End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- id
 license: apache-2.0
 base_model: LazarusNLP/IndoNanoT5-base
 tags:
@@ -9,7 +11,20 @@ metrics:
 - rouge
 model-index:
 - name: liputan6-pt-pl5
-  results: []
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: id_liputan6 canonical
+      type: id_liputan6
+      config: canonical
+      split: validation
+      args: canonical
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 29.4984
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -17,14 +32,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # liputan6-pt-pl5
 
-This model is a fine-tuned version of [LazarusNLP/IndoNanoT5-base](https://huggingface.co/LazarusNLP/IndoNanoT5-base) on the id_liputan6 dataset.
+This model is a fine-tuned version of [LazarusNLP/IndoNanoT5-base](https://huggingface.co/LazarusNLP/IndoNanoT5-base) on the id_liputan6 canonical dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.7526
-- Rouge1: 29.5518
-- Rouge2: 13.5408
-- Rougel: 24.834
-- Rougelsum: 27.6096
-- Gen Len: 51.956
+- Rouge1: 29.4984
+- Rouge2: 14.2115
+- Rougel: 24.9259
+- Rougelsum: 27.5126
+- Gen Len: 59.201
 
 ## Model description
 

all_results.json

@@ -1,29 +1,29 @@
 {
     "epoch": 5.0,
-    "eval_gen_len": 35.5,
-    "eval_loss": 3.820514440536499,
-    "eval_rouge1": 18.3412,
-    "eval_rouge2": 4.7361,
-    "eval_rougeL": 15.5136,
-    "eval_rougeLsum": 16.6913,
-    "eval_runtime": 1715.0696,
+    "eval_gen_len": 59.201,
+    "eval_loss": 2.752587080001831,
+    "eval_rouge1": 29.4984,
+    "eval_rouge2": 14.2115,
+    "eval_rougeL": 24.9259,
+    "eval_rougeLsum": 27.5126,
+    "eval_runtime": 2095.5163,
     "eval_samples": 1000,
-    "eval_samples_per_second": 0.583,
-    "eval_steps_per_second": 0.019,
-    "predict_gen_len": 35.234,
-    "predict_loss": 3.543308973312378,
-    "predict_rouge1": 21.9696,
-    "predict_rouge2": 6.2524,
-    "predict_rougeL": 18.1263,
-    "predict_rougeLsum": 20.1408,
-    "predict_runtime": 1735.6386,
+    "eval_samples_per_second": 0.477,
+    "eval_steps_per_second": 0.015,
+    "predict_gen_len": 57.369,
+    "predict_loss": 2.604276180267334,
+    "predict_rouge1": 33.6245,
+    "predict_rouge2": 17.4972,
+    "predict_rougeL": 29.0909,
+    "predict_rougeLsum": 31.7664,
+    "predict_runtime": 2090.3191,
     "predict_samples": 1000,
-    "predict_samples_per_second": 0.576,
-    "predict_steps_per_second": 0.018,
+    "predict_samples_per_second": 0.478,
+    "predict_steps_per_second": 0.015,
     "total_flos": 3877644533760000.0,
-    "train_loss": 4.081830463712177,
-    "train_runtime": 1435.3865,
+    "train_loss": 3.4973431299603175,
+    "train_runtime": 1609.8772,
     "train_samples": 1000,
-    "train_samples_per_second": 3.483,
-    "train_steps_per_second": 0.219
+    "train_samples_per_second": 3.106,
+    "train_steps_per_second": 0.196
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
-    "eval_gen_len": 35.5,
-    "eval_loss": 3.820514440536499,
-    "eval_rouge1": 18.3412,
-    "eval_rouge2": 4.7361,
-    "eval_rougeL": 15.5136,
-    "eval_rougeLsum": 16.6913,
-    "eval_runtime": 1715.0696,
+    "eval_gen_len": 59.201,
+    "eval_loss": 2.752587080001831,
+    "eval_rouge1": 29.4984,
+    "eval_rouge2": 14.2115,
+    "eval_rougeL": 24.9259,
+    "eval_rougeLsum": 27.5126,
+    "eval_runtime": 2095.5163,
     "eval_samples": 1000,
-    "eval_samples_per_second": 0.583,
-    "eval_steps_per_second": 0.019
+    "eval_samples_per_second": 0.477,
+    "eval_steps_per_second": 0.015
 }

