End of training

README.md

@@ -3,6 +3,8 @@ license: mit
 base_model: unicamp-dl/ptt5-small-t5-vocab
 tags:
 - generated_from_trainer
+datasets:
+- tiagoblima/qg_squad_v1_pt
 model-index:
 - name: t5_small-qg-aap
   results: []
@@ -13,7 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5_small-qg-aap
 
-This model is a fine-tuned version of [unicamp-dl/ptt5-small-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-small-t5-vocab) on an unknown dataset.
+This model is a fine-tuned version of [unicamp-dl/ptt5-small-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-small-t5-vocab) on the tiagoblima/qg_squad_v1_pt dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.4213
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 10.0,
-    "eval_loss": 1.4470528364181519,
-    "eval_runtime": 77.4953,
+    "epoch": 15.0,
+    "eval_loss": 1.4212826490402222,
+    "eval_runtime": 78.3846,
     "eval_samples": 8869,
-    "eval_samples_per_second": 114.446,
-    "eval_steps_per_second": 14.311,
-    "train_loss": 1.2936539687732658,
-    "train_runtime": 11629.1597,
+    "eval_samples_per_second": 113.147,
+    "eval_steps_per_second": 14.148,
+    "train_loss": 1.1541598279090604,
+    "train_runtime": 16985.3317,
     "train_samples": 51704,
-    "train_samples_per_second": 44.461,
-    "train_steps_per_second": 0.695
+    "train_samples_per_second": 45.661,
+    "train_steps_per_second": 0.714
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "eval_loss": 1.4470528364181519,
-    "eval_runtime": 77.4953,
+    "epoch": 15.0,
+    "eval_loss": 1.4212826490402222,
+    "eval_runtime": 78.3846,
     "eval_samples": 8869,
-    "eval_samples_per_second": 114.446,
-    "eval_steps_per_second": 14.311
+    "eval_samples_per_second": 113.147,
+    "eval_steps_per_second": 14.148
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 1.2936539687732658,
-    "train_runtime": 11629.1597,
+    "epoch": 15.0,
+    "train_loss": 1.1541598279090604,
+    "train_runtime": 16985.3317,
     "train_samples": 51704,
-    "train_samples_per_second": 44.461,
-    "train_steps_per_second": 0.695
+    "train_samples_per_second": 45.661,
+    "train_steps_per_second": 0.714
 }

