End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE QNLI
       type: glue
       config: qnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5722130697419
+      value: 0.580633351638294
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # distilbert_add_GLUE_Experiment_logit_kd_qnli_96
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4074
-- Accuracy: 0.5722
+- Loss: 0.3984
+- Accuracy: 0.5806
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.580633351638294,
+    "eval_loss": 0.39838820695877075,
+    "eval_runtime": 5.3106,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 1028.7,
+    "eval_steps_per_second": 4.143,
+    "train_loss": 0.396416024722381,
+    "train_runtime": 1171.9486,
+    "train_samples": 104743,
+    "train_samples_per_second": 4468.754,
+    "train_steps_per_second": 17.492
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.580633351638294,
+    "eval_loss": 0.39838820695877075,
+    "eval_runtime": 5.3106,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 1028.7,
+    "eval_steps_per_second": 4.143
+}

logs/events.out.tfevents.1674949306.serv-3334.540580.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ea2c6750f34c6d708ff44fd72e6869c8c1dc531a51bd368aa04f4be125e1505
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.396416024722381,
+    "train_runtime": 1171.9486,
+    "train_samples": 104743,
+    "train_samples_per_second": 4468.754,
+    "train_steps_per_second": 17.492
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 0.39838820695877075,
+  "best_model_checkpoint": "distilbert_add_GLUE_Experiment_logit_kd_qnli_96/checkpoint-1640",
+  "epoch": 9.0,
+  "global_step": 3690,
+  "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": 0.4153,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.4113771617412567,
+      "eval_runtime": 5.2932,
+      "eval_samples_per_second": 1032.078,
+      "eval_steps_per_second": 4.156,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.4152,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.4115099012851715,
+      "eval_runtime": 5.1775,
+      "eval_samples_per_second": 1055.137,
+      "eval_steps_per_second": 4.249,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.4129,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5685520776130332,
+      "eval_loss": 0.4019591510295868,
+      "eval_runtime": 5.1907,
+      "eval_samples_per_second": 1052.462,
+      "eval_steps_per_second": 4.238,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.3995,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.580633351638294,
+      "eval_loss": 0.39838820695877075,
+      "eval_runtime": 5.1978,
+      "eval_samples_per_second": 1051.02,
+      "eval_steps_per_second": 4.233,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3934,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5793520043931906,
+      "eval_loss": 0.3992016017436981,
+      "eval_runtime": 5.1986,
+      "eval_samples_per_second": 1050.862,
+      "eval_steps_per_second": 4.232,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3888,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5809994508511807,
+      "eval_loss": 0.4023948609828949,
+      "eval_runtime": 5.2149,
+      "eval_samples_per_second": 1047.584,
+      "eval_steps_per_second": 4.219,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.385,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5674537799743731,
+      "eval_loss": 0.4104597866535187,
+      "eval_runtime": 5.2043,
+      "eval_samples_per_second": 1049.712,
+      "eval_steps_per_second": 4.227,
+      "step": 2870
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.3808,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5777045579352005,
+      "eval_loss": 0.4049883186817169,
+      "eval_runtime": 5.1888,
+      "eval_samples_per_second": 1052.838,
+      "eval_steps_per_second": 4.24,
+      "step": 3280
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3768,
+      "step": 3690
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5722130697419,
+      "eval_loss": 0.4074430465698242,
+      "eval_runtime": 5.2968,
+      "eval_samples_per_second": 1031.375,
+      "eval_steps_per_second": 4.153,
+      "step": 3690
+    },
+    {
+      "epoch": 9.0,
+      "step": 3690,
+      "total_flos": 5415250056708096.0,
+      "train_loss": 0.396416024722381,
+      "train_runtime": 1171.9486,
+      "train_samples_per_second": 4468.754,
+      "train_steps_per_second": 17.492
+    }
+  ],
+  "max_steps": 20500,
+  "num_train_epochs": 50,
+  "total_flos": 5415250056708096.0,
+  "trial_name": null,
+  "trial_params": null
+}