predict_results.json

@@ -1,12 +1,12 @@
 {
-    "predict_gen_len": 35.234,
-    "predict_loss": 3.543308973312378,
-    "predict_rouge1": 21.9696,
-    "predict_rouge2": 6.2524,
-    "predict_rougeL": 18.1263,
-    "predict_rougeLsum": 20.1408,
-    "predict_runtime": 1735.6386,
+    "predict_gen_len": 57.369,
+    "predict_loss": 2.604276180267334,
+    "predict_rouge1": 33.6245,
+    "predict_rouge2": 17.4972,
+    "predict_rougeL": 29.0909,
+    "predict_rougeLsum": 31.7664,
+    "predict_runtime": 2090.3191,
     "predict_samples": 1000,
-    "predict_samples_per_second": 0.576,
-    "predict_steps_per_second": 0.018
+    "predict_samples_per_second": 0.478,
+    "predict_steps_per_second": 0.015
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 5.0,
     "total_flos": 3877644533760000.0,
-    "train_loss": 4.081830463712177,
-    "train_runtime": 1435.3865,
+    "train_loss": 3.4973431299603175,
+    "train_runtime": 1609.8772,
     "train_samples": 1000,
-    "train_samples_per_second": 3.483,
-    "train_steps_per_second": 0.219
+    "train_samples_per_second": 3.106,
+    "train_steps_per_second": 0.196
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 18.6328,
+  "best_metric": 29.5518,
   "best_model_checkpoint": "bin/liputan6-pt-pl5/checkpoint-315",
   "epoch": 5.0,
   "eval_steps": 500,
@@ -10,112 +10,112 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.9840264320373535,
+      "grad_norm": 0.8308019638061523,
       "learning_rate": 0.0008,
-      "loss": 4.799,
+      "loss": 4.309,
       "step": 63
     },
     {
       "epoch": 1.0,
-      "eval_gen_len": 40.873,
-      "eval_loss": 4.114248275756836,
-      "eval_rouge1": 13.0788,
-      "eval_rouge2": 2.2394,
-      "eval_rougeL": 10.8409,
-      "eval_rougeLsum": 11.8062,
-      "eval_runtime": 296.6779,
-      "eval_samples_per_second": 3.371,
-      "eval_steps_per_second": 0.108,
+      "eval_gen_len": 41.853,
+      "eval_loss": 3.4071803092956543,
+      "eval_rouge1": 20.861,
+      "eval_rouge2": 8.8494,
+      "eval_rougeL": 18.3118,
+      "eval_rougeLsum": 19.3333,
+      "eval_runtime": 253.5381,
+      "eval_samples_per_second": 3.944,
+      "eval_steps_per_second": 0.126,
       "step": 63
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.9562803506851196,
+      "grad_norm": 0.9169403910636902,
       "learning_rate": 0.0006,
-      "loss": 4.179,
+      "loss": 3.6184,
       "step": 126
     },
     {
       "epoch": 2.0,
-      "eval_gen_len": 32.962,
-      "eval_loss": 3.9927873611450195,
-      "eval_rouge1": 16.7604,
-      "eval_rouge2": 3.2541,
-      "eval_rougeL": 13.8889,
-      "eval_rougeLsum": 15.1654,
-      "eval_runtime": 249.03,
-      "eval_samples_per_second": 4.016,
-      "eval_steps_per_second": 0.128,
+      "eval_gen_len": 51.44,
+      "eval_loss": 3.1459953784942627,
+      "eval_rouge1": 25.8149,
+      "eval_rouge2": 9.9065,
+      "eval_rougeL": 21.2242,
+      "eval_rougeLsum": 23.6534,
+      "eval_runtime": 238.2583,
+      "eval_samples_per_second": 4.197,
+      "eval_steps_per_second": 0.134,
       "step": 126
