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: 0.5756910122643236
+      value: 0.5861248398315944
 ---
 
 <!-- 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_384
 
-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: 0.4081
-- Accuracy: 0.5757
+- Loss: 0.3959
+- Accuracy: 0.5861
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5861248398315944,
+    "eval_loss": 0.3959311246871948,
+    "eval_runtime": 5.504,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 992.554,
+    "eval_steps_per_second": 3.997,
+    "train_loss": 0.38917779158216737,
+    "train_runtime": 1070.966,
+    "train_samples": 104743,
+    "train_samples_per_second": 4890.118,
+    "train_steps_per_second": 19.142
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5861248398315944,
+    "eval_loss": 0.3959311246871948,
+    "eval_runtime": 5.504,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 992.554,
+    "eval_steps_per_second": 3.997
+}

logs/events.out.tfevents.1674949453.serv-3334.550938.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fdadb94d63d0c73a1af0f85b01b250b88821aedda28bf1051d4dcd78ba160d2
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.38917779158216737,
+    "train_runtime": 1070.966,
+    "train_samples": 104743,
+    "train_samples_per_second": 4890.118,
+    "train_steps_per_second": 19.142
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.3959311246871948,
+  "best_model_checkpoint": "distilbert_add_GLUE_Experiment_logit_kd_qnli_384/checkpoint-820",
+  "epoch": 7.0,
+  "global_step": 2870,
+  "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": 0.4145,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5749588138385503,
+      "eval_loss": 0.39820435643196106,
+      "eval_runtime": 5.5073,
+      "eval_samples_per_second": 991.948,
+      "eval_steps_per_second": 3.995,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.4004,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5861248398315944,
+      "eval_loss": 0.3959311246871948,
+      "eval_runtime": 5.5172,
+      "eval_samples_per_second": 990.178,
+      "eval_steps_per_second": 3.988,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.3922,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5771554091158704,
+      "eval_loss": 0.39736953377723694,
+      "eval_runtime": 5.5113,
+      "eval_samples_per_second": 991.236,
+      "eval_steps_per_second": 3.992,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.3865,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5815485996705108,
+      "eval_loss": 0.4004422724246979,
+      "eval_runtime": 5.5769,
+      "eval_samples_per_second": 979.574,
+      "eval_steps_per_second": 3.945,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3815,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5793520043931906,
+      "eval_loss": 0.4000557065010071,
+      "eval_runtime": 5.5231,
+      "eval_samples_per_second": 989.117,
+      "eval_steps_per_second": 3.983,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3767,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5777045579352005,
+      "eval_loss": 0.4072972238063812,
+      "eval_runtime": 5.5183,
+      "eval_samples_per_second": 989.977,
+      "eval_steps_per_second": 3.987,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3725,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5756910122643236,
+      "eval_loss": 0.4080561101436615,
+      "eval_runtime": 5.5141,
+      "eval_samples_per_second": 990.732,
+      "eval_steps_per_second": 3.99,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "step": 2870,
+      "total_flos": 1.917231202880717e+16,
+      "train_loss": 0.38917779158216737,
+      "train_runtime": 1070.966,
+      "train_samples_per_second": 4890.118,
+      "train_steps_per_second": 19.142
+    }
+  ],
+  "max_steps": 20500,
+  "num_train_epochs": 50,
+  "total_flos": 1.917231202880717e+16,
+  "trial_name": null,
+  "trial_params": null
+}