Upload trainer_state.json with huggingface_hub

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_global_step": 8271,
+  "best_metric": 0.8720680522378279,
+  "best_model_checkpoint": "outputs/hate-speech-detection/vihate-t5/checkpoint-8271",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 8271,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.998911860718172,
+      "grad_norm": 4.305943965911865,
+      "learning_rate": 1.9996091589167148e-05,
+      "loss": 0.4014,
+      "step": 918
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8637209302325581,
+      "eval_f1": 0.8576683113718502,
+      "eval_loss": 0.3640185594558716,
+      "eval_precision": 0.8555800625765106,
+      "eval_recall": 0.8637209302325581,
+      "eval_runtime": 21.0759,
+      "eval_samples_per_second": 204.025,
+      "eval_steps_per_second": 6.405,
+      "step": 919
+    },
+    {
+      "epoch": 1.997823721436344,
+      "grad_norm": 5.548094272613525,
+      "learning_rate": 1.9982377977480694e-05,
+      "loss": 0.2814,
+      "step": 1836
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8716279069767442,
+      "eval_f1": 0.8649693053637155,
+      "eval_loss": 0.3619186580181122,
+      "eval_precision": 0.8605904079045565,
+      "eval_recall": 0.8716279069767442,
+      "eval_runtime": 21.0838,
+      "eval_samples_per_second": 203.948,
+      "eval_steps_per_second": 6.403,
+      "step": 1838
+    },
+    {
+      "epoch": 2.996735582154516,
+      "grad_norm": 8.380730628967285,
+      "learning_rate": 1.9958812963917564e-05,
+      "loss": 0.191,
+      "step": 2754
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.87,
+      "eval_f1": 0.8628937123739281,
+      "eval_loss": 0.4661368429660797,
+      "eval_precision": 0.8588361297253848,
+      "eval_recall": 0.87,
+      "eval_runtime": 21.0807,
+      "eval_samples_per_second": 203.978,
+      "eval_steps_per_second": 6.404,
+      "step": 2757
+    },
+    {
+      "epoch": 3.995647442872688,
+      "grad_norm": 7.982128143310547,
+      "learning_rate": 1.992541980430111e-05,
+      "loss": 0.1179,
+      "step": 3672
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8588372093023255,
+      "eval_f1": 0.8598944782360041,
+      "eval_loss": 0.555857241153717,
+      "eval_precision": 0.8611954815216634,
+      "eval_recall": 0.8588372093023255,
+      "eval_runtime": 21.0689,
+      "eval_samples_per_second": 204.092,
+      "eval_steps_per_second": 6.408,
+      "step": 3676
+    },
+    {
+      "epoch": 4.99455930359086,
+      "grad_norm": 18.193098068237305,
+      "learning_rate": 1.9882231453648264e-05,
+      "loss": 0.0823,
+      "step": 4590
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8693023255813953,
+      "eval_f1": 0.8623605411584486,
+      "eval_loss": 0.8255857229232788,
+      "eval_precision": 0.8584079420364616,
+      "eval_recall": 0.8693023255813953,
+      "eval_runtime": 21.1139,
+      "eval_samples_per_second": 203.657,
+      "eval_steps_per_second": 6.394,
+      "step": 4595
+    },
+    {
+      "epoch": 5.993471164309032,
+      "grad_norm": 18.74983787536621,
+      "learning_rate": 1.982929053364693e-05,
+      "loss": 0.0574,
+      "step": 5508
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8744186046511628,
+      "eval_f1": 0.8661137464376901,
+      "eval_loss": 1.1137067079544067,
+      "eval_precision": 0.8612640241302489,
+      "eval_recall": 0.8744186046511628,
+      "eval_runtime": 21.0981,
+      "eval_samples_per_second": 203.81,
+      "eval_steps_per_second": 6.399,
+      "step": 5514
+    },
+    {
+      "epoch": 6.992383025027204,
+      "grad_norm": 26.94858169555664,
+      "learning_rate": 1.9766649290593513e-05,
+      "loss": 0.0508,
+      "step": 6426
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8586046511627907,
+      "eval_f1": 0.8607428207518112,
+      "eval_loss": 1.1956874132156372,
+      "eval_precision": 0.8631659432487684,
+      "eval_recall": 0.8586046511627907,
+      "eval_runtime": 21.1012,
+      "eval_samples_per_second": 203.78,
+      "eval_steps_per_second": 6.398,
+      "step": 6433
+    },
+    {
+      "epoch": 7.991294885745376,
+      "grad_norm": 0.0075432900339365005,
+      "learning_rate": 1.9694369543832138e-05,
+      "loss": 0.0413,
+      "step": 7344
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8758139534883721,
+      "eval_f1": 0.8665994257720897,
+      "eval_loss": 1.3685424327850342,
+      "eval_precision": 0.8617321415270166,
+      "eval_recall": 0.8758139534883721,
+      "eval_runtime": 21.0755,
+      "eval_samples_per_second": 204.029,
+      "eval_steps_per_second": 6.406,
+      "step": 7352
+    },
+    {
+      "epoch": 8.990206746463548,
+      "grad_norm": 0.007036261726170778,
+      "learning_rate": 1.9612522624746426e-05,
+      "loss": 0.0345,
+      "step": 8262
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.881860465116279,
+      "eval_f1": 0.8720680522378279,
+      "eval_loss": 1.4441194534301758,
+      "eval_precision": 0.8677296023078702,
+      "eval_recall": 0.881860465116279,
+      "eval_runtime": 21.0711,
+      "eval_samples_per_second": 204.071,
+      "eval_steps_per_second": 6.407,
+      "step": 8271
+    }
+  ],
+  "logging_steps": 918,
+  "max_steps": 91900,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 100,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}