detr_algae_0.25r_v0

This model is a fine-tuned version of facebook/detr-resnet-50 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.1346

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: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 200
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
5.1721	1.0	69	3.3857
3.3617	2.0	138	2.9312
3.0175	3.0	207	2.6671
2.9799	4.0	276	2.7044
2.835	5.0	345	2.7200
2.7184	6.0	414	2.4760
2.7039	7.0	483	2.5395
2.6316	8.0	552	2.4040
2.683	9.0	621	2.4622
2.6908	10.0	690	2.5163
2.5706	11.0	759	2.2428
2.5066	12.0	828	2.3053
2.5146	13.0	897	2.2936
2.446	14.0	966	2.1469
2.3444	15.0	1035	2.2260
2.3997	16.0	1104	2.2093
2.3823	17.0	1173	2.2598
2.3741	18.0	1242	2.1737
2.3437	19.0	1311	2.1670
2.3134	20.0	1380	2.2041
2.3636	21.0	1449	2.1305
2.2936	22.0	1518	2.0564
2.2117	23.0	1587	2.0024
2.1955	24.0	1656	2.0013
2.1724	25.0	1725	1.9785
2.1765	26.0	1794	2.0120
2.1489	27.0	1863	1.9891
2.157	28.0	1932	1.9267
2.1454	29.0	2001	1.9686
2.1545	30.0	2070	2.0207
2.0847	31.0	2139	1.8773
2.0731	32.0	2208	1.8358
2.0717	33.0	2277	1.9076
2.0616	34.0	2346	1.9287
2.1044	35.0	2415	1.8625
2.04	36.0	2484	1.8683
2.0489	37.0	2553	1.8393
2.0562	38.0	2622	1.9013
2.0433	39.0	2691	1.8472
2.0093	40.0	2760	1.8122
2.0149	41.0	2829	1.7668
2.049	42.0	2898	1.8419
1.9992	43.0	2967	1.8292
1.9494	44.0	3036	1.8329
2.0128	45.0	3105	1.8827
2.0712	46.0	3174	1.8425
1.9346	47.0	3243	1.8509
1.899	48.0	3312	1.7352
1.9576	49.0	3381	1.7825
1.9877	50.0	3450	1.7996
1.9176	51.0	3519	1.7754
1.9217	52.0	3588	1.7418
1.9365	53.0	3657	1.7711
1.9032	54.0	3726	1.7001
1.8404	55.0	3795	1.6628
1.8447	56.0	3864	1.6939
1.8418	57.0	3933	1.7099
1.7911	58.0	4002	1.6751
1.7899	59.0	4071	1.7471
1.8368	60.0	4140	1.7111
1.853	61.0	4209	1.7785
1.88	62.0	4278	1.7709
1.8734	63.0	4347	1.6597
1.8107	64.0	4416	1.6720
1.8329	65.0	4485	1.6868
1.8129	66.0	4554	1.6611
1.7972	67.0	4623	1.6452
1.7828	68.0	4692	1.6538
1.7653	69.0	4761	1.6246
1.7343	70.0	4830	1.5364
1.6567	71.0	4899	1.5308
1.6873	72.0	4968	1.5473
1.7233	73.0	5037	1.6096
1.6934	74.0	5106	1.5679
1.7263	75.0	5175	1.6542
1.7109	76.0	5244	1.5674
1.6977	77.0	5313	1.5367
1.6761	78.0	5382	1.5456
1.69	79.0	5451	1.5624
1.7241	80.0	5520	1.5067
1.643	81.0	5589	1.5723
1.6358	82.0	5658	1.5349
1.6511	83.0	5727	1.5321
1.6932	84.0	5796	1.5640
1.7214	85.0	5865	1.5118
1.6988	86.0	5934	1.5471
1.6697	87.0	6003	1.5650
1.6828	88.0	6072	1.5087
1.7211	89.0	6141	1.5302
1.6195	90.0	6210	1.5018
1.5924	91.0	6279	1.4886
1.5746	92.0	6348	1.4365
1.6277	93.0	6417	1.4995
1.5936	94.0	6486	1.4569
1.6132	95.0	6555	1.4982
1.5637	96.0	6624	1.4032
1.5502	97.0	6693	1.4388
1.5535	98.0	6762	1.4101
1.5306	99.0	6831	1.4048
