vit-base-game-icons Open-source Image Classification Model - Free Identification of Images Related to Game Ads

Home

Vit Base Game Icons

Developed by chromefan

An image classification model based on ViT architecture, fine-tuned on a game advertisement dataset

Image Classification

Transformers

Open Source License:Apache-2.0 #Advertisement Image Classification #ViT Fine-tuned Model #Low Accuracy Scenarios

Downloads 18

Release Time : 3/16/2023

Model Overview

This model is an image classification model fine-tuned on a game advertisement image dataset, based on the google/vit-base-patch16-224-in21k pre-trained model. It is primarily used for game advertisement image classification tasks.

Model Features

Based on ViT Architecture

Uses Vision Transformer (ViT) architecture, capable of effectively processing image data

Transfer Learning

Fine-tuned based on a pre-trained model, suitable for image classification tasks in specific domains

Model Capabilities

Image Classification

Game Advertisement Recognition

Use Cases

Advertisement Analysis

Game Advertisement Classification

Classify and recognize game advertisement images

Evaluation accuracy 30.24%

🚀 game-ad-0306_outputs

This model is a fine - tuned version of [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k) on the ./data/games - ad - 0306 dataset. It offers an effective solution for image classification tasks. On the evaluation set, it achieves a Loss of 2.6235 and an Accuracy of 0.3024.

🚀 Quick Start

This model is ready for use in image classification tasks. You can fine - tune it further based on your specific requirements.

✨ Features

Fine - tuned Model: Built on the foundation of [google/vit - base - patch16 - 224 - in21k], it has been fine - tuned on a specific dataset for better performance.
Evaluation Metrics: Provides clear evaluation metrics such as Loss and Accuracy, allowing users to assess its performance.

