End of training

README.md

@@ -3,11 +3,23 @@ license: mit
 base_model: facebook/bart-large-cnn
 tags:
 - generated_from_trainer
+datasets:
+- fedora-copr/pep-sum
 metrics:
 - rouge
 model-index:
 - name: pep_summarization
-  results: []
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: fedora-copr/pep-sum
+      type: fedora-copr/pep-sum
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 75.3806
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,7 +27,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # pep_summarization
 
-This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on an unknown dataset.
+This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on the fedora-copr/pep-sum dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.1242
 - Rouge1: 75.3806

all_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 5.0,
-    "eval_gen_len": 68.04347826086956,
-    "eval_loss": 0.9288389682769775,
-    "eval_rouge1": 39.7591,
-    "eval_rouge2": 24.1983,
-    "eval_rougeL": 32.2869,
-    "eval_rougeLsum": 36.1493,
-    "eval_runtime": 9.8095,
+    "eval_gen_len": 85.31884057971014,
+    "eval_loss": 0.12416736036539078,
+    "eval_rouge1": 75.3806,
+    "eval_rouge2": 74.6735,
+    "eval_rougeL": 75.5866,
+    "eval_rougeLsum": 75.5446,
+    "eval_runtime": 11.3295,
     "eval_samples": 69,
-    "eval_samples_per_second": 7.034,
-    "eval_steps_per_second": 1.835,
-    "train_loss": 0.8950336843297102,
-    "train_runtime": 114.301,
+    "eval_samples_per_second": 6.09,
+    "eval_steps_per_second": 1.589,
+    "train_loss": 0.036720043679942256,
+    "train_runtime": 132.663,
     "train_samples": 276,
-    "train_samples_per_second": 12.073,
-    "train_steps_per_second": 3.018
+    "train_samples_per_second": 10.402,
+    "train_steps_per_second": 2.601
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
-    "eval_gen_len": 68.04347826086956,
-    "eval_loss": 0.9288389682769775,
-    "eval_rouge1": 39.7591,
-    "eval_rouge2": 24.1983,
-    "eval_rougeL": 32.2869,
-    "eval_rougeLsum": 36.1493,
-    "eval_runtime": 9.8095,
+    "eval_gen_len": 85.31884057971014,
+    "eval_loss": 0.12416736036539078,
+    "eval_rouge1": 75.3806,
+    "eval_rouge2": 74.6735,
+    "eval_rougeL": 75.5866,
+    "eval_rougeLsum": 75.5446,
+    "eval_runtime": 11.3295,
     "eval_samples": 69,
-    "eval_samples_per_second": 7.034,
-    "eval_steps_per_second": 1.835
+    "eval_samples_per_second": 6.09,
+    "eval_steps_per_second": 1.589
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "train_loss": 0.8950336843297102,
-    "train_runtime": 114.301,
+    "train_loss": 0.036720043679942256,
+    "train_runtime": 132.663,
     "train_samples": 276,
-    "train_samples_per_second": 12.073,
-    "train_steps_per_second": 3.018
+    "train_samples_per_second": 10.402,
+    "train_steps_per_second": 2.601
 }

