.8047
0.1034	119.0001	16080	0.7535	0.9277	0.8023
0.0901	120.0001	16214	0.7161	0.9284	0.8053
0.0942	121.0001	16348	0.8485	0.9263	0.8041
0.1017	122.0001	16482	0.8011	0.9247	0.8016
0.1016	123.0001	16616	0.8531	0.9271	0.8060
0.0965	124.0001	16750	0.7779	0.9262	0.8075
0.1036	125.0001	16884	0.7804	0.9268	0.8064
0.0862	126.0001	17018	0.7374	0.9292	0.8099
0.0988	127.0001	17152	0.7485	0.9258	0.8068
0.1001	128.0000	17286	0.7971	0.9291	0.8123
0.0826	129.0000	17420	0.7788	0.9300	0.8102
0.0895	130.0000	17554	0.8128	0.9273	0.8093
0.086	131.0000	17688	0.6950	0.9308	0.8135
0.0834	132.0000	17822	0.7838	0.9287	0.8114
0.0847	133.0000	17956	0.8893	0.9295	0.8104
0.0837	133.0010	18090	0.7676	0.9319	0.8132
0.0811	134.0010	18224	0.7416	0.9279	0.8130
0.0783	135.0010	18358	0.7640	0.9318	0.8146
0.0881	136.0010	18492	0.8202	0.9310	0.8129
0.0807	137.0010	18626	0.8029	0.9268	0.8103
0.0901	138.0010	18760	0.8089	0.9294	0.8118
0.0881	139.0010	18894	0.8343	0.9311	0.8106
0.0794	140.0010	19028	0.8175	0.9263	0.8096
0.0831	141.0009	19162	0.7771	0.9320	0.8150
0.0802	142.0009	19296	0.8411	0.9266	0.8109
0.0756	143.0009	19430	0.7972	0.9261	0.8096
0.0775	144.0009	19564	0.8106	0.9324	0.8163
0.0791	145.0009	19698	0.7681	0.9317	0.8173
0.0828	146.0009	19832	0.8258	0.9275	0.8170
0.0776	147.0009	19966	0.8004	0.9274	0.8133
0.0835	148.0009	20100	0.7324	0.9322	0.8187
0.0746	149.0009	20234	0.7475	0.9320	0.8241
0.0692	150.0009	20368	0.8213	0.9299	0.8163
0.0754	151.0009	20502	0.9008	0.9300	0.8177
0.0787	152.0009	20636	0.7721	0.9339	0.8203
0.0701	153.0009	20770	0.7928	0.9312	0.8187
0.0851	154.0008	20904	0.7540	0.9297	0.8163
0.0733	155.0008	21038	0.7263	0.9352	0.8241
0.0775	156.0008	21172	0.7911	0.9317	0.8200
0.0697	157.0008	21306	0.8417	0.9342	0.8267
0.0845	158.0008	21440	0.8306	0.9319	0.8214
0.0706	159.0008	21574	0.8212	0.9316	0.8233
0.0692	160.0008	21708	0.8235	0.9298	0.8200
0.0778	161.0008	21842	0.8269	0.9329	0.8225
0.0869	162.0008	21976	0.8551	0.9324	0.8219
0.0662	163.0008	22110	0.8053	0.9324	0.8228
0.07	164.0008	22244	0.8416	0.9352	0.8264
0.069	165.0008	22378	0.8288	0.9342	0.8228
0.0705	166.0008	22512	0.7717	0.9339	0.8198
0.0672	167.0008	22646	0.7860	0.9355	0.8252
0.0714	168.0007	22780	0.7715	0.9353	0.8252
0.0669	169.0007	22914	0.9343	0.9330	0.8262
0.0636	170.0007	23048	0.8008	0.9304	0.8209
0.0691	171.0007	23182	0.7393	0.9316	0.8243
0.0658	172.0007	23316	0.7584	0.9328	0.8263
0.0683	173.0007	23450	0.7970	0.9320	0.8265
0.0718	174.0007	23584	0.9420	0.9332	0.8270
0.0714	175.0007	23718	0.8105	0.9336	0.8223
0.0673	176.0007	23852	0.8063	0.9315	0.8254
0.0685	177.0007	23986	0.8507	0.9324	0.8232
0.0672	178.0007	24120	0.8262	0.9354	0.8296
0.0666	179.0007	24254	0.8841	0.9358	0.8272
0.0623	180.0007	24388	0.8183	0.9354	0.8235