trainer_state.json

@@ -1,124 +1,172 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 15.0,
   "eval_steps": 500,
-  "global_step": 8080,
+  "global_step": 12120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.62,
-      "learning_rate": 9.381188118811881e-05,
-      "loss": 1.7918,
+      "learning_rate": 9.690594059405941e-05,
+      "loss": 1.7898,
       "step": 500
     },
     {
       "epoch": 1.24,
-      "learning_rate": 8.762376237623763e-05,
-      "loss": 1.4547,
+      "learning_rate": 9.174917491749175e-05,
+      "loss": 1.4524,
       "step": 1000
     },
     {
       "epoch": 1.86,
-      "learning_rate": 8.143564356435644e-05,
-      "loss": 1.3849,
+      "learning_rate": 8.762376237623763e-05,
+      "loss": 1.3819,
       "step": 1500
     },
     {
       "epoch": 2.48,
-      "learning_rate": 7.524752475247526e-05,
-      "loss": 1.3378,
+      "learning_rate": 8.34983498349835e-05,
+      "loss": 1.3341,
       "step": 2000
     },
     {
       "epoch": 3.09,
-      "learning_rate": 6.905940594059406e-05,
-      "loss": 1.3125,
+      "learning_rate": 7.937293729372938e-05,
+      "loss": 1.3073,
       "step": 2500
     },
     {
       "epoch": 3.71,
-      "learning_rate": 6.287128712871287e-05,
-      "loss": 1.2841,
+      "learning_rate": 7.524752475247526e-05,
+      "loss": 1.2771,
       "step": 3000
     },
     {
       "epoch": 4.33,
-      "learning_rate": 5.668316831683168e-05,
-      "loss": 1.2655,
+      "learning_rate": 7.112211221122112e-05,
+      "loss": 1.2567,
       "step": 3500
     },
     {
       "epoch": 4.95,
-      "learning_rate": 5.0495049504950497e-05,
-      "loss": 1.2478,
+      "learning_rate": 6.6996699669967e-05,
+      "loss": 1.2378,
       "step": 4000
     },
     {
       "epoch": 5.57,
-      "learning_rate": 4.430693069306931e-05,
-      "loss": 1.2333,
+      "learning_rate": 6.287128712871287e-05,
+      "loss": 1.2201,
       "step": 4500
     },
     {
       "epoch": 6.19,
-      "learning_rate": 3.811881188118812e-05,
-      "loss": 1.2248,
+      "learning_rate": 5.874587458745875e-05,
+      "loss": 1.2095,
       "step": 5000
     },
     {
       "epoch": 6.81,
-      "learning_rate": 3.1930693069306936e-05,
-      "loss": 1.2144,
+      "learning_rate": 5.462046204620462e-05,
+      "loss": 1.1963,
       "step": 5500
     },
     {
       "epoch": 7.43,
-      "learning_rate": 2.5742574257425746e-05,
-      "loss": 1.1982,
+      "learning_rate": 5.0495049504950497e-05,
+      "loss": 1.1768,
       "step": 6000
     },
     {
       "epoch": 8.04,
-      "learning_rate": 1.9554455445544556e-05,
-      "loss": 1.2003,
+      "learning_rate": 4.636963696369637e-05,
+      "loss": 1.1763,
       "step": 6500
     },
     {
       "epoch": 8.66,
-      "learning_rate": 1.3366336633663367e-05,
-      "loss": 1.19,
+      "learning_rate": 4.224422442244225e-05,
+      "loss": 1.1606,
       "step": 7000
     },
     {
       "epoch": 9.28,
-      "learning_rate": 7.178217821782178e-06,
-      "loss": 1.1888,
+      "learning_rate": 3.811881188118812e-05,
+      "loss": 1.156,
       "step": 7500
     },
     {
       "epoch": 9.9,
-      "learning_rate": 9.900990099009902e-07,
-      "loss": 1.1881,
+      "learning_rate": 3.3993399339933996e-05,
+      "loss": 1.1516,
       "step": 8000
     },
     {
-      "epoch": 10.0,
-      "step": 8080,
-      "total_flos": 5.248284377481216e+16,
-      "train_loss": 1.2936539687732658,
-      "train_runtime": 11629.1597,
-      "train_samples_per_second": 44.461,
-      "train_steps_per_second": 0.695
+      "epoch": 10.52,
+      "learning_rate": 2.986798679867987e-05,
+      "loss": 1.1371,
+      "step": 8500
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 2.5742574257425746e-05,
+      "loss": 1.1417,
+      "step": 9000
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 2.161716171617162e-05,
+      "loss": 1.1285,
+      "step": 9500
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 1.7491749174917492e-05,
+      "loss": 1.1302,
+      "step": 10000
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.3366336633663367e-05,
+      "loss": 1.1223,
+      "step": 10500
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 9.24092409240924e-06,
+      "loss": 1.1192,
+      "step": 11000
+    },
+    {
+      "epoch": 14.23,
+      "learning_rate": 5.115511551155116e-06,
+      "loss": 1.122,
+      "step": 11500
+    },
+    {
+      "epoch": 14.85,
+      "learning_rate": 9.900990099009902e-07,
+      "loss": 1.1116,
+      "step": 12000
+    },
+    {
+      "epoch": 15.0,
+      "step": 12120,
+      "total_flos": 7.872426566221824e+16,
+      "train_loss": 1.1541598279090604,
+      "train_runtime": 16985.3317,
+      "train_samples_per_second": 45.661,
+      "train_steps_per_second": 0.714
     }
   ],
   "logging_steps": 500,
-  "max_steps": 8080,
-  "num_train_epochs": 10,
+  "max_steps": 12120,
+  "num_train_epochs": 15,
   "save_steps": 500,
-  "total_flos": 5.248284377481216e+16,
+  "total_flos": 7.872426566221824e+16,
   "trial_name": null,
   "trial_params": null
 }