trainer_state.json

@@ -10,77 +10,77 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_gen_len": 68.04347826086956,
-      "eval_loss": 0.9288747310638428,
-      "eval_rouge1": 39.7591,
-      "eval_rouge2": 24.1983,
-      "eval_rougeL": 32.2869,
-      "eval_rougeLsum": 36.1493,
-      "eval_runtime": 9.3009,
-      "eval_samples_per_second": 7.419,
-      "eval_steps_per_second": 1.935,
+      "eval_gen_len": 95.44927536231884,
+      "eval_loss": 0.09570997208356857,
+      "eval_rouge1": 72.6601,
+      "eval_rouge2": 71.6824,
+      "eval_rougeL": 72.6858,
+      "eval_rougeLsum": 72.4668,
+      "eval_runtime": 12.0064,
+      "eval_samples_per_second": 5.747,
+      "eval_steps_per_second": 1.499,
       "step": 69
     },
     {
       "epoch": 2.0,
-      "eval_gen_len": 68.04347826086956,
-      "eval_loss": 0.9288538694381714,
-      "eval_rouge1": 39.7591,
-      "eval_rouge2": 24.1983,
-      "eval_rougeL": 32.2869,
-      "eval_rougeLsum": 36.1493,
-      "eval_runtime": 9.2625,
-      "eval_samples_per_second": 7.449,
-      "eval_steps_per_second": 1.943,
+      "eval_gen_len": 92.01449275362319,
+      "eval_loss": 0.13447459042072296,
+      "eval_rouge1": 75.0063,
+      "eval_rouge2": 74.0782,
+      "eval_rougeL": 75.0597,
+      "eval_rougeLsum": 74.8943,
+      "eval_runtime": 11.945,
+      "eval_samples_per_second": 5.776,
+      "eval_steps_per_second": 1.507,
       "step": 138
     },
     {
       "epoch": 3.0,
-      "eval_gen_len": 68.04347826086956,
-      "eval_loss": 0.9288433790206909,
-      "eval_rouge1": 39.7591,
-      "eval_rouge2": 24.1983,
-      "eval_rougeL": 32.2869,
-      "eval_rougeLsum": 36.1493,
-      "eval_runtime": 9.3044,
-      "eval_samples_per_second": 7.416,
-      "eval_steps_per_second": 1.935,
+      "eval_gen_len": 85.46376811594203,
+      "eval_loss": 0.14119356870651245,
+      "eval_rouge1": 75.3012,
+      "eval_rouge2": 74.5492,
+      "eval_rougeL": 75.4246,
+      "eval_rougeLsum": 75.324,
+      "eval_runtime": 10.9494,
+      "eval_samples_per_second": 6.302,
+      "eval_steps_per_second": 1.644,
       "step": 207
     },
     {
       "epoch": 4.0,
-      "eval_gen_len": 68.04347826086956,
-      "eval_loss": 0.9288394451141357,
-      "eval_rouge1": 39.7591,
-      "eval_rouge2": 24.1983,
-      "eval_rougeL": 32.2869,
-      "eval_rougeLsum": 36.1493,
-      "eval_runtime": 9.3367,
-      "eval_samples_per_second": 7.39,
-      "eval_steps_per_second": 1.928,
+      "eval_gen_len": 85.04347826086956,
+      "eval_loss": 0.10889122635126114,
+      "eval_rouge1": 74.8426,
+      "eval_rouge2": 74.0317,
+      "eval_rougeL": 74.8939,
+      "eval_rougeLsum": 74.8128,
+      "eval_runtime": 11.2109,
+      "eval_samples_per_second": 6.155,
+      "eval_steps_per_second": 1.606,
       "step": 276
     },
     {
       "epoch": 5.0,
-      "eval_gen_len": 68.04347826086956,
-      "eval_loss": 0.9288389682769775,
-      "eval_rouge1": 39.7591,
-      "eval_rouge2": 24.1983,
-      "eval_rougeL": 32.2869,
-      "eval_rougeLsum": 36.1493,
-      "eval_runtime": 9.2668,
-      "eval_samples_per_second": 7.446,
-      "eval_steps_per_second": 1.942,
+      "eval_gen_len": 85.31884057971014,
+      "eval_loss": 0.12416736036539078,
+      "eval_rouge1": 75.3806,
+      "eval_rouge2": 74.6735,
+      "eval_rougeL": 75.5866,
+      "eval_rougeLsum": 75.5446,
+      "eval_runtime": 11.151,
+      "eval_samples_per_second": 6.188,
+      "eval_steps_per_second": 1.614,
       "step": 345
     },
     {
       "epoch": 5.0,
       "step": 345,
-      "total_flos": 1680723777945600.0,
-      "train_loss": 0.8950336843297102,
-      "train_runtime": 114.301,
-      "train_samples_per_second": 12.073,
-      "train_steps_per_second": 3.018
+      "total_flos": 2990604350914560.0,
+      "train_loss": 0.036720043679942256,
+      "train_runtime": 132.663,
+      "train_samples_per_second": 10.402,
+      "train_steps_per_second": 2.601
     }
   ],
   "logging_steps": 500,
@@ -88,7 +88,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 1680723777945600.0,
+  "total_flos": 2990604350914560.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null