End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -14,7 +16,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -22,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6862745098039216
+      value: 0.7107843137254902
     - name: F1
       type: f1
-      value: 0.7808219178082192
+      value: 0.8233532934131738
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,12 +35,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # distilbert_add_GLUE_Experiment_mrpc_256
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the glue dataset.
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7583
-- Accuracy: 0.6863
-- F1: 0.7808
-- Combined Score: 0.7335
+- Loss: 0.5932
+- Accuracy: 0.7108
+- F1: 0.8234
+- Combined Score: 0.7671
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 21.0,
+    "eval_accuracy": 0.7107843137254902,
+    "eval_combined_score": 0.767068803569332,
+    "eval_f1": 0.8233532934131738,
+    "eval_loss": 0.5931658744812012,
+    "eval_runtime": 0.1999,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2040.737,
+    "eval_steps_per_second": 10.004,
+    "train_loss": 0.5765140730237204,
+    "train_runtime": 112.7881,
+    "train_samples": 3668,
+    "train_samples_per_second": 1626.058,
+    "train_steps_per_second": 6.65
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 21.0,
+    "eval_accuracy": 0.7107843137254902,
+    "eval_combined_score": 0.767068803569332,
+    "eval_f1": 0.8233532934131738,
+    "eval_loss": 0.5931658744812012,
+    "eval_runtime": 0.1999,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2040.737,
+    "eval_steps_per_second": 10.004
+}

logs/events.out.tfevents.1674736493.glendale.2597965.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbb22f0797ed2147be72e81c5c4c58792d1582c792c93f2d761df32e489cd101
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 21.0,
+    "train_loss": 0.5765140730237204,
+    "train_runtime": 112.7881,
+    "train_samples": 3668,
+    "train_samples_per_second": 1626.058,
+    "train_steps_per_second": 6.65
+}