0.0616	181.0006	24522	0.7900	0.9350	0.8220
0.0625	182.0006	24656	0.7423	0.9360	0.8282
0.059	183.0006	24790	0.7458	0.9327	0.8281
0.0586	184.0006	24924	0.8024	0.9353	0.8282
0.0613	185.0006	25058	0.8241	0.9369	0.8297
0.0637	186.0006	25192	0.7518	0.9345	0.8327
0.0724	187.0006	25326	0.8642	0.9331	0.8256
0.0664	188.0006	25460	0.8348	0.9344	0.8317
0.0592	189.0006	25594	0.7909	0.9362	0.8312
0.0659	190.0006	25728	0.7717	0.9373	0.8328
0.0662	191.0006	25862	0.7877	0.9367	0.8308
0.0569	192.0006	25996	0.7889	0.9334	0.8325
0.0681	193.0006	26130	0.8186	0.9332	0.8280
0.0625	194.0005	26264	0.8294	0.9338	0.8287
0.0586	195.0005	26398	0.8620	0.9330	0.8279
0.0591	196.0005	26532	0.7812	0.9334	0.8297
0.0562	197.0005	26666	0.8911	0.9327	0.8291
0.0537	198.0005	26800	0.8063	0.9335	0.8324
0.0653	199.0005	26934	0.7315	0.9355	0.8354
0.0586	200.0005	27068	0.8556	0.9358	0.8360
0.0662	201.0005	27202	0.7125	0.9357	0.8332
0.0544	202.0005	27336	0.7896	0.9375	0.8347
0.058	203.0005	27470	0.7741	0.9346	0.8331
0.0569	204.0005	27604	0.8569	0.9358	0.8327
0.0541	205.0005	27738	0.8176	0.9360	0.8341
0.0639	206.0005	27872	0.8133	0.9347	0.8314
0.0544	207.0005	28006	0.8501	0.9379	0.8380
0.056	208.0004	28140	0.7397	0.9357	0.8334
0.0605	209.0004	28274	0.8294	0.9380	0.8344
0.0598	210.0004	28408	0.8295	0.9360	0.8314
0.0621	211.0004	28542	0.9268	0.9325	0.8285
0.0638	212.0004	28676	0.7732	0.9363	0.8370
0.0652	213.0004	28810	0.8035	0.9354	0.8364
0.0623	214.0004	28944	0.7354	0.9342	0.8313
0.0601	215.0004	29078	0.8142	0.9362	0.8346
0.0487	216.0004	29212	0.8914	0.9359	0.8362
0.0522	217.0004	29346	0.8071	0.9359	0.8349
0.0509	218.0004	29480	0.8832	0.9365	0.8364
0.059	219.0004	29614	0.9041	0.9349	0.8338
0.0548	220.0004	29748	0.7905	0.9387	0.8374
0.0615	221.0003	29882	0.9597	0.9338	0.8348
0.0539	222.0003	30016	0.8774	0.9338	0.8349
0.0558	223.0003	30150	0.7731	0.9358	0.8342
0.0627	224.0003	30284	0.7959	0.9410	0.8400
0.0515	225.0003	30418	0.9352	0.9356	0.8396
0.0519	226.0003	30552	0.8323	0.9339	0.8364
0.058	227.0003	30686	0.8642	0.9330	0.8365
0.0556	228.0003	30820	0.9974	0.9284	0.8343
0.0537	229.0003	30954	0.7829	0.9356	0.8361
0.0545	230.0003	31088	0.7687	0.9375	0.8351
0.0493	231.0003	31222	0.8174	0.9381	0.8359
0.0509	232.0003	31356	0.8749	0.9352	0.8365
0.0578	233.0003	31490	0.8092	0.9351	0.8389
0.054	234.0003	31624	0.9021	0.9320	0.8412
0.0546	235.0002	31758	0.8734	0.9371	0.8402
0.048	236.0002	31892	0.8324	0.9379	0.8424
0.0501	237.0002	32026	0.8272	0.9366	0.8400
0.0503	238.0002	32160	0.8997	0.9369	0.8430
0.047	239.0002	32294	0.8471	0.9380	0.8415
0.0449	240.0002	32428	0.8074	0.9359	0.8373
0.0485	241.0002	32562	0.8875	0.9318	0.8347
0.0546	242.0002	32696	0.7826	0.9346	0.8399
0.0601	243.0002	32830	0.8505	0.9390	0.8397
0.0485	244.0002	32964	0.8046	0.9395	0.8439
