End of training

README.md

@@ -3,6 +3,8 @@ license: apache-2.0
 base_model: google-t5/t5-small
 tags:
 - generated_from_trainer
+metrics:
+- rouge
 model-index:
 - name: tst-summarization
   results: []
@@ -15,6 +17,13 @@ should probably proofread and complete it, then remove this comment. -->
 # tst-summarization
 
 This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5129
+- Rouge1: 69.4813
+- Rouge2: 53.8739
+- Rougel: 69.3727
+- Rougelsum: 69.2986
+- Gen Len: 19.4911
 
 ## Model description
 

all_results.json

@@ -1,19 +1,19 @@
 {
-    "epoch": 3.0,
-    "eval_gen_len": 21.107355864811133,
-    "eval_loss": 0.6219637989997864,
-    "eval_rouge1": 65.6704,
-    "eval_rouge2": 51.3605,
-    "eval_rougeL": 65.6565,
-    "eval_rougeLsum": 65.3846,
-    "eval_runtime": 17.9666,
+    "epoch": 10.0,
+    "eval_gen_len": 19.491053677932406,
+    "eval_loss": 0.5128998756408691,
+    "eval_rouge1": 69.4813,
+    "eval_rouge2": 53.8739,
+    "eval_rougeL": 69.3727,
+    "eval_rougeLsum": 69.2986,
+    "eval_runtime": 16.6312,
     "eval_samples": 503,
-    "eval_samples_per_second": 27.996,
-    "eval_steps_per_second": 3.507,
-    "total_flos": 454479091040256.0,
-    "train_loss": 1.124249507675013,
-    "train_runtime": 54.6803,
+    "eval_samples_per_second": 30.244,
+    "eval_steps_per_second": 3.788,
+    "total_flos": 1519376405004288.0,
+    "train_loss": 0.8102951523680124,
+    "train_runtime": 150.8906,
     "train_samples": 5149,
-    "train_samples_per_second": 282.497,
-    "train_steps_per_second": 8.833
+    "train_samples_per_second": 341.241,
+    "train_steps_per_second": 10.67
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 3.0,
-    "eval_gen_len": 21.107355864811133,
-    "eval_loss": 0.6219637989997864,
-    "eval_rouge1": 65.6704,
-    "eval_rouge2": 51.3605,
-    "eval_rougeL": 65.6565,
-    "eval_rougeLsum": 65.3846,
-    "eval_runtime": 17.9666,
+    "epoch": 10.0,
+    "eval_gen_len": 19.491053677932406,
+    "eval_loss": 0.5128998756408691,
+    "eval_rouge1": 69.4813,
+    "eval_rouge2": 53.8739,
+    "eval_rougeL": 69.3727,
+    "eval_rougeLsum": 69.2986,
+    "eval_runtime": 16.6312,
     "eval_samples": 503,
-    "eval_samples_per_second": 27.996,
-    "eval_steps_per_second": 3.507
+    "eval_samples_per_second": 30.244,
+    "eval_steps_per_second": 3.788
 }

generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
   "pad_token_id": 0,

runs/Jun30_18-13-45_5bcf15f9-cbd7-4847-8c67-530b27aedefb/events.out.tfevents.1719771405.5bcf15f9-cbd7-4847-8c67-530b27aedefb.1162.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d92eff840e2b1436043c4d0e7e00a275e3e67ef7f143b093e4a12b5485ac0081
+size 613

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 3.0,
-    "total_flos": 454479091040256.0,
-    "train_loss": 1.124249507675013,
-    "train_runtime": 54.6803,
+    "epoch": 10.0,
+    "total_flos": 1519376405004288.0,
+    "train_loss": 0.8102951523680124,
+    "train_runtime": 150.8906,
     "train_samples": 5149,
-    "train_samples_per_second": 282.497,
-    "train_steps_per_second": 8.833
+    "train_samples_per_second": 341.241,
+    "train_steps_per_second": 10.67
 }

trainer_state.json

@@ -1,27 +1,48 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 483,
+  "global_step": 1610,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 3.0,
-      "step": 483,
-      "total_flos": 454479091040256.0,
-      "train_loss": 1.124249507675013,
-      "train_runtime": 54.6803,
-      "train_samples_per_second": 282.497,
-      "train_steps_per_second": 8.833
+      "epoch": 3.1055900621118013,
+      "grad_norm": 2.380723476409912,
+      "learning_rate": 3.4472049689440996e-05,
+      "loss": 1.0678,
+      "step": 500
+    },
+    {
+      "epoch": 6.211180124223603,
+      "grad_norm": 2.099438428878784,
+      "learning_rate": 1.894409937888199e-05,
+      "loss": 0.7287,
+      "step": 1000
+    },
+    {
+      "epoch": 9.316770186335404,
+      "grad_norm": 1.4192956686019897,
+      "learning_rate": 3.4161490683229816e-06,
+      "loss": 0.6669,
+      "step": 1500
+    },
+    {
+      "epoch": 10.0,
+      "step": 1610,
+      "total_flos": 1519376405004288.0,
+      "train_loss": 0.8102951523680124,
+      "train_runtime": 150.8906,
+      "train_samples_per_second": 341.241,
+      "train_steps_per_second": 10.67
     }
   ],
   "logging_steps": 500,
-  "max_steps": 483,
+  "max_steps": 1610,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -35,7 +56,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 454479091040256.0,
+  "total_flos": 1519376405004288.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null