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 MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6007131940906776
+      value: 0.5957078925956062
 ---
 
 <!-- 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_sa_GLUE_Experiment_mnli_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 MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9131
-- Accuracy: 0.6007
+- Loss: 0.8825
+- Accuracy: 0.5957
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 14.0,
+    "epoch_mm": 14.0,
+    "eval_accuracy": 0.6022414671421293,
+    "eval_accuracy_mm": 0.5957078925956062,
+    "eval_loss": 0.8776137828826904,
+    "eval_loss_mm": 0.8824665546417236,
+    "eval_runtime": 21.7552,
+    "eval_runtime_mm": 21.6962,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 451.156,
+    "eval_samples_per_second_mm": 453.166,
+    "eval_steps_per_second": 3.539,
+    "eval_steps_per_second_mm": 3.549,
+    "train_loss": 0.8308279570709018,
+    "train_runtime": 35893.6313,
+    "train_samples": 392702,
+    "train_samples_per_second": 547.036,
+    "train_steps_per_second": 4.274
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 14.0,
+    "epoch_mm": 14.0,
+    "eval_accuracy": 0.6022414671421293,
+    "eval_accuracy_mm": 0.5957078925956062,
+    "eval_loss": 0.8776137828826904,
+    "eval_loss_mm": 0.8824665546417236,
+    "eval_runtime": 21.7552,
+    "eval_runtime_mm": 21.6962,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 451.156,
+    "eval_samples_per_second_mm": 453.166,
+    "eval_steps_per_second": 3.539,
+    "eval_steps_per_second_mm": 3.549
+}

logs/events.out.tfevents.1674721449.garda.227848.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ac25b189d6f1c8196636a5d0d3b26ff651810e924b22a65b18522ea70dcf61e
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.8308279570709018,
+    "train_runtime": 35893.6313,
+    "train_samples": 392702,
+    "train_samples_per_second": 547.036,
+    "train_steps_per_second": 4.274
+}

trainer_state.json

@@ -0,0 +1,235 @@
+{
+  "best_metric": 0.8776137828826904,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_mnli_128/checkpoint-27612",
+  "epoch": 14.0,
+  "global_step": 42952,
+  "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": 1.0129,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5437595517065715,
+      "eval_loss": 0.9528754353523254,
+      "eval_runtime": 21.6888,
+      "eval_samples_per_second": 452.537,
+      "eval_steps_per_second": 3.55,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9284,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5593479368313805,
+      "eval_loss": 0.9265554547309875,
+      "eval_runtime": 21.6883,
+      "eval_samples_per_second": 452.548,
+      "eval_steps_per_second": 3.55,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8999,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5774834437086093,
+      "eval_loss": 0.9054868817329407,
+      "eval_runtime": 21.6795,
+      "eval_samples_per_second": 452.731,
+      "eval_steps_per_second": 3.552,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8803,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5854304635761589,
+      "eval_loss": 0.8951208591461182,
+      "eval_runtime": 21.7169,
+      "eval_samples_per_second": 451.952,
+      "eval_steps_per_second": 3.546,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.8637,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5885888945491594,
+      "eval_loss": 0.8991015553474426,
+      "eval_runtime": 21.7239,
+      "eval_samples_per_second": 451.806,
+      "eval_steps_per_second": 3.544,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.8472,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5913397860417728,
+      "eval_loss": 0.8906782269477844,
+      "eval_runtime": 21.7654,
+      "eval_samples_per_second": 450.945,
+      "eval_steps_per_second": 3.538,
+      "step": 18408
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.8309,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5927661742231278,
+      "eval_loss": 0.8940486907958984,
+      "eval_runtime": 21.665,
+      "eval_samples_per_second": 453.036,
+      "eval_steps_per_second": 3.554,
+      "step": 21476
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.814,
+      "step": 24544
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5987773815588385,
+      "eval_loss": 0.8879620432853699,
+      "eval_runtime": 21.7009,
+      "eval_samples_per_second": 452.286,
+      "eval_steps_per_second": 3.548,
+      "step": 24544
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.7988,
+      "step": 27612
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6022414671421293,
+      "eval_loss": 0.8776137828826904,
+      "eval_runtime": 21.7212,
+      "eval_samples_per_second": 451.863,
+      "eval_steps_per_second": 3.545,
+      "step": 27612
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.7825,
+      "step": 30680
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6022414671421293,
+      "eval_loss": 0.8958155512809753,
+      "eval_runtime": 21.7108,
+      "eval_samples_per_second": 452.08,
+      "eval_steps_per_second": 3.547,
+      "step": 30680
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.7662,
+      "step": 33748
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.6061130922058074,
+      "eval_loss": 0.8834749460220337,
+      "eval_runtime": 21.6874,
+      "eval_samples_per_second": 452.567,
+      "eval_steps_per_second": 3.55,
+      "step": 33748
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.7504,
+      "step": 36816
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.6040753948038716,
+      "eval_loss": 0.9004417657852173,
+      "eval_runtime": 21.7036,
+      "eval_samples_per_second": 452.229,
+      "eval_steps_per_second": 3.548,
+      "step": 36816
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.7359,
+      "step": 39884
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.6,
+      "eval_loss": 0.9251638650894165,
+      "eval_runtime": 21.7106,
+      "eval_samples_per_second": 452.083,
+      "eval_steps_per_second": 3.547,
+      "step": 39884
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7204,
+      "step": 42952
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.6007131940906776,
+      "eval_loss": 0.9130809903144836,
+      "eval_runtime": 21.7023,
+      "eval_samples_per_second": 452.257,
+      "eval_steps_per_second": 3.548,
+      "step": 42952
+    },
+    {
+      "epoch": 14.0,
+      "step": 42952,
+      "total_flos": 1.3103958963991347e+17,
+      "train_loss": 0.8308279570709018,
+      "train_runtime": 35893.6313,
+      "train_samples_per_second": 547.036,
+      "train_steps_per_second": 4.274
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 1.3103958963991347e+17,
+  "trial_name": null,
+  "trial_params": null
+}