results

This model is a fine-tuned version of laszlokiss27/doodle-dash2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7177
• Accuracy: 0.8121

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: 0.0008
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9709	0.0256	5000	0.9170	0.7612
0.9635	0.0513	10000	0.9147	0.7623
0.9518	0.0769	15000	0.9081	0.7646
0.9472	0.1026	20000	0.9044	0.7656
0.9443	0.1282	25000	0.9061	0.7660
0.93	0.1538	30000	0.9071	0.7651
0.9206	0.1795	35000	0.8963	0.7680
0.9214	0.2051	40000	0.8910	0.7693
0.912	0.2308	45000	0.8914	0.7687
0.9113	0.2564	50000	0.8801	0.7719
0.9035	0.2820	55000	0.8803	0.7723
0.9035	0.3077	60000	0.8798	0.7717
0.8898	0.3333	65000	0.8822	0.7719
0.8874	0.3590	70000	0.8703	0.7748
0.8848	0.3846	75000	0.8623	0.7764
0.8817	0.4102	80000	0.8609	0.7766
0.8765	0.4359	85000	0.8599	0.7769
0.8763	0.4615	90000	0.8532	0.7787
0.8714	0.4872	95000	0.8572	0.7774
0.869	0.5128	100000	0.8479	0.7796
0.8672	0.5384	105000	0.8480	0.7798
0.8632	0.5641	110000	0.8520	0.7792
0.8592	0.5897	115000	0.8433	0.7811
0.8607	0.6154	120000	0.8428	0.7811
0.853	0.6410	125000	0.8375	0.7827
0.8541	0.6666	130000	0.8455	0.7805
0.8473	0.6923	135000	0.8330	0.7838
0.8449	0.7179	140000	0.8305	0.7838
0.8465	0.7436	145000	0.8274	0.7850
0.8423	0.7692	150000	0.8325	0.7836
0.8454	0.7948	155000	0.8270	0.7849
0.8358	0.8205	160000	0.8328	0.7838
0.8389	0.8461	165000	0.8209	0.7868
0.8332	0.8718	170000	0.8340	0.7834
0.8357	0.8974	175000	0.8200	0.7864
0.8356	0.9230	180000	0.8162	0.7877
0.835	0.9487	185000	0.8181	0.7874
0.8298	0.9743	190000	0.8180	0.7874
0.8285	1.0000	195000	0.8154	0.7878
0.8138	1.0256	200000	0.8119	0.7889
0.8104	1.0512	205000	0.8087	0.7887
0.8162	1.0769	210000	0.8073	0.7895
0.8122	1.1025	215000	0.8053	0.7902
0.807	1.1282	220000	0.8064	0.7900
0.8114	1.1538	225000	0.8043	0.7907
0.8165	1.1794	230000	0.8042	0.7911
0.8124	1.2051	235000	0.8009	0.7910
0.8092	1.2307	240000	0.8019	0.7914
0.8023	1.2564	245000	0.7979	0.7921
0.8058	1.2820	250000	0.7988	0.7922
0.8057	1.3076	255000	0.7976	0.7923
0.8076	1.3333	260000	0.7976	0.7921
0.805	1.3589	265000	0.7953	0.7930
0.797	1.3846	270000	0.7990	0.7926
0.7997	1.4102	275000	0.7929	0.7935
0.8028	1.4358	280000	0.7933	0.7933
0.7981	1.4615	285000	0.7905	0.7934
0.8002	1.4871	290000	0.7965	0.7924
0.7984	1.5128	295000	0.7915	0.7933
0.7973	1.5384	300000	0.7950	0.7932
0.7933	1.5640	305000	0.7865	0.7950
0.7927	1.5897	310000	0.7886	0.7946
0.799	1.6153	315000	0.7840	0.7954
0.7961	1.6410	320000	0.8132	0.7901
0.7866	1.6666	325000	0.7829	0.7958
0.7898	1.6922	330000	0.7813	0.7959
0.7885	1.7179	335000	0.7796	0.7969
0.7901	1.7435	340000	0.7817	0.7958
0.7916	1.7692	345000	0.7823	0.7962
0.787	1.7948	350000	0.7789	0.7969
0.7822	1.8204	355000	0.7787	0.7968
0.7844	1.8461	360000	0.7754	0.7981
0.7849	1.8717	365000	0.7775	0.7972
0.7845	1.8974	370000	0.7761	0.7973
0.7905	1.9230	375000	0.7736	0.7983
0.788	1.9486	380000	0.7738	0.7978
0.7832	1.9743	385000	0.7719	0.7980
0.7787	1.9999	390000	0.7710	0.7986
0.767	2.0256	395000	0.7717	0.7985
0.7666	2.0512	400000	0.7698	0.7989
0.7631	2.0768	405000	0.7719	0.7982
0.7634	2.1025	410000	0.7684	0.7994
0.7621	2.1281	415000	0.7707	0.7987
0.7694	2.1538	420000	0.7700	0.7994
0.7648	2.1794	425000	0.7678	0.7995
0.7612	2.2050	430000	0.7673	0.7995
0.7627	2.2307	435000	0.7671	0.7997
0.766	2.2563	440000	0.7649	0.8003
0.7635	2.2820	445000	0.7653	0.8000
0.761	2.3076	450000	0.7647	0.8000
0.7649	2.3332	455000	0.7661	0.8001
0.7589	2.3589	460000	0.7630	0.8005
0.7586	2.3845	465000	0.7703	0.7988
0.7595	2.4102	470000	0.7640	0.8003
0.7622	2.4358	475000	0.7627	0.8005
0.7593	2.4614	480000	0.7605	0.8013
