End of training

README.md

@@ -17,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [aubmindlab/aragpt2-base](https://huggingface.co/aubmindlab/aragpt2-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1062
-- Bleu: 0.1550
-- Rouge1: 0.4724
-- Rouge2: 0.2523
-- Rougel: 0.4657
+- Loss: 0.1032
+- Bleu: 0.1405
+- Rouge1: 0.4455
+- Rouge2: 0.2251
+- Rougel: 0.4383
 
 ## Model description
 

all_results.json

@@ -0,0 +1,19 @@
+{
+    "epoch": 10.0,
+    "eval_bleu": 0.14049162127130865,
+    "eval_loss": 0.10316114127635956,
+    "eval_rouge1": 0.44551198545007975,
+    "eval_rouge2": 0.22506890852974587,
+    "eval_rougeL": 0.4382572142917238,
+    "eval_runtime": 324.6518,
+    "eval_samples": 14209,
+    "eval_samples_per_second": 43.767,
+    "eval_steps_per_second": 5.474,
+    "perplexity": 1.1086700471890996,
+    "total_flos": 2.9701587861504e+17,
+    "train_loss": 0.09969830163342276,
+    "train_runtime": 40778.4516,
+    "train_samples": 56836,
+    "train_samples_per_second": 27.876,
+    "train_steps_per_second": 3.485
+}

egy_training_log.txt

@@ -1934,3 +1934,5 @@ INFO:root:Epoch 9.0: Train Loss = 0.0722, Eval Loss = 0.10443145781755447
 INFO:absl:Using default tokenizer.
 INFO:root:Epoch 10.0: Train Loss = 0.069, Eval Loss = 0.10583677142858505
 INFO:absl:Using default tokenizer.
+INFO:__main__:*** Evaluate ***
+INFO:absl:Using default tokenizer.

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 10.0,
+    "eval_bleu": 0.14049162127130865,
+    "eval_loss": 0.10316114127635956,
+    "eval_rouge1": 0.44551198545007975,
+    "eval_rouge2": 0.22506890852974587,
+    "eval_rougeL": 0.4382572142917238,
+    "eval_runtime": 324.6518,
+    "eval_samples": 14209,
+    "eval_samples_per_second": 43.767,
+    "eval_steps_per_second": 5.474,
+    "perplexity": 1.1086700471890996
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 2.9701587861504e+17,
+    "train_loss": 0.09969830163342276,
+    "train_runtime": 40778.4516,
+    "train_samples": 56836,
+    "train_samples_per_second": 27.876,
+    "train_steps_per_second": 3.485
+}

