videomae-base-finetuned-chickenbehaviour-2

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1357
• Accuracy: 0.6697
• Precision: 0.6429
• Recall: 0.6697
• F1: 0.6354

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: 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_ratio: 0.1
• training_steps: 127240

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
2.035	0.01	1591	1.6961	0.5116	0.4264	0.5116	0.4139
1.5431	1.01	3182	1.4395	0.5898	0.5240	0.5898	0.5167
1.4118	2.01	4773	1.3632	0.6051	0.5553	0.6051	0.5535
1.3413	3.01	6364	1.3312	0.6021	0.5516	0.6021	0.5384
1.2969	4.01	7955	1.2739	0.6212	0.6122	0.6212	0.5663
1.2636	5.01	9546	1.3212	0.6058	0.6187	0.6058	0.5430
1.2231	6.01	11137	1.2543	0.6242	0.6461	0.6242	0.5747
1.1989	7.01	12728	1.2378	0.6405	0.6356	0.6405	0.5869
1.1566	8.01	14319	1.2124	0.6528	0.6199	0.6528	0.5942
1.1145	9.01	15910	1.1803	0.6476	0.6341	0.6476	0.6052
1.0567	10.01	17501	1.2577	0.6266	0.6279	0.6266	0.5969
1.0172	11.01	19092	1.1961	0.6570	0.6369	0.6570	0.6083
0.9817	12.01	20683	1.2287	0.6620	0.6499	0.6620	0.6049
0.9279	13.01	22274	1.2358	0.6549	0.6504	0.6549	0.6213
0.8913	14.01	23865	1.1815	0.6681	0.6325	0.6681	0.6308
0.8559	15.01	25456	1.3212	0.6391	0.6392	0.6391	0.6037
0.8083	16.01	27047	1.3073	0.6231	0.6251	0.6231	0.6006
0.7662	17.01	28638	1.2982	0.6462	0.6252	0.6462	0.6214
0.7363	18.01	30229	1.3019	0.6575	0.6428	0.6575	0.6264
0.6787	19.01	31820	1.3867	0.6511	0.6368	0.6511	0.6230
0.6433	20.01	33411	1.4019	0.6365	0.6375	0.6365	0.6139
0.5969	21.01	35002	1.4419	0.6341	0.6212	0.6341	0.6104
0.563	22.01	36593	1.4778	0.6509	0.6293	0.6509	0.6170
0.5252	23.01	38184	1.4864	0.6433	0.6316	0.6433	0.6214
0.5	24.01	39775	1.6704	0.6233	0.6273	0.6233	0.6023
0.4622	25.01	41366	1.6658	0.6488	0.6260	0.6488	0.6119
0.4292	26.01	42957	1.6428	0.6495	0.6287	0.6495	0.6243
0.4044	27.01	44548	1.6703	0.6587	0.6311	0.6587	0.6387
0.3952	28.01	46139	1.7576	0.6330	0.6171	0.6330	0.6123
0.3681	29.01	47730	1.9032	0.6554	0.6349	0.6554	0.6231
0.3541	30.01	49321	1.9508	0.6445	0.6320	0.6445	0.6207
0.322	31.01	50912	2.1317	0.6226	0.6277	0.6226	0.6099
0.3239	32.01	52503	1.9785	0.6509	0.6321	0.6509	0.6328
0.301	33.01	54094	2.2050	0.6436	0.6259	0.6436	0.6097
0.28	34.01	55685	2.2268	0.6320	0.6319	0.6320	0.6174
0.2742	35.01	57276	2.3538	0.6419	0.6239	0.6419	0.6158
0.2433	36.01	58867	2.3947	0.6478	0.6237	0.6478	0.6184
0.2677	37.01	60458	2.4007	0.6455	0.6285	0.6455	0.6234
