End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7622549019607843
+      value: 0.6936274509803921
     - name: F1
       type: f1
-      value: 0.8380634390651085
+      value: 0.8085758039816232
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9954
-- Accuracy: 0.7623
-- F1: 0.8381
-- Combined Score: 0.8002
+- Loss: 0.5772
+- Accuracy: 0.6936
+- F1: 0.8086
+- Combined Score: 0.7511
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6936274509803921,
+    "eval_combined_score": 0.7511016274810076,
+    "eval_f1": 0.8085758039816232,
+    "eval_loss": 0.5771955847740173,
+    "eval_runtime": 0.5133,
+    "eval_samples": 408,
+    "eval_samples_per_second": 794.855,
+    "eval_steps_per_second": 3.896,
+    "train_loss": 0.4106310119231542,
+    "train_runtime": 128.2586,
+    "train_samples": 3668,
+    "train_samples_per_second": 1429.924,
+    "train_steps_per_second": 5.848
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6936274509803921,
+    "eval_combined_score": 0.7511016274810076,
+    "eval_f1": 0.8085758039816232,
+    "eval_loss": 0.5771955847740173,
+    "eval_runtime": 0.5133,
+    "eval_samples": 408,
+    "eval_samples_per_second": 794.855,
+    "eval_steps_per_second": 3.896
+}

logs/events.out.tfevents.1677661058.kyoto.1072791.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2493fd940b16cacf2fdb429efdc5a0d0a3979785f1d508da3b988a2a95825344
+size 459

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.4106310119231542,
+    "train_runtime": 128.2586,
+    "train_samples": 3668,
+    "train_samples_per_second": 1429.924,
+    "train_steps_per_second": 5.848
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.5771955847740173,
+  "best_model_checkpoint": "hBERTv2_mrpc/checkpoint-45",
+  "epoch": 8.0,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.906666666666667e-05,
+      "loss": 0.6388,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6297235488891602,
+      "eval_runtime": 0.4865,
+      "eval_samples_per_second": 838.69,
+      "eval_steps_per_second": 4.111,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.806666666666667e-05,
+      "loss": 0.612,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.7511175174915782,
+      "eval_f1": 0.8135095447870779,
+      "eval_loss": 0.631518542766571,
+      "eval_runtime": 0.4865,
+      "eval_samples_per_second": 838.716,
+      "eval_steps_per_second": 4.111,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.706666666666667e-05,
+      "loss": 0.5725,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7511016274810076,
+      "eval_f1": 0.8085758039816232,
+      "eval_loss": 0.5771955847740173,
+      "eval_runtime": 0.485,
+      "eval_samples_per_second": 841.168,
+      "eval_steps_per_second": 4.123,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.606666666666667e-05,
+      "loss": 0.512,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7580659536541889,
+      "eval_f1": 0.8151515151515151,
+      "eval_loss": 0.6260703206062317,
+      "eval_runtime": 0.4882,
+      "eval_samples_per_second": 835.793,
+      "eval_steps_per_second": 4.097,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5066666666666667e-05,
+      "loss": 0.3924,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7279411764705882,
+      "eval_combined_score": 0.7737266857962697,
+      "eval_f1": 0.8195121951219512,
+      "eval_loss": 0.6433467268943787,
+      "eval_runtime": 0.4891,
+      "eval_samples_per_second": 834.113,
+      "eval_steps_per_second": 4.089,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.406666666666667e-05,
+      "loss": 0.2592,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7228365030222615,
+      "eval_f1": 0.7593984962406015,
+      "eval_loss": 0.7530680894851685,
+      "eval_runtime": 0.4871,
+      "eval_samples_per_second": 837.615,
+      "eval_steps_per_second": 4.106,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3066666666666665e-05,
+      "loss": 0.1689,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7377450980392157,
+      "eval_combined_score": 0.7767899328406062,
+      "eval_f1": 0.8158347676419966,
+      "eval_loss": 0.7904398441314697,
+      "eval_runtime": 0.4881,
+      "eval_samples_per_second": 835.949,
+      "eval_steps_per_second": 4.098,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.206666666666667e-05,
+      "loss": 0.1292,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7622549019607843,
+      "eval_combined_score": 0.8001591705129464,
+      "eval_f1": 0.8380634390651085,
+      "eval_loss": 0.9954336285591125,
+      "eval_runtime": 0.4868,
+      "eval_samples_per_second": 838.079,
+      "eval_steps_per_second": 4.108,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "step": 120,
+      "total_flos": 3654895836069888.0,
+      "train_loss": 0.4106310119231542,
+      "train_runtime": 128.2586,
+      "train_samples_per_second": 1429.924,
+      "train_steps_per_second": 5.848
+    }
+  ],
+  "max_steps": 750,
+  "num_train_epochs": 50,
+  "total_flos": 3654895836069888.0,
+  "trial_name": null,
+  "trial_params": null
+}