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
@@ -29,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mobilebert_add_GLUE_Experiment_logit_kd_qnli
 
-This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on the glue dataset.
+This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0611
+- Loss: 1.0610
 - Accuracy: 0.5054
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 1.0610356330871582,
+    "eval_runtime": 12.0493,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 453.387,
+    "eval_steps_per_second": 3.569,
+    "train_loss": 1.139348039535214,
+    "train_runtime": 5578.6907,
+    "train_samples": 104743,
+    "train_samples_per_second": 938.778,
+    "train_steps_per_second": 7.34
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 1.0610356330871582,
+    "eval_runtime": 12.0493,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 453.387,
+    "eval_steps_per_second": 3.569
+}

logs/events.out.tfevents.1674955275.serv-3332.3626077.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:210e2455e87ba44ea0769f2a040bcb402abe5f5c975ec56f7efec1040a8bcac4
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 1.139348039535214,
+    "train_runtime": 5578.6907,
+    "train_samples": 104743,
+    "train_samples_per_second": 938.778,
+    "train_steps_per_second": 7.34
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 1.0610356330871582,
+  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_qnli/checkpoint-3276",
+  "epoch": 9.0,
+  "global_step": 7371,
+  "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": 1.1397,
+      "step": 819
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0611544847488403,
+      "eval_runtime": 12.0737,
+      "eval_samples_per_second": 452.471,
+      "eval_steps_per_second": 3.561,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 1.1394,
+      "step": 1638
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0610606670379639,
+      "eval_runtime": 12.0421,
+      "eval_samples_per_second": 453.657,
+      "eval_steps_per_second": 3.571,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 1.1393,
+      "step": 2457
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0616439580917358,
+      "eval_runtime": 12.0412,
+      "eval_samples_per_second": 453.691,
+      "eval_steps_per_second": 3.571,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 1.1393,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0610356330871582,
+      "eval_runtime": 12.2467,
+      "eval_samples_per_second": 446.079,
+      "eval_steps_per_second": 3.511,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 1.1394,
+      "step": 4095
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0612379312515259,
+      "eval_runtime": 12.0192,
+      "eval_samples_per_second": 454.522,
+      "eval_steps_per_second": 3.578,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 1.1393,
+      "step": 4914
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0612590312957764,
+      "eval_runtime": 12.209,
+      "eval_samples_per_second": 447.458,
+      "eval_steps_per_second": 3.522,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 1.1393,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0614322423934937,
+      "eval_runtime": 12.0416,
+      "eval_samples_per_second": 453.678,
+      "eval_steps_per_second": 3.571,
+      "step": 5733
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 1.1393,
+      "step": 6552
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0612900257110596,
+      "eval_runtime": 12.028,
+      "eval_samples_per_second": 454.19,
+      "eval_steps_per_second": 3.575,
+      "step": 6552
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 1.1392,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0610575675964355,
+      "eval_runtime": 12.0528,
+      "eval_samples_per_second": 453.258,
+      "eval_steps_per_second": 3.568,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "step": 7371,
+      "total_flos": 2.788502631402701e+16,
+      "train_loss": 1.139348039535214,
+      "train_runtime": 5578.6907,
+      "train_samples_per_second": 938.778,
+      "train_steps_per_second": 7.34
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 2.788502631402701e+16,
+  "trial_name": null,
+  "trial_params": null
+}