0.7558	2.4871	485000	0.7609	0.8012
0.7599	2.5127	490000	0.7651	0.8002
0.7587	2.5384	495000	0.7589	0.8016
0.7588	2.5640	500000	0.7570	0.8024
0.762	2.5896	505000	0.7566	0.8020
0.7526	2.6153	510000	0.7602	0.8013
0.7587	2.6409	515000	0.7560	0.8021
0.7522	2.6666	520000	0.7557	0.8026
0.7546	2.6922	525000	0.7542	0.8026
0.7542	2.7178	530000	0.7543	0.8029
0.7509	2.7435	535000	0.7542	0.8029
0.7515	2.7691	540000	0.7585	0.8016
0.7508	2.7948	545000	0.7553	0.8024
0.7523	2.8204	550000	0.7531	0.8028
0.756	2.8460	555000	0.7511	0.8035
0.7559	2.8717	560000	0.7500	0.8038
0.75	2.8973	565000	0.7494	0.8038
0.7492	2.9230	570000	0.7511	0.8035
0.7481	2.9486	575000	0.7471	0.8044
0.751	2.9742	580000	0.7478	0.8043
0.7545	2.9999	585000	0.7595	0.8019
0.7299	3.0255	590000	0.7478	0.8042
0.7305	3.0512	595000	0.7487	0.8047
0.7343	3.0768	600000	0.7466	0.8047
0.731	3.1024	605000	0.7472	0.8045
0.733	3.1281	610000	0.7460	0.8046
0.7351	3.1537	615000	0.7486	0.8043
0.7372	3.1794	620000	0.7446	0.8052
0.7299	3.2050	625000	0.7478	0.8045
0.7351	3.2306	630000	0.7458	0.8047
0.7304	3.2563	635000	0.7460	0.8049
0.7335	3.2819	640000	0.7451	0.8049
0.7351	3.3076	645000	0.7416	0.8058
0.7324	3.3332	650000	0.7420	0.8058
0.732	3.3588	655000	0.7426	0.8057
0.7286	3.3845	660000	0.7418	0.8062
0.7331	3.4101	665000	0.7420	0.8059
0.729	3.4358	670000	0.7402	0.8065
0.7336	3.4614	675000	0.7409	0.8063
0.7275	3.4870	680000	0.7398	0.8064
0.7298	3.5127	685000	0.7388	0.8069
0.724	3.5383	690000	0.7365	0.8070
0.7266	3.5640	695000	0.7373	0.8072
0.7282	3.5896	700000	0.7371	0.8074
0.7272	3.6152	705000	0.7360	0.8073
0.7227	3.6409	710000	0.7360	0.8072
0.7275	3.6665	715000	0.7358	0.8073
0.7299	3.6922	720000	0.7422	0.8063
0.7363	3.7178	725000	0.7361	0.8072
0.7274	3.7434	730000	0.7334	0.8082
0.7282	3.7691	735000	0.7347	0.8081
0.7239	3.7947	740000	0.7326	0.8085
0.7225	3.8204	745000	0.7352	0.8076
0.7242	3.8460	750000	0.7320	0.8086
0.7291	3.8716	755000	0.7317	0.8089
0.7292	3.8973	760000	0.7310	0.8087
0.7247	3.9229	765000	0.7310	0.8083
0.7286	3.9486	770000	0.7326	0.8084
0.7237	3.9742	775000	0.7303	0.8088
0.7187	3.9998	780000	0.7298	0.8090
0.7077	4.0255	785000	0.7316	0.8084
0.7108	4.0511	790000	0.7316	0.8084
0.7025	4.0768	795000	0.7300	0.8093
0.708	4.1024	800000	0.7295	0.8093
0.7067	4.1280	805000	0.7288	0.8094
0.7123	4.1537	810000	0.7287	0.8094
0.707	4.1793	815000	0.7283	0.8095
0.7033	4.2050	820000	0.7282	0.8099
0.7128	4.2306	825000	0.7272	0.8099
0.7053	4.2562	830000	0.7284	0.8095
0.7097	4.2819	835000	0.7268	0.8098
0.7101	4.3075	840000	0.7267	0.8097
0.7074	4.3332	845000	0.7261	0.8102
0.7034	4.3588	850000	0.7257	0.8101
0.7059	4.3844	855000	0.7262	0.8098
0.7008	4.4101	860000	0.7247	0.8100
0.7021	4.4357	865000	0.7241	0.8103
0.707	4.4614	870000	0.7243	0.8105
0.7034	4.4870	875000	0.7238	0.8106
0.7055	4.5126	880000	0.7233	0.8106
0.7056	4.5383	885000	0.7231	0.8107
0.7029	4.5639	890000	0.7226	0.8108
0.7048	4.5896	895000	0.7224	0.8111
0.7031	4.6152	900000	0.7221	0.8110
0.7034	4.6408	905000	0.7216	0.8112
0.7012	4.6665	910000	0.7218	0.8113
0.702	4.6921	915000	0.7209	0.8114
0.7018	4.7178	920000	0.7207	0.8115
0.7056	4.7434	925000	0.7201	0.8116
0.7005	4.7690	930000	0.7199	0.8118
0.7005	4.7947	935000	0.7197	0.8117
0.708	4.8203	940000	0.7189	0.8117
0.6956	4.8460	945000	0.7190	0.8118
0.7074	4.8716	950000	0.7185	0.8120
0.6964	4.8972	955000	0.7184	0.8121
0.7048	4.9229	960000	0.7188	0.8120
0.7018	4.9485	965000	0.7178	0.8122
0.7006	4.9742	970000	0.7177	0.8121
0.7005	4.9998	975000	0.7177	0.8121

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.2+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1