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-aap
   results: []
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5_large-qg-aap
 
-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: 7.4208
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 9.98930835723877,
-    "eval_runtime": 1084.569,
+    "epoch": 5.0,
+    "eval_loss": 7.420807361602783,
+    "eval_runtime": 587.2121,
     "eval_samples": 8869,
-    "eval_samples_per_second": 8.177,
-    "eval_steps_per_second": 4.089,
-    "train_loss": 8.507731522663985,
-    "train_runtime": 5914.0661,
+    "eval_samples_per_second": 15.104,
+    "eval_steps_per_second": 7.553,
+    "train_loss": 6.77048750395822,
+    "train_runtime": 16676.4775,
     "train_samples": 51704,
-    "train_samples_per_second": 8.743,
-    "train_steps_per_second": 0.068
+    "train_samples_per_second": 15.502,
+    "train_steps_per_second": 0.121
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 9.98930835723877,
-    "eval_runtime": 1084.569,
+    "epoch": 5.0,
+    "eval_loss": 7.420807361602783,
+    "eval_runtime": 587.2121,
     "eval_samples": 8869,
-    "eval_samples_per_second": 8.177,
-    "eval_steps_per_second": 4.089
+    "eval_samples_per_second": 15.104,
+    "eval_steps_per_second": 7.553
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "train_loss": 8.507731522663985,
-    "train_runtime": 5914.0661,
+    "epoch": 5.0,
+    "train_loss": 6.77048750395822,
+    "train_runtime": 16676.4775,
     "train_samples": 51704,
-    "train_samples_per_second": 8.743,
-    "train_steps_per_second": 0.068
+    "train_samples_per_second": 15.502,
+    "train_steps_per_second": 0.121
 }

trainer_state.json

@@ -1,30 +1,188 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 5.0,
   "eval_steps": 500,
-  "global_step": 404,
+  "global_step": 2020,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0028514851485148514,
+      "loss": 8.3611,
+      "step": 100
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.002702970297029703,
+      "loss": 8.0607,
+      "step": 200
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.0025544554455445546,
+      "loss": 7.7784,
+      "step": 300
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.0024059405940594063,
+      "loss": 7.5413,
+      "step": 400
+    },
     {
       "epoch": 1.0,
-      "step": 404,
-      "total_flos": 8.815639295478989e+16,
-      "train_loss": 8.507731522663985,
-      "train_runtime": 5914.0661,
-      "train_samples_per_second": 8.743,
-      "train_steps_per_second": 0.068
+      "eval_loss": 8.850217819213867,
+      "eval_runtime": 584.1219,
+      "eval_samples_per_second": 15.183,
+      "eval_steps_per_second": 7.593,
+      "step": 404
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0022574257425742577,
+      "loss": 7.3312,
+      "step": 500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.0021089108910891086,
+      "loss": 7.1317,
+      "step": 600
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.0019603960396039604,
+      "loss": 6.9568,
+      "step": 700
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.0018118811881188118,
+      "loss": 6.7965,
+      "step": 800
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 8.118351936340332,
+      "eval_runtime": 586.5561,
+      "eval_samples_per_second": 15.12,
+      "eval_steps_per_second": 7.561,
+      "step": 808
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.0016633663366336635,
+      "loss": 6.6621,
+      "step": 900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.001514851485148515,
+      "loss": 6.5571,
+      "step": 1000
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.0013663366336633665,
+      "loss": 6.4736,
+      "step": 1100
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 0.001217821782178218,
+      "loss": 6.3963,
+      "step": 1200
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 7.695041179656982,
+      "eval_runtime": 586.9195,
+      "eval_samples_per_second": 15.111,
+      "eval_steps_per_second": 7.556,
+      "step": 1212
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 0.0010693069306930692,
+      "loss": 6.3235,
+      "step": 1300
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 0.0009207920792079207,
+      "loss": 6.283,
+      "step": 1400
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 0.0007722772277227723,
+      "loss": 6.2224,
+      "step": 1500
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 0.0006237623762376238,
+      "loss": 6.1664,
+      "step": 1600
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 7.485514163970947,
+      "eval_runtime": 587.006,
+      "eval_samples_per_second": 15.109,
+      "eval_steps_per_second": 7.555,
+      "step": 1616
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 0.0004752475247524753,
+      "loss": 6.1536,
+      "step": 1700
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 0.00032673267326732675,
+      "loss": 6.1402,
+      "step": 1800
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 0.0001782178217821782,
+      "loss": 6.0973,
+      "step": 1900
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 2.9702970297029706e-05,
+      "loss": 6.1028,
+      "step": 2000
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 7.420807361602783,
+      "eval_runtime": 587.2049,
+      "eval_samples_per_second": 15.104,
+      "eval_steps_per_second": 7.553,
+      "step": 2020
+    },
+    {
+      "epoch": 5.0,
+      "step": 2020,
+      "total_flos": 4.4078196477394944e+17,
+      "train_loss": 6.77048750395822,
+      "train_runtime": 16676.4775,
+      "train_samples_per_second": 15.502,
+      "train_steps_per_second": 0.121
     }
   ],
-  "logging_steps": 500,
-  "max_steps": 404,
-  "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "logging_steps": 100,
+  "max_steps": 2020,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 8.815639295478989e+16,
-  "train_batch_size": 128,
+  "total_flos": 4.4078196477394944e+17,
   "trial_name": null,
   "trial_params": null
 }