Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +92 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 241,
|
| 3 |
+
"best_metric": 1.9624056816101074,
|
| 4 |
+
"best_model_checkpoint": "./mt5_base/indic_loss/pan/checkpoint-241",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 723,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"eval_bleu": 7.8449,
|
| 15 |
+
"eval_gen_len": 19.9686,
|
| 16 |
+
"eval_loss": 1.9624056816101074,
|
| 17 |
+
"eval_runtime": 23.9819,
|
| 18 |
+
"eval_samples_per_second": 23.893,
|
| 19 |
+
"eval_steps_per_second": 3.002,
|
| 20 |
+
"eval_wer": 0.8248,
|
| 21 |
+
"step": 241
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 2.0,
|
| 25 |
+
"eval_bleu": 2.2342,
|
| 26 |
+
"eval_gen_len": 17.2356,
|
| 27 |
+
"eval_loss": 2.161780834197998,
|
| 28 |
+
"eval_runtime": 23.892,
|
| 29 |
+
"eval_samples_per_second": 23.983,
|
| 30 |
+
"eval_steps_per_second": 3.014,
|
| 31 |
+
"eval_wer": 0.8018,
|
| 32 |
+
"step": 482
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.074688796680498,
|
| 36 |
+
"grad_norm": 1.437437653541565,
|
| 37 |
+
"learning_rate": 4.482365145228216e-05,
|
| 38 |
+
"loss": 1.8496,
|
| 39 |
+
"step": 500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 3.0,
|
| 43 |
+
"eval_bleu": 3.4713,
|
| 44 |
+
"eval_gen_len": 19.5602,
|
| 45 |
+
"eval_loss": 2.269686698913574,
|
| 46 |
+
"eval_runtime": 23.8354,
|
| 47 |
+
"eval_samples_per_second": 24.04,
|
| 48 |
+
"eval_steps_per_second": 3.021,
|
| 49 |
+
"eval_wer": 0.7771,
|
| 50 |
+
"step": 723
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"step": 723,
|
| 55 |
+
"total_flos": 1410808919629824.0,
|
| 56 |
+
"train_loss": 1.330777443955713,
|
| 57 |
+
"train_runtime": 254.2669,
|
| 58 |
+
"train_samples_per_second": 151.258,
|
| 59 |
+
"train_steps_per_second": 18.956
|
| 60 |
+
}
|
| 61 |
+
],
|
| 62 |
+
"logging_steps": 500,
|
| 63 |
+
"max_steps": 4820,
|
| 64 |
+
"num_input_tokens_seen": 0,
|
| 65 |
+
"num_train_epochs": 20,
|
| 66 |
+
"save_steps": 50000.0,
|
| 67 |
+
"stateful_callbacks": {
|
| 68 |
+
"EarlyStoppingCallback": {
|
| 69 |
+
"args": {
|
| 70 |
+
"early_stopping_patience": 2,
|
| 71 |
+
"early_stopping_threshold": 0.0
|
| 72 |
+
},
|
| 73 |
+
"attributes": {
|
| 74 |
+
"early_stopping_patience_counter": 2
|
| 75 |
+
}
|
| 76 |
+
},
|
| 77 |
+
"TrainerControl": {
|
| 78 |
+
"args": {
|
| 79 |
+
"should_epoch_stop": false,
|
| 80 |
+
"should_evaluate": false,
|
| 81 |
+
"should_log": false,
|
| 82 |
+
"should_save": true,
|
| 83 |
+
"should_training_stop": true
|
| 84 |
+
},
|
| 85 |
+
"attributes": {}
|
| 86 |
+
}
|
| 87 |
+
},
|
| 88 |
+
"total_flos": 1410808919629824.0,
|
| 89 |
+
"train_batch_size": 8,
|
| 90 |
+
"trial_name": null,
|
| 91 |
+
"trial_params": null
|
| 92 |
+
}
|