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_256
 
-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": 1.0610467195510864,
+    "eval_runtime": 12.07,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 452.609,
+    "eval_steps_per_second": 3.563,
+    "train_loss": 1.1393204408137676,
+    "train_runtime": 5554.7339,
+    "train_samples": 104743,
+    "train_samples_per_second": 942.826,
+    "train_steps_per_second": 7.372
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 1.0610467195510864,
+    "eval_runtime": 12.07,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 452.609,
+    "eval_steps_per_second": 3.563
+}

logs/events.out.tfevents.1674954550.serv-3328.2501856.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd49be75fcd1d12ddbf39db332b66f1fab1eda489daec5a0f8c27558ac3a3a49
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 1.1393204408137676,
+    "train_runtime": 5554.7339,
+    "train_samples": 104743,
+    "train_samples_per_second": 942.826,
+    "train_steps_per_second": 7.372
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 1.0610467195510864,
+  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_qnli_256/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": 1.1396,
+      "step": 819
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.061155080795288,
+      "eval_runtime": 11.9927,
+      "eval_samples_per_second": 455.526,
+      "eval_steps_per_second": 3.586,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 1.1393,
+      "step": 1638
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0610756874084473,
+      "eval_runtime": 11.9855,
+      "eval_samples_per_second": 455.799,
+      "eval_steps_per_second": 3.588,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 1.1393,
+      "step": 2457
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0616600513458252,
+      "eval_runtime": 12.0036,
+      "eval_samples_per_second": 455.112,
+      "eval_steps_per_second": 3.582,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 1.1393,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0610467195510864,
+      "eval_runtime": 12.0579,
+      "eval_samples_per_second": 453.063,
+      "eval_steps_per_second": 3.566,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 1.1394,
+      "step": 4095
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0612071752548218,
+      "eval_runtime": 12.0792,
+      "eval_samples_per_second": 452.264,
+      "eval_steps_per_second": 3.56,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 1.1393,
+      "step": 4914
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0612789392471313,
+      "eval_runtime": 11.9193,
+      "eval_samples_per_second": 458.331,
+      "eval_steps_per_second": 3.608,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 1.1393,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0614013671875,
+      "eval_runtime": 11.9153,
+      "eval_samples_per_second": 458.485,
+      "eval_steps_per_second": 3.609,
+      "step": 5733
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 1.1393,
+      "step": 6552
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.061530351638794,
+      "eval_runtime": 11.9396,
+      "eval_samples_per_second": 457.555,
+      "eval_steps_per_second": 3.601,
+      "step": 6552
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 1.1392,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 1.0610854625701904,
+      "eval_runtime": 12.0583,
+      "eval_samples_per_second": 453.047,
+      "eval_steps_per_second": 3.566,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "step": 7371,
+      "total_flos": 2.4297959017414656e+16,
+      "train_loss": 1.1393204408137676,
+      "train_runtime": 5554.7339,
+      "train_samples_per_second": 942.826,
+      "train_steps_per_second": 7.372
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 2.4297959017414656e+16,
+  "trial_name": null,
+  "trial_params": null
+}