trainer_state.json

@@ -0,0 +1,382 @@
+{
+  "best_metric": 0.5931658744812012,
+  "best_model_checkpoint": "distilbert_add_GLUE_Experiment_mrpc_256/checkpoint-240",
+  "epoch": 21.0,
+  "global_step": 315,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.637,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6241778135299683,
+      "eval_runtime": 0.2066,
+      "eval_samples_per_second": 1975.28,
+      "eval_steps_per_second": 9.683,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.629,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6240377426147461,
+      "eval_runtime": 0.1942,
+      "eval_samples_per_second": 2100.738,
+      "eval_steps_per_second": 10.298,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6302,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6248239874839783,
+      "eval_runtime": 0.1917,
+      "eval_samples_per_second": 2128.019,
+      "eval_steps_per_second": 10.431,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.63,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.624061107635498,
+      "eval_runtime": 0.1945,
+      "eval_samples_per_second": 2097.442,
+      "eval_steps_per_second": 10.282,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6323,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6240324378013611,
+      "eval_runtime": 0.1944,
+      "eval_samples_per_second": 2098.803,
+      "eval_steps_per_second": 10.288,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6299,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6242926120758057,
+      "eval_runtime": 0.2052,
+      "eval_samples_per_second": 1988.12,
+      "eval_steps_per_second": 9.746,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6325,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6239174604415894,
+      "eval_runtime": 0.2195,
+      "eval_samples_per_second": 1859.06,
+      "eval_steps_per_second": 9.113,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6301,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.623868465423584,
+      "eval_runtime": 0.1993,
+      "eval_samples_per_second": 2047.174,
+      "eval_steps_per_second": 10.035,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6324,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6240455508232117,
+      "eval_runtime": 0.1991,
+      "eval_samples_per_second": 2049.734,
+      "eval_steps_per_second": 10.048,
+      "step": 135
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.6293,
+      "step": 150
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6240090727806091,
+      "eval_runtime": 0.1918,
+      "eval_samples_per_second": 2127.006,
+      "eval_steps_per_second": 10.426,
+      "step": 150
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.6307,
+      "step": 165
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6239336729049683,
+      "eval_runtime": 0.1917,
+      "eval_samples_per_second": 2128.045,
+      "eval_steps_per_second": 10.432,
+      "step": 165
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.6302,
+      "step": 180
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6240288019180298,
+      "eval_runtime": 0.2014,
+      "eval_samples_per_second": 2025.663,
+      "eval_steps_per_second": 9.93,
+      "step": 180
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.6338,
+      "step": 195
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6237308382987976,
+      "eval_runtime": 0.196,
+      "eval_samples_per_second": 2081.206,
+      "eval_steps_per_second": 10.202,
+      "step": 195
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6281,
+      "step": 210
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6225215792655945,
+      "eval_runtime": 0.1949,
+      "eval_samples_per_second": 2092.958,
+      "eval_steps_per_second": 10.26,
+      "step": 210
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 3.5e-05,
+      "loss": 0.6263,
+      "step": 225
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6182957291603088,
+      "eval_runtime": 0.2058,
+      "eval_samples_per_second": 1982.67,
+      "eval_steps_per_second": 9.719,
+      "step": 225
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.6017,
+      "step": 240
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.7107843137254902,
+      "eval_combined_score": 0.767068803569332,
+      "eval_f1": 0.8233532934131738,
+      "eval_loss": 0.5931658744812012,
+      "eval_runtime": 0.1933,
+      "eval_samples_per_second": 2110.847,
+      "eval_steps_per_second": 10.347,
+      "step": 240
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.3e-05,
+      "loss": 0.5213,
+      "step": 255
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.6642156862745098,
+      "eval_combined_score": 0.7091275917907558,
+      "eval_f1": 0.7540394973070018,
+      "eval_loss": 0.6146347522735596,
+      "eval_runtime": 0.2025,
+      "eval_samples_per_second": 2014.987,
+      "eval_steps_per_second": 9.877,
+      "step": 255
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.4383,
+      "step": 270
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.6911764705882353,
+      "eval_combined_score": 0.7377115229653506,
+      "eval_f1": 0.7842465753424658,
+      "eval_loss": 0.6404514908790588,
+      "eval_runtime": 0.2015,
+      "eval_samples_per_second": 2025.167,
+      "eval_steps_per_second": 9.927,
+      "step": 270
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 3.1e-05,
+      "loss": 0.3903,
+      "step": 285
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.6911764705882353,
+      "eval_combined_score": 0.7391693163751987,
+      "eval_f1": 0.7871621621621622,
+      "eval_loss": 0.6909675002098083,
+      "eval_runtime": 0.1929,
+      "eval_samples_per_second": 2115.29,
+      "eval_steps_per_second": 10.369,
+      "step": 285
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 3e-05,
+      "loss": 0.363,
+      "step": 300
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.6544117647058824,
+      "eval_combined_score": 0.695920966151824,
+      "eval_f1": 0.7374301675977655,
+      "eval_loss": 0.7221016883850098,
+      "eval_runtime": 0.2057,
+      "eval_samples_per_second": 1983.587,
+      "eval_steps_per_second": 9.723,
+      "step": 300
+    },
+    {
+      "epoch": 21.0,
+      "learning_rate": 2.9e-05,
+      "loss": 0.3306,
+      "step": 315
+    },
+    {
+      "epoch": 21.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7335482138060704,
+      "eval_f1": 0.7808219178082192,
+      "eval_loss": 0.7583230137825012,
+      "eval_runtime": 0.1959,
+      "eval_samples_per_second": 2082.795,
+      "eval_steps_per_second": 10.21,
+      "step": 315
+    },
+    {
+      "epoch": 21.0,
+      "step": 315,
+      "total_flos": 1269130346889216.0,
+      "train_loss": 0.5765140730237204,
+      "train_runtime": 112.7881,
+      "train_samples_per_second": 1626.058,
+      "train_steps_per_second": 6.65
+    }
+  ],
+  "max_steps": 750,
+  "num_train_epochs": 50,
+  "total_flos": 1269130346889216.0,
+  "trial_name": null,
+  "trial_params": null
+}