1.5425	100.0	6900	1.4133
1.529	101.0	6969	1.4244
1.5659	102.0	7038	1.4268
1.5234	103.0	7107	1.3829
1.498	104.0	7176	1.3884
1.4838	105.0	7245	1.3627
1.4774	106.0	7314	1.3501
1.479	107.0	7383	1.3738
1.475	108.0	7452	1.3537
1.4592	109.0	7521	1.3621
1.5015	110.0	7590	1.4022
1.4948	111.0	7659	1.4069
1.4875	112.0	7728	1.3325
1.437	113.0	7797	1.3080
1.4276	114.0	7866	1.3100
1.4489	115.0	7935	1.3712
1.4951	116.0	8004	1.4256
1.4585	117.0	8073	1.3720
1.4736	118.0	8142	1.4397
1.4664	119.0	8211	1.4036
1.4569	120.0	8280	1.3672
1.4627	121.0	8349	1.3809
1.4924	122.0	8418	1.3420
1.4487	123.0	8487	1.3142
1.4341	124.0	8556	1.3238
1.4025	125.0	8625	1.2891
1.4013	126.0	8694	1.3140
1.3909	127.0	8763	1.3329
1.4305	128.0	8832	1.3489
1.3771	129.0	8901	1.3640
1.4442	130.0	8970	1.3695
1.4272	131.0	9039	1.3752
1.4087	132.0	9108	1.3145
1.3648	133.0	9177	1.3222
1.3981	134.0	9246	1.3483
1.4116	135.0	9315	1.2986
1.4117	136.0	9384	1.3789
1.4416	137.0	9453	1.3491
1.3753	138.0	9522	1.3026
1.3721	139.0	9591	1.3292
1.3951	140.0	9660	1.2946
1.3406	141.0	9729	1.2646
1.3336	142.0	9798	1.3247
1.3182	143.0	9867	1.2960
1.3293	144.0	9936	1.2845
1.3242	145.0	10005	1.2849
1.3171	146.0	10074	1.2662
1.3193	147.0	10143	1.2827
1.3298	148.0	10212	1.2776
1.3014	149.0	10281	1.2603
1.3419	150.0	10350	1.2484
1.3385	151.0	10419	1.2477
1.3029	152.0	10488	1.2408
1.2803	153.0	10557	1.2191
1.2562	154.0	10626	1.2264
1.2667	155.0	10695	1.2250
1.2669	156.0	10764	1.2095
1.2643	157.0	10833	1.1964
1.2548	158.0	10902	1.2019
1.262	159.0	10971	1.2393
1.2596	160.0	11040	1.2018
1.2476	161.0	11109	1.2308
1.2755	162.0	11178	1.2103
1.237	163.0	11247	1.2020
1.226	164.0	11316	1.2173
1.2278	165.0	11385	1.1912
1.244	166.0	11454	1.2036
1.2467	167.0	11523	1.1831
1.2063	168.0	11592	1.1720
1.2141	169.0	11661	1.1580
1.224	170.0	11730	1.1897
1.2171	171.0	11799	1.1546
1.2151	172.0	11868	1.1621
1.1623	173.0	11937	1.1899
1.2037	174.0	12006	1.1649
1.1741	175.0	12075	1.1794
1.1921	176.0	12144	1.1584
1.1811	177.0	12213	1.1589
1.1956	178.0	12282	1.1555
1.1703	179.0	12351	1.1510
1.1727	180.0	12420	1.1363
1.1747	181.0	12489	1.1570
1.1524	182.0	12558	1.1655
1.1645	183.0	12627	1.1324
1.1549	184.0	12696	1.1529
1.1432	185.0	12765	1.1396
1.1552	186.0	12834	1.1406
1.1568	187.0	12903	1.1585
1.1407	188.0	12972	1.1417
1.1419	189.0	13041	1.1542
1.1451	190.0	13110	1.1330
1.1421	191.0	13179	1.1309
1.1283	192.0	13248	1.1271
1.1528	193.0	13317	1.1195
1.1367	194.0	13386	1.1300
1.1407	195.0	13455	1.1144
1.1456	196.0	13524	1.1584
1.1072	197.0	13593	1.1334
1.1081	198.0	13662	1.1378
1.1205	199.0	13731	1.1327
1.1275	200.0	13800	1.1346

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu117
• Datasets 2.9.0
• Tokenizers 0.13.2