longformer-spans / checkpoint-405 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 5, checkpoint
94f859d verified
raw
history blame
6.12 kB
{
"best_metric": 0.2252955436706543,
"best_model_checkpoint": "longformer-spans/checkpoint-162",
"epoch": 5.0,
"eval_steps": 500,
"global_step": 405,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_B": {
"f1-score": 0.816,
"precision": 0.7461594732991953,
"recall": 0.9002647837599294,
"support": 1133.0
},
"eval_I": {
"f1-score": 0.9321376763813793,
"precision": 0.9024103768767235,
"recall": 0.9638902525500463,
"support": 18333.0
},
"eval_O": {
"f1-score": 0.8560784528570645,
"precision": 0.931782945736434,
"recall": 0.7917511147142278,
"support": 9868.0
},
"eval_accuracy": 0.9035249198881844,
"eval_loss": 0.2619660496711731,
"eval_macro avg": {
"f1-score": 0.8680720430794812,
"precision": 0.860117598637451,
"recall": 0.8853020503414012,
"support": 29334.0
},
"eval_runtime": 5.5032,
"eval_samples_per_second": 14.537,
"eval_steps_per_second": 1.817,
"eval_weighted avg": {
"f1-score": 0.9020655278480035,
"precision": 0.9062562975065145,
"recall": 0.9035249198881844,
"support": 29334.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_B": {
"f1-score": 0.8512256973795435,
"precision": 0.8167072181670721,
"recall": 0.8887908208296558,
"support": 1133.0
},
"eval_I": {
"f1-score": 0.9416781438711729,
"precision": 0.9152551099212274,
"recall": 0.9696721758577429,
"support": 18333.0
},
"eval_O": {
"f1-score": 0.8778173190984578,
"precision": 0.9380041484212952,
"recall": 0.8248885285772193,
"support": 9868.0
},
"eval_accuracy": 0.9178427763005387,
"eval_loss": 0.2252955436706543,
"eval_macro avg": {
"f1-score": 0.8902403867830581,
"precision": 0.8899888255031981,
"recall": 0.8944505084215394,
"support": 29334.0
},
"eval_runtime": 5.4972,
"eval_samples_per_second": 14.553,
"eval_steps_per_second": 1.819,
"eval_weighted avg": {
"f1-score": 0.9167016237671239,
"precision": 0.9191015935430046,
"recall": 0.9178427763005387,
"support": 29334.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_B": {
"f1-score": 0.8531120331950207,
"precision": 0.8050117462803446,
"recall": 0.9073256840247131,
"support": 1133.0
},
"eval_I": {
"f1-score": 0.9469915571230095,
"precision": 0.9280963603037444,
"recall": 0.9666721213112965,
"support": 18333.0
},
"eval_O": {
"f1-score": 0.8903876792352629,
"precision": 0.9353938852934612,
"recall": 0.8495135792460479,
"support": 9868.0
},
"eval_accuracy": 0.9249676143724006,
"eval_loss": 0.22786012291908264,
"eval_macro avg": {
"f1-score": 0.8968304231844311,
"precision": 0.8895006639591835,
"recall": 0.9078371281940192,
"support": 29334.0
},
"eval_runtime": 5.5036,
"eval_samples_per_second": 14.536,
"eval_steps_per_second": 1.817,
"eval_weighted avg": {
"f1-score": 0.9243239165827936,
"precision": 0.9257972230878861,
"recall": 0.9249676143724006,
"support": 29334.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_B": {
"f1-score": 0.8567807351077312,
"precision": 0.8217179902755267,
"recall": 0.8949691085613416,
"support": 1133.0
},
"eval_I": {
"f1-score": 0.9472597903427299,
"precision": 0.9432635621180161,
"recall": 0.9512900234549719,
"support": 18333.0
},
"eval_O": {
"f1-score": 0.8979927100980543,
"precision": 0.9099989595255437,
"recall": 0.8862991487636805,
"support": 9868.0
},
"eval_accuracy": 0.9272516533715143,
"eval_loss": 0.23897655308246613,
"eval_macro avg": {
"f1-score": 0.9006777451828384,
"precision": 0.8916601706396955,
"recall": 0.910852760259998,
"support": 29334.0
},
"eval_runtime": 5.4904,
"eval_samples_per_second": 14.571,
"eval_steps_per_second": 1.821,
"eval_weighted avg": {
"f1-score": 0.9271915992526735,
"precision": 0.9273787107073643,
"recall": 0.9272516533715143,
"support": 29334.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_B": {
"f1-score": 0.8624407072013798,
"precision": 0.8431703204047217,
"recall": 0.8826125330979699,
"support": 1133.0
},
"eval_I": {
"f1-score": 0.948190701170407,
"precision": 0.9335059992600032,
"recall": 0.9633447880870561,
"support": 18333.0
},
"eval_O": {
"f1-score": 0.8955333298423835,
"precision": 0.9265359193845487,
"recall": 0.8665383056343737,
"support": 9868.0
},
"eval_accuracy": 0.9276607349832958,
"eval_loss": 0.25390708446502686,
"eval_macro avg": {
"f1-score": 0.9020549127380568,
"precision": 0.9010707463497579,
"recall": 0.9041652089397999,
"support": 29334.0
},
"eval_runtime": 5.4919,
"eval_samples_per_second": 14.567,
"eval_steps_per_second": 1.821,
"eval_weighted avg": {
"f1-score": 0.9271646670996412,
"precision": 0.9276721180179627,
"recall": 0.9276607349832958,
"support": 29334.0
},
"step": 405
}
],
"logging_steps": 500,
"max_steps": 1620,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 500,
"total_flos": 718933263678000.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}