trainer_state.json

@@ -0,0 +1,241 @@
+{
+  "best_metric": 0.10316114127635956,
+  "best_model_checkpoint": "/home/iais_marenpielka/Bouthaina/res_nw_eg_aragpt2-base/checkpoint-35525",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 71050,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.1684599220752716,
+      "learning_rate": 4.766772598870057e-05,
+      "loss": 0.2542,
+      "step": 7105
+    },
+    {
+      "epoch": 1.0,
+      "eval_bleu": 0.07285026466195073,
+      "eval_loss": 0.11986471712589264,
+      "eval_rouge1": 0.31870746501286634,
+      "eval_rouge2": 0.11025143574709506,
+      "eval_rougeL": 0.30940400851788263,
+      "eval_runtime": 445.2495,
+      "eval_samples_per_second": 31.912,
+      "eval_steps_per_second": 3.991,
+      "step": 7105
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.16581259667873383,
+      "learning_rate": 4.515889830508475e-05,
+      "loss": 0.1078,
+      "step": 14210
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 0.104385167515318,
+      "eval_loss": 0.1118762344121933,
+      "eval_rouge1": 0.38033922316590574,
+      "eval_rouge2": 0.16357204386475152,
+      "eval_rougeL": 0.37196606997888715,
+      "eval_runtime": 445.4751,
+      "eval_samples_per_second": 31.896,
+      "eval_steps_per_second": 3.989,
+      "step": 14210
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.18644841015338898,
+      "learning_rate": 4.265007062146893e-05,
+      "loss": 0.0972,
+      "step": 21315
+    },
+    {
+      "epoch": 3.0,
+      "eval_bleu": 0.12224033944483013,
+      "eval_loss": 0.10767202824354172,
+      "eval_rouge1": 0.4109049665753015,
+      "eval_rouge2": 0.19329040527739555,
+      "eval_rougeL": 0.40326193036349967,
+      "eval_runtime": 383.8172,
+      "eval_samples_per_second": 37.02,
+      "eval_steps_per_second": 4.63,
+      "step": 21315
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.2712990939617157,
+      "learning_rate": 4.014124293785311e-05,
+      "loss": 0.0902,
+      "step": 28420
+    },
+    {
+      "epoch": 4.0,
+      "eval_bleu": 0.13124006032752472,
+      "eval_loss": 0.10514508932828903,
+      "eval_rouge1": 0.42940503923153844,
+      "eval_rouge2": 0.2090469696784658,
+      "eval_rougeL": 0.4223466730872162,
+      "eval_runtime": 323.7733,
+      "eval_samples_per_second": 43.886,
+      "eval_steps_per_second": 5.488,
+      "step": 28420
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.16391794383525848,
+      "learning_rate": 3.763241525423729e-05,
+      "loss": 0.0846,
+      "step": 35525
+    },
+    {
+      "epoch": 5.0,
+      "eval_bleu": 0.14049162127130865,
+      "eval_loss": 0.10316114127635956,
+      "eval_rouge1": 0.44551198545007975,
+      "eval_rouge2": 0.22506890852974587,
+      "eval_rougeL": 0.4382572142917238,
+      "eval_runtime": 340.0971,
+      "eval_samples_per_second": 41.779,
+      "eval_steps_per_second": 5.225,
+      "step": 35525
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.1612280309200287,
+      "learning_rate": 3.5123587570621466e-05,
+      "loss": 0.0799,
+      "step": 42630
+    },
+    {
+      "epoch": 6.0,
+      "eval_bleu": 0.14535221713569074,
+      "eval_loss": 0.10411085933446884,
+      "eval_rouge1": 0.45365603658460285,
+      "eval_rouge2": 0.23383881662198475,
+      "eval_rougeL": 0.4465500235966283,
+      "eval_runtime": 384.6849,
+      "eval_samples_per_second": 36.937,
+      "eval_steps_per_second": 4.619,
+      "step": 42630
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.252353310585022,
+      "learning_rate": 3.261475988700565e-05,
+      "loss": 0.0759,
+      "step": 49735
+    },
+    {
+      "epoch": 7.0,
+      "eval_bleu": 0.1493733281238007,
+      "eval_loss": 0.10441984981298447,
+      "eval_rouge1": 0.4622737132167348,
+      "eval_rouge2": 0.24252644756195563,
+      "eval_rougeL": 0.45529247029346154,
+      "eval_runtime": 324.6182,
+      "eval_samples_per_second": 43.771,
+      "eval_steps_per_second": 5.474,
+      "step": 49735
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.21136653423309326,
+      "learning_rate": 3.010593220338983e-05,
+      "loss": 0.0722,
+      "step": 56840
+    },
+    {
+      "epoch": 8.0,
+      "eval_bleu": 0.1526924816405305,
+      "eval_loss": 0.10443145781755447,
+      "eval_rouge1": 0.4655122846049303,
+      "eval_rouge2": 0.24695475478185897,
+      "eval_rougeL": 0.45872867266974005,
+      "eval_runtime": 324.8018,
+      "eval_samples_per_second": 43.747,
+      "eval_steps_per_second": 5.471,
+      "step": 56840
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.26080313324928284,
+      "learning_rate": 2.7597104519774014e-05,
+      "loss": 0.069,
+      "step": 63945
+    },
+    {
+      "epoch": 9.0,
+      "eval_bleu": 0.153555473627703,
+      "eval_loss": 0.10583677142858505,
+      "eval_rouge1": 0.4688625732021522,
+      "eval_rouge2": 0.24885496454985231,
+      "eval_rougeL": 0.46205637495730445,
+      "eval_runtime": 324.9582,
+      "eval_samples_per_second": 43.726,
+      "eval_steps_per_second": 5.468,
+      "step": 63945
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.27384456992149353,
+      "learning_rate": 2.5088276836158192e-05,
+      "loss": 0.066,
+      "step": 71050
+    },
+    {
+      "epoch": 10.0,
+      "eval_bleu": 0.1549549290214687,
+      "eval_loss": 0.10621096938848495,
+      "eval_rouge1": 0.4724064043822283,
+      "eval_rouge2": 0.25225313492301393,
+      "eval_rougeL": 0.46574710538787245,
+      "eval_runtime": 327.3388,
+      "eval_samples_per_second": 43.408,
+      "eval_steps_per_second": 5.429,
+      "step": 71050
+    },
+    {
+      "epoch": 10.0,
+      "step": 71050,
+      "total_flos": 2.9701587861504e+17,
+      "train_loss": 0.09969830163342276,
+      "train_runtime": 40778.4516,
+      "train_samples_per_second": 27.876,
+      "train_steps_per_second": 3.485
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 142100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9701587861504e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}