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 WNLI
       type: glue
       config: wnli
       split: validation
@@ -29,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mobilebert_add_GLUE_Experiment_logit_kd_wnli_128
 
-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 WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3460
+- Loss: 0.3455
 - Accuracy: 0.5634
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.3455486595630646,
+    "eval_runtime": 0.1386,
+    "eval_samples": 71,
+    "eval_samples_per_second": 512.195,
+    "eval_steps_per_second": 7.214,
+    "train_loss": 0.3466017405192057,
+    "train_runtime": 61.4918,
+    "train_samples": 635,
+    "train_samples_per_second": 516.329,
+    "train_steps_per_second": 4.066
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.3455486595630646,
+    "eval_runtime": 0.1386,
+    "eval_samples": 71,
+    "eval_samples_per_second": 512.195,
+    "eval_steps_per_second": 7.214
+}

logs/events.out.tfevents.1674975875.serv-3334.518319.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b2d4971d3164d539058f54eb84d42973aa35fbdc4c333ea2748d0d61db97014
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 0.3466017405192057,
+    "train_runtime": 61.4918,
+    "train_samples": 635,
+    "train_samples_per_second": 516.329,
+    "train_steps_per_second": 4.066
+}

trainer_state.json

@@ -0,0 +1,205 @@
+{
+  "best_metric": 0.3455486595630646,
+  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_wnli_128/checkpoint-35",
+  "epoch": 12.0,
+  "global_step": 60,
+  "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": 0.3469,
+      "step": 5
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3455769717693329,
+      "eval_runtime": 0.1243,
+      "eval_samples_per_second": 571.247,
+      "eval_steps_per_second": 8.046,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3467,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3458103537559509,
+      "eval_runtime": 0.1242,
+      "eval_samples_per_second": 571.857,
+      "eval_steps_per_second": 8.054,
+      "step": 10
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.3466,
+      "step": 15
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34594178199768066,
+      "eval_runtime": 0.1244,
+      "eval_samples_per_second": 570.716,
+      "eval_steps_per_second": 8.038,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.3466,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3457295894622803,
+      "eval_runtime": 0.1244,
+      "eval_samples_per_second": 570.787,
+      "eval_steps_per_second": 8.039,
+      "step": 20
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3466,
+      "step": 25
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34556901454925537,
+      "eval_runtime": 0.1262,
+      "eval_samples_per_second": 562.776,
+      "eval_steps_per_second": 7.926,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3466,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3455694019794464,
+      "eval_runtime": 0.1236,
+      "eval_samples_per_second": 574.427,
+      "eval_steps_per_second": 8.091,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3466,
+      "step": 35
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3455486595630646,
+      "eval_runtime": 0.1248,
+      "eval_samples_per_second": 569.047,
+      "eval_steps_per_second": 8.015,
+      "step": 35
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.3466,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34575846791267395,
+      "eval_runtime": 0.1262,
+      "eval_samples_per_second": 562.639,
+      "eval_steps_per_second": 7.924,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3466,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.345662385225296,
+      "eval_runtime": 0.1238,
+      "eval_samples_per_second": 573.305,
+      "eval_steps_per_second": 8.075,
+      "step": 45
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.3465,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34582069516181946,
+      "eval_runtime": 0.1253,
+      "eval_samples_per_second": 566.847,
+      "eval_steps_per_second": 7.984,
+      "step": 50
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.3466,
+      "step": 55
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3460276126861572,
+      "eval_runtime": 0.1243,
+      "eval_samples_per_second": 571.348,
+      "eval_steps_per_second": 8.047,
+      "step": 55
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.3465,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3459770977497101,
+      "eval_runtime": 0.1258,
+      "eval_samples_per_second": 564.353,
+      "eval_steps_per_second": 7.949,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "step": 60,
+      "total_flos": 181909497839616.0,
+      "train_loss": 0.3466017405192057,
+      "train_runtime": 61.4918,
+      "train_samples_per_second": 516.329,
+      "train_steps_per_second": 4.066
+    }
+  ],
+  "max_steps": 250,
+  "num_train_epochs": 50,
+  "total_flos": 181909497839616.0,
+  "trial_name": null,
+  "trial_params": null
+}