0.0536	245.0002	33098	0.8315	0.9372	0.8416
0.0463	246.0002	33232	0.8194	0.9383	0.8418
0.0536	247.0002	33366	0.8217	0.9375	0.8385
0.0461	248.0001	33500	0.7930	0.9388	0.8389
0.0487	249.0001	33634	0.8873	0.9390	0.8381
0.0603	250.0001	33768	0.7746	0.9331	0.8330
0.051	251.0001	33902	0.8490	0.9340	0.8359
0.0498	252.0001	34036	0.8098	0.9395	0.8417
0.0466	253.0001	34170	0.8295	0.9388	0.8410
0.0497	254.0001	34304	0.7591	0.9390	0.8442
0.0594	255.0001	34438	0.9901	0.9394	0.8402
0.0514	256.0001	34572	0.8113	0.9360	0.8419
0.0461	257.0001	34706	0.8267	0.9334	0.8364
0.0434	258.0001	34840	0.7760	0.9379	0.8386
0.0506	259.0001	34974	0.8669	0.9382	0.8408
0.0471	260.0001	35108	0.7772	0.9390	0.8402
0.0486	261.0001	35242	0.8321	0.9370	0.8405
0.0518	262.0000	35376	0.9146	0.9367	0.8392
0.0413	263.0000	35510	0.8359	0.9360	0.8413
0.0457	264.0000	35644	0.8839	0.9401	0.8442
0.0456	265.0000	35778	0.8783	0.9391	0.8431
0.0458	266.0000	35912	0.8283	0.9410	0.8461
0.0494	267.0000	36046	0.8519	0.9373	0.8374
0.0529	267.0010	36180	0.8561	0.9363	0.8440
0.0412	268.0010	36314	0.9018	0.9403	0.8450
0.0529	269.0010	36448	0.8497	0.9356	0.8282
0.0505	270.0010	36582	0.8640	0.9365	0.8379
0.0461	271.0010	36716	0.7976	0.9393	0.8413
0.0517	272.0010	36850	0.8718	0.9346	0.8372
0.0467	273.0010	36984	0.8357	0.9361	0.8399
0.0439	274.0010	37118	0.8862	0.9382	0.8425
0.0451	275.0009	37252	0.8501	0.9379	0.8431
0.047	276.0009	37386	0.7937	0.9384	0.8417
0.0495	277.0009	37520	0.8400	0.9368	0.8382
0.0463	278.0009	37654	0.7838	0.9355	0.8373
0.0489	279.0009	37788	0.8209	0.9400	0.8430
0.0442	280.0009	37922	0.8982	0.9344	0.8431
0.0495	281.0009	38056	0.7733	0.9387	0.8428
0.0414	282.0009	38190	0.8784	0.9394	0.8442
0.0392	283.0009	38324	0.9188	0.9371	0.8427
0.0431	284.0009	38458	0.8236	0.9404	0.8493
0.0487	285.0009	38592	0.8183	0.9396	0.8392
0.0464	286.0009	38726	0.8479	0.9374	0.8434
0.0416	287.0009	38860	0.8527	0.9396	0.8429
0.0454	288.0008	38994	0.8712	0.9354	0.8364
0.0514	289.0008	39128	0.8558	0.9394	0.8439
0.048	290.0008	39262	0.8637	0.9370	0.8394
0.0535	291.0008	39396	0.8481	0.9382	0.8410
0.048	292.0008	39530	0.8196	0.9404	0.8455
0.0406	293.0008	39664	0.8391	0.9399	0.8464
0.0473	294.0008	39798	0.8275	0.9406	0.8451
0.0468	295.0008	39932	0.7980	0.9398	0.8484
0.0419	296.0008	40066	0.8415	0.9403	0.8483
0.051	297.0008	40200	0.8705	0.9412	0.8453
0.0457	298.0008	40334	0.7646	0.9393	0.8454
0.0488	299.0008	40468	0.8858	0.9366	0.8446
0.0399	300.0008	40602	0.8592	0.9366	0.8453
0.0398	301.0008	40736	0.9023	0.9396	0.8458
0.0372	302.0007	40870	0.8517	0.9380	0.8453
0.0384	303.0007	41004	0.7671	0.9409	0.8462
0.0426	304.0007	41138	0.8307	0.9378	0.8466

Framework versions

• Transformers 4.46.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.1