🔧 Technical Details

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e - 05
train_batch_size: 16
eval_batch_size: 16
seed: 13373
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
num_epochs: 1000.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.2891	1.0	103	3.0266	0.2165
2.9971	2.0	206	2.9194	0.2302
2.9151	3.0	309	2.8731	0.2474
2.8579	4.0	412	2.8072	0.2715
2.7768	5.0	515	2.7918	0.2577
2.7184	6.0	618	2.7296	0.2818
2.648	7.0	721	2.7044	0.2921
2.5884	8.0	824	2.7190	0.2680
2.5146	9.0	927	2.6942	0.2784
2.4384	10.0	1030	2.6877	0.2921
2.442	11.0	1133	2.6412	0.2818
2.3099	12.0	1236	2.6331	0.2852
2.2685	13.0	1339	2.6451	0.2990
2.182	14.0	1442	2.6927	0.2715
2.1421	15.0	1545	2.6615	0.3162
2.0483	16.0	1648	2.6500	0.3230
1.9884	17.0	1751	2.6527	0.2990
1.9316	18.0	1854	2.6736	0.2990
1.8785	19.0	1957	2.6391	0.2921
1.788	20.0	2060	2.7002	0.3127
1.7115	21.0	2163	2.8321	0.2715
1.6929	22.0	2266	2.6235	0.3024
1.6239	23.0	2369	2.6378	0.3058
1.5387	24.0	2472	2.6888	0.3127
1.5095	25.0	2575	2.6888	0.3127
1.4153	26.0	2678	2.6771	0.2715
1.4254	27.0	2781	2.7354	0.2887
1.3351	28.0	2884	2.7175	0.2990
1.2955	29.0	2987	2.7679	0.2818
1.2232	30.0	3090	2.7784	0.2921
1.2115	31.0	3193	2.8496	0.2749
1.1656	32.0	3296	2.7899	0.2818
1.1419	33.0	3399	2.7646	0.2715
1.0481	34.0	3502	2.8416	0.2715
0.9763	35.0	3605	2.8370	0.3024
0.9452	36.0	3708	2.7904	0.2955
0.9178	37.0	3811	2.8309	0.2715
0.9115	38.0	3914	2.8584	0.3093
0.8472	39.0	4017	2.9066	0.2612
0.8323	40.0	4120	2.8630	0.2921
0.7622	41.0	4223	3.0020	0.2680
0.7531	42.0	4326	2.8885	0.2921
0.7054	43.0	4429	2.8820	0.2818
0.685	44.0	4532	2.8764	0.3162
0.7206	45.0	4635	2.8659	0.3162
0.6304	46.0	4738	2.9537	0.2887
0.6369	47.0	4841	2.9660	0.2509
0.6161	48.0	4944	3.1112	0.2543
0.618	49.0	5047	2.9729	0.2990
0.556	50.0	5150	2.9870	0.2921
0.5314	51.0	5253	2.9934	0.3093
0.5502	52.0	5356	2.9379	0.2818
0.4958	53.0	5459	3.0344	0.3024
0.4896	54.0	5562	2.9924	0.2749
0.4803	55.0	5665	3.0161	0.3127
0.4554	56.0	5768	3.0221	0.2818
0.4591	57.0	5871	3.0461	0.3024
0.4349	58.0	5974	3.1377	0.3265
0.4127	59.0	6077	3.0169	0.2955
0.3973	60.0	6180	3.0338	0.2818
0.4109	61.0	6283	3.0638	0.2818
0.3872	62.0	6386	3.0810	0.2818
0.3693	63.0	6489	3.2003	0.2715
0.3457	64.0	6592	3.0843	0.2990
0.3521	65.0	6695	3.1623	0.3058
0.3625	66.0	6798	3.0036	0.3299
0.3339	67.0	6901	3.2389	0.2921
0.3378	68.0	7004	3.2493	0.2990
0.2981	69.0	7107	3.1308	0.2955
0.3023	70.0	7210	3.2455	0.3093
0.3076	71.0	7313	3.2725	0.2887
0.3201	72.0	7416	3.2563	0.2887
0.3083	73.0	7519	3.2520	0.2921
0.2906	74.0	7622	3.3344	0.3093
0.2721	75.0	7725	3.1952	0.2852
0.2873	76.0	7828	3.2529	0.3058
0.278	77.0	7931	3.3428	0.2818
0.2573	78.0	8034	3.3216	0.2784
0.2578	79.0	8137	3.4178	0.2955
0.2774	80.0	8240	3.3449	0.2818
0.2762	81.0	8343	3.3452	0.2749
0.2504	82.0	8446	3.5792	0.2955
0.2552	83.0	8549	3.3478	0.2818
0.2541	84.0	8652	3.4902	0.2784
0.2616	85.0	8755	3.2829	0.3127
0.2079	86.0	8858	3.5287	0.3162
0.2538	87.0	8961	3.4731	0.3196
0.2485	88.0	9064	3.5998	0.2646
0.2714	89.0	9167	3.4567	0.2921
0.232	90.0	9270	3.5061	0.2818
0.2577	91.0	9373	3.5370	0.2921
0.2232	92.0	9476	3.5062	0.2509
0.2351	93.0	9579	3.5592	0.2784
0.2299	94.0	9682	3.5167	0.3333
0.2415	95.0	9785	3.6283	0.2887
0.2265	96.0	9888	3.4819	0.2852
0.2448	97.0	9991	3.5793	0.2990
0.2141	98.0	10094	3.5728	0.2887
0.1979	99.0	10197	3.4685	0.2921
0.2077	100.0	10300	3.5586	0.3230
0.1854	101.0	10403	3.5650	0.3162
0.2017	102.0	10506	3.4760	0.2921
0.2119	103.0	10609	3.5531	0.2784
0.2314	104.0	10712	3.5118	0.3024
0.212	105.0	10815	3.5496	0.3196
0.197	106.0	10918	3.6080	0.2543
0.2067	107.0	11021	3.6217	0.2887
0.1896	108.0	11124	3.6446	0.3230
0.198	109.0	11227	3.7699	0.2784
0.2152	110.0	11330	3.6709	0.3162
0.2121	111.0	11433	3.6266	0.3368
0.1869	112.0	11536	3.6681	0.2955
0.1927	113.0	11639	3.7305	0.3162
0.2259	114.0	11742	3.6302	0.3127
0.1809	115.0	11845	3.6301	0.3093
0.2071	116.0	11948	3.7288	0.3127
0.1977	117.0	12051	3.6467	0.3058
0.1902	118.0	12154	3.7039	0.3093
0.1996	119.0	12257	3.9013	0.3093
0.2122	120.0	12360	3.8228	0.2990
0.1702	121.0	12463	3.7118	0.3162
0.1889	122.0	12566	3.7211	0.3162
0.1857	123.0	12669	3.8894	0.2509
0.2003	124.0	12772	3.6575	0.3093
0.202	125.0	12875	3.7925	0.3333
0.1722	126.0	12978	3.8188	0.2818
0.1716	127.0	13081	3.9584	0.3162
0.1598	128.0	13184	3.7732	0.3265
0.1825	129.0	13287	3.8038	0.3196
0.1716	130.0	13390	3.7606	0.3196
0.179	131.0	13493	3.7458	0.2955
0.1817	132.0	13596	3.8413	0.2955
0.1606	133.0	13699	3.8766	0.3196
0.1625	134.0	13802	3.8188	0.3230
0.1622	135.0	13905	3.7223	0.2955
0.1852	136.0	14008	3.7774	0.3024
0.1671	137.0	14111	3.8407	0.2612
0.1862	138.0	14214	3.7442	0.3196
0.1808	139.0	14317	3.8458	0.3093
0.1375	140.0	14420	3.7372	0.3024
0.1876	141.0	14523	3.9925	0.2990
0.1693	142.0	14626	3.9364	0.3058
0.1719	143.0	14729	3.9149	0.2818
0.1406	144.0	14832	3.8603	0.2955
0.1709	145.0	14935	3.9216	0.3196
0.1794	146.0	15038	3.8934	0.3058
0.1455	147.0	15141	4.0086	0.2784
0.1959	148.0	15244	3.9358	0.3024
0.1664	149.0	15347	3.9775	0.2921
0.1455	150.0	15450	3.9304	0.2990
0.1819	151.0	15553	4.0299	0.2715
0.1532	152.0	15656	4.1219	0.2680
0.1638	153.0	15759	4.1465	0.3093
0.1579	154.0	15862	4.0596	0.2955
0.1668	155.0	15965	4.0857	0.3127
0.1401	156.0	16068	4.1669	0.2921
0.1452	157.0	16171	4.0430	0.2887
0.1568	158.0	16274	4.0157	0.2990
0.1771	159.0	16377	4.0770	0.3093
0.1383	160.0	16480	4.0888	0.2680
0.1572	161.0	16583	4.2271	0.2646
0.1472	162.0	16686	4.0215	0.2852
0.1534	163.0	16789	4.2248	0.3058
0.136	164.0	16892	4.2159	0.2852
0.1525	165.0	16995	4.0565	0.2990
0.1418	166.0	17098	4.1175	0.2852
0.1374	167.0	17201	4.1708	0.2921
0.1538	168.0	17304	4.2566	0.2784
0.1365	169.0	17407	4.3063	0.2577
0.1661	170.0	17510	4.2231	0.2887
0.1278	171.0	17613	4.3125	0.2646
0.1418	172.0	17716	4.3337	0.2646
0.1538	173.0	17819	4.3129	0.2852
0.1315	174.0	17922	4.3102	0.2680
0.128	175.0	18025	4.2853	0.2749
0.1398	176.0	18128	4.1560	0.2921

📄 License

This project is licensed under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご