vit-gpt-diffusion-zh / trainer_state.json
svjack's picture
Upload with huggingface_hub
7d9efca
{
"best_metric": 1.1318830251693726,
"best_model_checkpoint": "./image_captioning_checkpoint/checkpoint-16000",
"epoch": 1.702127659574468,
"global_step": 20000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.09,
"learning_rate": 4.9150638297872343e-05,
"loss": 1.6148,
"step": 1000
},
{
"epoch": 0.09,
"eval_loss": 1.3645786046981812,
"eval_runtime": 33.5956,
"eval_samples_per_second": 10.329,
"eval_steps_per_second": 5.179,
"step": 1000
},
{
"epoch": 0.17,
"learning_rate": 4.829957446808511e-05,
"loss": 1.262,
"step": 2000
},
{
"epoch": 0.17,
"eval_loss": 1.2903200387954712,
"eval_runtime": 33.3017,
"eval_samples_per_second": 10.42,
"eval_steps_per_second": 5.225,
"step": 2000
},
{
"epoch": 0.26,
"learning_rate": 4.744851063829788e-05,
"loss": 1.1841,
"step": 3000
},
{
"epoch": 0.26,
"eval_loss": 1.3015347719192505,
"eval_runtime": 33.3088,
"eval_samples_per_second": 10.418,
"eval_steps_per_second": 5.224,
"step": 3000
},
{
"epoch": 0.34,
"learning_rate": 4.659744680851064e-05,
"loss": 1.1072,
"step": 4000
},
{
"epoch": 0.34,
"eval_loss": 1.203688383102417,
"eval_runtime": 32.4391,
"eval_samples_per_second": 10.697,
"eval_steps_per_second": 5.364,
"step": 4000
},
{
"epoch": 0.43,
"learning_rate": 4.5746382978723406e-05,
"loss": 1.0512,
"step": 5000
},
{
"epoch": 0.43,
"eval_loss": 1.1912097930908203,
"eval_runtime": 32.1233,
"eval_samples_per_second": 10.802,
"eval_steps_per_second": 5.417,
"step": 5000
},
{
"epoch": 0.51,
"learning_rate": 4.489531914893617e-05,
"loss": 1.0235,
"step": 6000
},
{
"epoch": 0.51,
"eval_loss": 1.186132788658142,
"eval_runtime": 32.9752,
"eval_samples_per_second": 10.523,
"eval_steps_per_second": 5.277,
"step": 6000
},
{
"epoch": 0.6,
"learning_rate": 4.404510638297872e-05,
"loss": 0.9872,
"step": 7000
},
{
"epoch": 0.6,
"eval_loss": 1.2038156986236572,
"eval_runtime": 32.8457,
"eval_samples_per_second": 10.565,
"eval_steps_per_second": 5.298,
"step": 7000
},
{
"epoch": 0.68,
"learning_rate": 4.3194042553191494e-05,
"loss": 0.949,
"step": 8000
},
{
"epoch": 0.68,
"eval_loss": 1.2001100778579712,
"eval_runtime": 31.8799,
"eval_samples_per_second": 10.885,
"eval_steps_per_second": 5.458,
"step": 8000
},
{
"epoch": 0.77,
"learning_rate": 4.234297872340426e-05,
"loss": 0.9352,
"step": 9000
},
{
"epoch": 0.77,
"eval_loss": 1.1584669351577759,
"eval_runtime": 31.0962,
"eval_samples_per_second": 11.159,
"eval_steps_per_second": 5.596,
"step": 9000
},
{
"epoch": 0.85,
"learning_rate": 4.149276595744681e-05,
"loss": 0.9224,
"step": 10000
},
{
"epoch": 0.85,
"eval_loss": 1.1474857330322266,
"eval_runtime": 32.1504,
"eval_samples_per_second": 10.793,
"eval_steps_per_second": 5.412,
"step": 10000
},
{
"epoch": 0.94,
"learning_rate": 4.0641702127659574e-05,
"loss": 0.8977,
"step": 11000
},
{
"epoch": 0.94,
"eval_loss": 1.1433526277542114,
"eval_runtime": 32.2243,
"eval_samples_per_second": 10.768,
"eval_steps_per_second": 5.4,
"step": 11000
},
{
"epoch": 1.02,
"learning_rate": 3.979063829787234e-05,
"loss": 0.8627,
"step": 12000
},
{
"epoch": 1.02,
"eval_loss": 1.1658977270126343,
"eval_runtime": 32.8198,
"eval_samples_per_second": 10.573,
"eval_steps_per_second": 5.302,
"step": 12000
},
{
"epoch": 1.11,
"learning_rate": 3.893957446808511e-05,
"loss": 0.7574,
"step": 13000
},
{
"epoch": 1.11,
"eval_loss": 1.1645604372024536,
"eval_runtime": 33.8017,
"eval_samples_per_second": 10.266,
"eval_steps_per_second": 5.148,
"step": 13000
},
{
"epoch": 1.19,
"learning_rate": 3.808936170212766e-05,
"loss": 0.7611,
"step": 14000
},
{
"epoch": 1.19,
"eval_loss": 1.158371925354004,
"eval_runtime": 32.1766,
"eval_samples_per_second": 10.784,
"eval_steps_per_second": 5.408,
"step": 14000
},
{
"epoch": 1.28,
"learning_rate": 3.7238297872340426e-05,
"loss": 0.7333,
"step": 15000
},
{
"epoch": 1.28,
"eval_loss": 1.150078535079956,
"eval_runtime": 31.774,
"eval_samples_per_second": 10.921,
"eval_steps_per_second": 5.476,
"step": 15000
},
{
"epoch": 1.36,
"learning_rate": 3.638723404255319e-05,
"loss": 0.7424,
"step": 16000
},
{
"epoch": 1.36,
"eval_loss": 1.1318830251693726,
"eval_runtime": 30.8015,
"eval_samples_per_second": 11.266,
"eval_steps_per_second": 5.649,
"step": 16000
},
{
"epoch": 1.45,
"learning_rate": 3.5536170212765954e-05,
"loss": 0.7364,
"step": 17000
},
{
"epoch": 1.45,
"eval_loss": 1.1392625570297241,
"eval_runtime": 32.7681,
"eval_samples_per_second": 10.59,
"eval_steps_per_second": 5.31,
"step": 17000
},
{
"epoch": 1.53,
"learning_rate": 3.468595744680851e-05,
"loss": 0.7189,
"step": 18000
},
{
"epoch": 1.53,
"eval_loss": 1.1607253551483154,
"eval_runtime": 31.4151,
"eval_samples_per_second": 11.046,
"eval_steps_per_second": 5.539,
"step": 18000
},
{
"epoch": 1.62,
"learning_rate": 3.383489361702128e-05,
"loss": 0.7131,
"step": 19000
},
{
"epoch": 1.62,
"eval_loss": 1.161035180091858,
"eval_runtime": 32.0907,
"eval_samples_per_second": 10.813,
"eval_steps_per_second": 5.422,
"step": 19000
},
{
"epoch": 1.7,
"learning_rate": 3.298468085106383e-05,
"loss": 0.7012,
"step": 20000
},
{
"epoch": 1.7,
"eval_loss": 1.1332006454467773,
"eval_runtime": 32.2943,
"eval_samples_per_second": 10.745,
"eval_steps_per_second": 5.388,
"step": 20000
}
],
"max_steps": 58750,
"num_train_epochs": 5,
"total_flos": 1.443711128961024e+19,
"trial_name": null,
"trial_params": null
}