End of training

README.md

@@ -3,6 +3,8 @@ license: mit
 base_model: unicamp-dl/ptt5-large-t5-vocab
 tags:
 - generated_from_trainer
+datasets:
+- tiagoblima/qg_squad_v1_pt
 model-index:
 - name: t5_large-qg-ap-nopeft
   results: []
@@ -13,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5_large-qg-ap-nopeft
 
-This model is a fine-tuned version of [unicamp-dl/ptt5-large-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-large-t5-vocab) on an unknown dataset.
+This model is a fine-tuned version of [unicamp-dl/ptt5-large-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-large-t5-vocab) on the tiagoblima/qg_squad_v1_pt dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2048
+- Loss: 1.1994
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 5.0,
+    "eval_loss": 1.1993927955627441,
+    "eval_runtime": 37.8229,
+    "eval_samples": 3585,
+    "eval_samples_per_second": 94.784,
+    "eval_steps_per_second": 5.949,
+    "train_loss": 1.0571302187324751,
+    "train_runtime": 9858.1834,
+    "train_samples": 51704,
+    "train_samples_per_second": 26.224,
+    "train_steps_per_second": 0.41
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "eval_loss": 1.1993927955627441,
+    "eval_runtime": 37.8229,
+    "eval_samples": 3585,
+    "eval_samples_per_second": 94.784,
+    "eval_steps_per_second": 5.949
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 1.0571302187324751,
+    "train_runtime": 9858.1834,
+    "train_samples": 51704,
+    "train_samples_per_second": 26.224,
+    "train_steps_per_second": 0.41
+}

trainer_state.json

@@ -0,0 +1,308 @@
+{
+  "best_metric": 1.1993927955627441,
+  "best_model_checkpoint": "/temp/t5_large-qg-ap-nopeft/checkpoint-2424",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 4040,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.752475247524753e-05,
+      "loss": 1.9922,
+      "step": 100
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.504950495049505e-05,
+      "loss": 1.3295,
+      "step": 200
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.257425742574259e-05,
+      "loss": 1.2696,
+      "step": 300
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.009900990099011e-05,
+      "loss": 1.2353,
+      "step": 400
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 8.762376237623763e-05,
+      "loss": 1.2092,
+      "step": 500
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 8.514851485148515e-05,
+      "loss": 1.1913,
+      "step": 600
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 8.267326732673268e-05,
+      "loss": 1.1831,
+      "step": 700
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 8.019801980198021e-05,
+      "loss": 1.1714,
+      "step": 800
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.2384811639785767,
+      "eval_runtime": 37.8804,
+      "eval_samples_per_second": 94.64,
+      "eval_steps_per_second": 5.94,
+      "step": 808
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 7.772277227722773e-05,
+      "loss": 1.0963,
+      "step": 900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 7.524752475247526e-05,
+      "loss": 1.0913,
+      "step": 1000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7.277227722772278e-05,
+      "loss": 1.0815,
+      "step": 1100
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.02970297029703e-05,
+      "loss": 1.0873,
+      "step": 1200
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 6.782178217821783e-05,
+      "loss": 1.0841,
+      "step": 1300
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 6.534653465346535e-05,
+      "loss": 1.0671,
+      "step": 1400
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.287128712871287e-05,
+      "loss": 1.0734,
+      "step": 1500
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 6.03960396039604e-05,
+      "loss": 1.0782,
+      "step": 1600
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.2065564393997192,
+      "eval_runtime": 37.776,
+      "eval_samples_per_second": 94.901,
+      "eval_steps_per_second": 5.956,
+      "step": 1616
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 5.792079207920792e-05,
+      "loss": 1.0134,
+      "step": 1700
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 5.544554455445545e-05,
+      "loss": 1.0127,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 5.2970297029702974e-05,
+      "loss": 0.9995,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 5.0495049504950497e-05,
+      "loss": 1.004,
+      "step": 2000
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.801980198019802e-05,
+      "loss": 1.0019,
+      "step": 2100
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.554455445544555e-05,
+      "loss": 0.9972,
+      "step": 2200
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 4.306930693069307e-05,
+      "loss": 1.0135,
+      "step": 2300
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.05940594059406e-05,
+      "loss": 1.0064,
+      "step": 2400
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.1993927955627441,
+      "eval_runtime": 37.796,
+      "eval_samples_per_second": 94.851,
+      "eval_steps_per_second": 5.953,
+      "step": 2424
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 3.811881188118812e-05,
+      "loss": 0.9637,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 3.5643564356435645e-05,
+      "loss": 0.9496,
+      "step": 2600
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 3.3168316831683175e-05,
+      "loss": 0.9621,
+      "step": 2700
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 3.06930693069307e-05,
+      "loss": 0.9592,
+      "step": 2800
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 2.8217821782178216e-05,
+      "loss": 0.9571,
+      "step": 2900
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 2.5742574257425746e-05,
+      "loss": 0.94,
+      "step": 3000
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 2.326732673267327e-05,
+      "loss": 0.9553,
+      "step": 3100
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 2.079207920792079e-05,
+      "loss": 0.9555,
+      "step": 3200
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.2030982971191406,
+      "eval_runtime": 37.7841,
+      "eval_samples_per_second": 94.881,
+      "eval_steps_per_second": 5.955,
+      "step": 3232
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 1.8316831683168317e-05,
+      "loss": 0.939,
+      "step": 3300
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 1.5841584158415843e-05,
+      "loss": 0.9216,
+      "step": 3400
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 1.3366336633663367e-05,
+      "loss": 0.9303,
+      "step": 3500
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 1.0891089108910891e-05,
+      "loss": 0.9122,
+      "step": 3600
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 8.415841584158417e-06,
+      "loss": 0.9262,
+      "step": 3700
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 5.940594059405941e-06,
+      "loss": 0.9135,
+      "step": 3800
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 3.4653465346534657e-06,
+      "loss": 0.929,
+      "step": 3900
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 9.900990099009902e-07,
+      "loss": 0.9261,
+      "step": 4000
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.2047841548919678,
+      "eval_runtime": 37.8264,
+      "eval_samples_per_second": 94.775,
+      "eval_steps_per_second": 5.948,
+      "step": 4040
+    },
+    {
+      "epoch": 5.0,
+      "step": 4040,
+      "total_flos": 5.5970802696192e+17,
+      "train_loss": 1.0571302187324751,
+      "train_runtime": 9858.1834,
+      "train_samples_per_second": 26.224,
+      "train_steps_per_second": 0.41
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 4040,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 5.5970802696192e+17,
+  "trial_name": null,
+  "trial_params": null
+}