0.2316	38.01	62049	2.5197	0.6297	0.6246	0.6297	0.6120
0.2229	39.01	63640	2.5478	0.6506	0.6322	0.6506	0.6235
0.215	40.01	65231	2.5168	0.6445	0.6455	0.6445	0.6209
0.2032	41.01	66822	2.6607	0.6443	0.6304	0.6443	0.6161
0.1957	42.01	68413	2.6434	0.6219	0.6206	0.6219	0.6059
0.1839	43.01	70004	2.6378	0.6480	0.6182	0.6480	0.6202
0.1672	44.01	71595	2.8355	0.6330	0.6175	0.6330	0.6095
0.1554	45.01	73186	2.8833	0.6297	0.6180	0.6297	0.6090
0.1525	46.01	74777	2.8732	0.6499	0.6212	0.6499	0.6247
0.1443	47.01	76368	2.7936	0.6513	0.6240	0.6513	0.6297
0.1361	48.01	77959	2.8815	0.6443	0.6187	0.6443	0.6230
0.1351	49.01	79550	3.0703	0.6429	0.6244	0.6429	0.6175
0.1196	50.01	81141	3.0275	0.6424	0.6250	0.6424	0.6190
0.111	51.01	82732	3.1255	0.6419	0.6281	0.6419	0.6189
0.1119	52.01	84323	3.1854	0.6471	0.6299	0.6471	0.6215
0.1069	53.01	85914	3.2136	0.6384	0.6251	0.6384	0.6195
0.093	54.01	87505	3.3125	0.6506	0.6145	0.6506	0.6155
0.0901	55.01	89096	3.3028	0.6384	0.6277	0.6384	0.6217
0.0776	56.01	90687	3.3315	0.6488	0.6272	0.6488	0.6298
0.0837	57.01	92278	3.4385	0.6558	0.6374	0.6558	0.6242
0.0701	58.01	93869	3.3800	0.6440	0.6321	0.6440	0.6286
0.0682	59.01	95460	3.4473	0.6542	0.6344	0.6542	0.6262
0.0763	60.01	97051	3.4505	0.6315	0.6149	0.6315	0.6148
0.0629	61.01	98642	3.4402	0.6504	0.6233	0.6504	0.6253
0.0552	62.01	100233	3.4402	0.6537	0.6324	0.6537	0.6315
0.0463	63.01	101824	3.5300	0.6466	0.6217	0.6466	0.6217
0.0471	64.01	103415	3.6793	0.6511	0.6346	0.6511	0.6223
0.0448	65.01	105006	3.6850	0.6450	0.6265	0.6450	0.6170
0.0362	66.01	106597	3.6585	0.6483	0.6265	0.6483	0.6242
0.0419	67.01	108188	3.6285	0.6344	0.6192	0.6344	0.6169
0.0309	68.01	109779	3.6657	0.6490	0.6264	0.6490	0.6269
0.0312	69.01	111370	3.7123	0.6417	0.6239	0.6417	0.6205
0.0315	70.01	112961	3.7538	0.6490	0.6224	0.6490	0.6189
0.0294	71.01	114552	3.7064	0.6483	0.6234	0.6483	0.6237
0.0282	72.01	116143	3.7945	0.6429	0.6247	0.6429	0.6192
0.0275	73.01	117734	3.7550	0.6528	0.6297	0.6528	0.6272
0.0319	74.01	119325	3.7407	0.6509	0.6289	0.6509	0.6234
0.021	75.01	120916	3.7527	0.6532	0.6290	0.6532	0.6270
0.0159	76.01	122507	3.7780	0.6516	0.6241	0.6516	0.6243
0.0133	77.01	124098	3.7923	0.6499	0.6272	0.6499	0.6240
0.0125	78.01	125689	3.8070	0.6504	0.6263	0.6504	0.6217
0.0132	79.01	127240	3.7964	0.6506	0.6264	0.6506	0.6225

Framework versions

• Transformers 4.39.1
• Pytorch 2.1.0
• Datasets 2.18.0
• Tokenizers 0.15.2