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_sa_GLUE_Experiment_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: 0.6918
+- Loss: 0.6907
 - Accuracy: 0.5634
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.690684974193573,
+    "eval_runtime": 0.1748,
+    "eval_samples": 71,
+    "eval_samples_per_second": 406.198,
+    "eval_steps_per_second": 5.721,
+    "train_loss": 0.6931104858716329,
+    "train_runtime": 100.6647,
+    "train_samples": 635,
+    "train_samples_per_second": 315.404,
+    "train_steps_per_second": 2.483
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.690684974193573,
+    "eval_runtime": 0.1748,
+    "eval_samples": 71,
+    "eval_samples_per_second": 406.198,
+    "eval_steps_per_second": 5.721
+}

logs/events.out.tfevents.1674685327.garda.227848.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:738dc560d9e7d4e397af9adb4f7858d845fc5de9ec14399cb092558cebf40234
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 0.6931104858716329,
+    "train_runtime": 100.6647,
+    "train_samples": 635,
+    "train_samples_per_second": 315.404,
+    "train_steps_per_second": 2.483
+}

trainer_state.json

@@ -0,0 +1,205 @@
+{
+  "best_metric": 0.690684974193573,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_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": 0.6938,
+      "step": 5
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.69107985496521,
+      "eval_runtime": 0.1722,
+      "eval_samples_per_second": 412.297,
+      "eval_steps_per_second": 5.807,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6933,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6917158961296082,
+      "eval_runtime": 0.2039,
+      "eval_samples_per_second": 348.164,
+      "eval_steps_per_second": 4.904,
+      "step": 10
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6931,
+      "step": 15
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6920373439788818,
+      "eval_runtime": 0.1716,
+      "eval_samples_per_second": 413.718,
+      "eval_steps_per_second": 5.827,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.693,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6915233135223389,
+      "eval_runtime": 0.1715,
+      "eval_samples_per_second": 414.07,
+      "eval_steps_per_second": 5.832,
+      "step": 20
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.693,
+      "step": 25
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6910518407821655,
+      "eval_runtime": 0.2025,
+      "eval_samples_per_second": 350.629,
+      "eval_steps_per_second": 4.938,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.693,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6909132599830627,
+      "eval_runtime": 0.1719,
+      "eval_samples_per_second": 413.098,
+      "eval_steps_per_second": 5.818,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.693,
+      "step": 35
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.690684974193573,
+      "eval_runtime": 0.1714,
+      "eval_samples_per_second": 414.154,
+      "eval_steps_per_second": 5.833,
+      "step": 35
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.693,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6910773515701294,
+      "eval_runtime": 0.1726,
+      "eval_samples_per_second": 411.386,
+      "eval_steps_per_second": 5.794,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6931,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6907783150672913,
+      "eval_runtime": 0.173,
+      "eval_samples_per_second": 410.457,
+      "eval_steps_per_second": 5.781,
+      "step": 45
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.693,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6911693811416626,
+      "eval_runtime": 0.1716,
+      "eval_samples_per_second": 413.752,
+      "eval_steps_per_second": 5.827,
+      "step": 50
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.693,
+      "step": 55
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6917507648468018,
+      "eval_runtime": 0.1721,
+      "eval_samples_per_second": 412.624,
+      "eval_steps_per_second": 5.812,
+      "step": 55
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.693,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.691800057888031,
+      "eval_runtime": 0.1714,
+      "eval_samples_per_second": 414.313,
+      "eval_steps_per_second": 5.835,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "step": 60,
+      "total_flos": 181619612712960.0,
+      "train_loss": 0.6931104858716329,
+      "train_runtime": 100.6647,
+      "train_samples_per_second": 315.404,
+      "train_steps_per_second": 2.483
+    }
+  ],
+  "max_steps": 250,
+  "num_train_epochs": 50,
+  "total_flos": 181619612712960.0,
+  "trial_name": null,
+  "trial_params": null
+}