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 SST2
       type: glue
       config: sst2
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7878440366972477
+      value: 0.801605504587156
 ---
 
 <!-- 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. -->
 
 # mobilebert_add_GLUE_Experiment_logit_kd_sst2
 
-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 SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9595
-- Accuracy: 0.7878
+- Loss: 0.7778
+- Accuracy: 0.8016
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 13.0,
+    "eval_accuracy": 0.801605504587156,
+    "eval_loss": 0.7777931094169617,
+    "eval_runtime": 1.9584,
+    "eval_samples": 872,
+    "eval_samples_per_second": 445.265,
+    "eval_steps_per_second": 3.574,
+    "train_loss": 0.8076285107642817,
+    "train_runtime": 5189.2983,
+    "train_samples": 67349,
+    "train_samples_per_second": 648.922,
+    "train_steps_per_second": 5.078
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 13.0,
+    "eval_accuracy": 0.801605504587156,
+    "eval_loss": 0.7777931094169617,
+    "eval_runtime": 1.9584,
+    "eval_samples": 872,
+    "eval_samples_per_second": 445.265,
+    "eval_steps_per_second": 3.574
+}

logs/events.out.tfevents.1674985095.serv-3332.3626077.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afa935d2e096aa8871e7eafc42fd778d3b350afc7d2ee9c6e798bf8f359038e7
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.0,
+    "train_loss": 0.8076285107642817,
+    "train_runtime": 5189.2983,
+    "train_samples": 67349,
+    "train_samples_per_second": 648.922,
+    "train_steps_per_second": 5.078
+}

trainer_state.json

@@ -0,0 +1,220 @@
+{
+  "best_metric": 0.7777931094169617,
+  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_sst2/checkpoint-4216",
+  "epoch": 13.0,
+  "global_step": 6851,
+  "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": 1.5405,
+      "step": 527
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5538990825688074,
+      "eval_loss": 1.4224907159805298,
+      "eval_runtime": 1.9444,
+      "eval_samples_per_second": 448.472,
+      "eval_steps_per_second": 3.6,
+      "step": 527
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 1.3567,
+      "step": 1054
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5481651376146789,
+      "eval_loss": 1.470656394958496,
+      "eval_runtime": 1.958,
+      "eval_samples_per_second": 445.343,
+      "eval_steps_per_second": 3.575,
+      "step": 1054
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 1.2859,
+      "step": 1581
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5676605504587156,
+      "eval_loss": 1.4661372900009155,
+      "eval_runtime": 1.9409,
+      "eval_samples_per_second": 449.269,
+      "eval_steps_per_second": 3.607,
+      "step": 1581
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 1.2563,
+      "step": 2108
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5665137614678899,
+      "eval_loss": 1.4136215448379517,
+      "eval_runtime": 1.9617,
+      "eval_samples_per_second": 444.52,
+      "eval_steps_per_second": 3.568,
+      "step": 2108
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 1.2414,
+      "step": 2635
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5940366972477065,
+      "eval_loss": 1.4238868951797485,
+      "eval_runtime": 1.9469,
+      "eval_samples_per_second": 447.897,
+      "eval_steps_per_second": 3.596,
+      "step": 2635
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 1.2288,
+      "step": 3162
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5745412844036697,
+      "eval_loss": 1.4442880153656006,
+      "eval_runtime": 1.9462,
+      "eval_samples_per_second": 448.059,
+      "eval_steps_per_second": 3.597,
+      "step": 3162
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.7679,
+      "step": 3689
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7878440366972477,
+      "eval_loss": 0.7869645357131958,
+      "eval_runtime": 1.9376,
+      "eval_samples_per_second": 450.039,
+      "eval_steps_per_second": 3.613,
+      "step": 3689
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4135,
+      "step": 4216
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.801605504587156,
+      "eval_loss": 0.7777931094169617,
+      "eval_runtime": 1.9457,
+      "eval_samples_per_second": 448.16,
+      "eval_steps_per_second": 3.598,
+      "step": 4216
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3376,
+      "step": 4743
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7993119266055045,
+      "eval_loss": 0.8672550320625305,
+      "eval_runtime": 2.0222,
+      "eval_samples_per_second": 431.216,
+      "eval_steps_per_second": 3.462,
+      "step": 4743
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.2972,
+      "step": 5270
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7901376146788991,
+      "eval_loss": 0.8790308237075806,
+      "eval_runtime": 1.9375,
+      "eval_samples_per_second": 450.065,
+      "eval_steps_per_second": 3.613,
+      "step": 5270
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.2734,
+      "step": 5797
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.7912844036697247,
+      "eval_loss": 0.9524618983268738,
+      "eval_runtime": 1.9456,
+      "eval_samples_per_second": 448.195,
+      "eval_steps_per_second": 3.598,
+      "step": 5797
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.2569,
+      "step": 6324
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7935779816513762,
+      "eval_loss": 0.9557108879089355,
+      "eval_runtime": 1.9404,
+      "eval_samples_per_second": 449.382,
+      "eval_steps_per_second": 3.607,
+      "step": 6324
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.2431,
+      "step": 6851
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.7878440366972477,
+      "eval_loss": 0.959472119808197,
+      "eval_runtime": 1.9369,
+      "eval_samples_per_second": 450.192,
+      "eval_steps_per_second": 3.614,
+      "step": 6851
+    },
+    {
+      "epoch": 13.0,
+      "step": 6851,
+      "total_flos": 2.5898703529181184e+16,
+      "train_loss": 0.8076285107642817,
+      "train_runtime": 5189.2983,
+      "train_samples_per_second": 648.922,
+      "train_steps_per_second": 5.078
+    }
+  ],
+  "max_steps": 26350,
+  "num_train_epochs": 50,
+  "total_flos": 2.5898703529181184e+16,
+  "trial_name": null,
+  "trial_params": null
+}