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.9820513725280762,
+      "grad_norm": 0.7842546105384827,
       "learning_rate": 0.0004,
-      "loss": 3.9656,
+      "loss": 3.3773,
       "step": 189
     },
     {
       "epoch": 3.0,
-      "eval_gen_len": 30.549,
-      "eval_loss": 3.883206367492676,
-      "eval_rouge1": 18.1366,
-      "eval_rouge2": 3.9918,
-      "eval_rougeL": 15.2392,
-      "eval_rougeLsum": 16.4266,
-      "eval_runtime": 192.373,
-      "eval_samples_per_second": 5.198,
-      "eval_steps_per_second": 0.166,
+      "eval_gen_len": 55.215,
+      "eval_loss": 2.9716238975524902,
+      "eval_rouge1": 27.0263,
+      "eval_rouge2": 11.563,
+      "eval_rougeL": 22.6018,
+      "eval_rougeLsum": 25.0075,
+      "eval_runtime": 260.5928,
+      "eval_samples_per_second": 3.837,
+      "eval_steps_per_second": 0.123,
       "step": 189
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.9895921945571899,
+      "grad_norm": 1.033412218093872,
       "learning_rate": 0.0002,
-      "loss": 3.8038,
+      "loss": 3.1793,
       "step": 252
     },
     {
       "epoch": 4.0,
-      "eval_gen_len": 28.411,
-      "eval_loss": 3.855151414871216,
-      "eval_rouge1": 18.2504,
-      "eval_rouge2": 4.0948,
-      "eval_rougeL": 15.4777,
-      "eval_rougeLsum": 16.7374,
-      "eval_runtime": 138.9665,
-      "eval_samples_per_second": 7.196,
-      "eval_steps_per_second": 0.23,
+      "eval_gen_len": 52.119,
+      "eval_loss": 2.826566219329834,
+      "eval_rouge1": 28.6871,
+      "eval_rouge2": 12.9802,
+      "eval_rougeL": 24.038,
+      "eval_rougeLsum": 26.6558,
+      "eval_runtime": 245.0999,
+      "eval_samples_per_second": 4.08,
+      "eval_steps_per_second": 0.131,
       "step": 252
     },
     {
       "epoch": 5.0,
-      "grad_norm": 1.030814290046692,
+      "grad_norm": 0.9423825740814209,
       "learning_rate": 0.0,
-      "loss": 3.6617,
+      "loss": 3.0027,
       "step": 315
     },
     {
       "epoch": 5.0,
-      "eval_gen_len": 30.177,
-      "eval_loss": 3.820514440536499,
-      "eval_rouge1": 18.6328,
-      "eval_rouge2": 4.2703,
-      "eval_rougeL": 15.6625,
-      "eval_rougeLsum": 16.9103,
-      "eval_runtime": 173.6125,
-      "eval_samples_per_second": 5.76,
-      "eval_steps_per_second": 0.184,
+      "eval_gen_len": 51.956,
+      "eval_loss": 2.752587080001831,
+      "eval_rouge1": 29.5518,
+      "eval_rouge2": 13.5408,
+      "eval_rougeL": 24.834,
+      "eval_rougeLsum": 27.6096,
+      "eval_runtime": 228.7726,
+      "eval_samples_per_second": 4.371,
+      "eval_steps_per_second": 0.14,
       "step": 315
     },
     {
       "epoch": 5.0,
       "step": 315,
       "total_flos": 3877644533760000.0,
-      "train_loss": 4.081830463712177,
-      "train_runtime": 1435.3865,
-      "train_samples_per_second": 3.483,
-      "train_steps_per_second": 0.219
+      "train_loss": 3.4973431299603175,
+      "train_runtime": 1609.8772,
+      "train_samples_per_second": 3.106,
+      "train_steps_per_second": 0.196
     }
   ],
   "logging_steps": 500,