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: 0.3273560876209883
+      value: 0.3295362082994304
 ---
 
 <!-- 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_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: 1.7940
-- Accuracy: 0.3274
+- Loss: 1.7834
+- Accuracy: 0.3295
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.3273560876209883,
+    "eval_accuracy_mm": 0.3295362082994304,
+    "eval_loss": 1.793838381767273,
+    "eval_loss_mm": 1.7834389209747314,
+    "eval_runtime": 16.1504,
+    "eval_runtime_mm": 16.1104,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 607.724,
+    "eval_samples_per_second_mm": 610.287,
+    "eval_steps_per_second": 4.768,
+    "eval_steps_per_second_mm": 4.78,
+    "train_loss": 1.886379419995722,
+    "train_runtime": 15224.9542,
+    "train_samples": 392702,
+    "train_samples_per_second": 1289.666,
+    "train_steps_per_second": 10.076
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.3273560876209883,
+    "eval_accuracy_mm": 0.3295362082994304,
+    "eval_loss": 1.793838381767273,
+    "eval_loss_mm": 1.7834389209747314,
+    "eval_runtime": 16.1504,
+    "eval_runtime_mm": 16.1104,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 607.724,
+    "eval_samples_per_second_mm": 610.287,
+    "eval_steps_per_second": 4.768,
+    "eval_steps_per_second_mm": 4.78
+}

logs/events.out.tfevents.1674991288.serv-3334.518319.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1797654d3592fe95251f54d2647d80144e62f48722fec8385434a0759a2c0455
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 1.886379419995722,
+    "train_runtime": 15224.9542,
+    "train_samples": 392702,
+    "train_samples_per_second": 1289.666,
+    "train_steps_per_second": 10.076
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": 1.793838381767273,
+  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_mnli_128/checkpoint-15340",
+  "epoch": 10.0,
+  "global_step": 30680,
+  "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": 1.8865,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.793958306312561,
+      "eval_runtime": 15.8224,
+      "eval_samples_per_second": 620.322,
+      "eval_steps_per_second": 4.867,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 1.8864,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7939366102218628,
+      "eval_runtime": 15.837,
+      "eval_samples_per_second": 619.752,
+      "eval_steps_per_second": 4.862,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 1.8864,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7943456172943115,
+      "eval_runtime": 16.027,
+      "eval_samples_per_second": 612.403,
+      "eval_steps_per_second": 4.804,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 1.8864,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7939870357513428,
+      "eval_runtime": 15.825,
+      "eval_samples_per_second": 620.221,
+      "eval_steps_per_second": 4.866,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 1.8864,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.793838381767273,
+      "eval_runtime": 15.8831,
+      "eval_samples_per_second": 617.951,
+      "eval_steps_per_second": 4.848,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 1.8864,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7939422130584717,
+      "eval_runtime": 15.8814,
+      "eval_samples_per_second": 618.02,
+      "eval_steps_per_second": 4.848,
+      "step": 18408
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 1.8864,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7942163944244385,
+      "eval_runtime": 16.1327,
+      "eval_samples_per_second": 608.393,
+      "eval_steps_per_second": 4.773,
+      "step": 21476
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 1.8864,
+      "step": 24544
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7938553094863892,
+      "eval_runtime": 16.1396,
+      "eval_samples_per_second": 608.131,
+      "eval_steps_per_second": 4.771,
+      "step": 24544
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 1.8864,
+      "step": 27612
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.793870210647583,
+      "eval_runtime": 16.1202,
+      "eval_samples_per_second": 608.865,
+      "eval_steps_per_second": 4.777,
+      "step": 27612
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 1.8863,
+      "step": 30680
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.7940114736557007,
+      "eval_runtime": 16.1365,
+      "eval_samples_per_second": 608.249,
+      "eval_steps_per_second": 4.772,
+      "step": 30680
+    },
+    {
+      "epoch": 10.0,
+      "step": 30680,
+      "total_flos": 9.374910732304384e+16,
+      "train_loss": 1.886379419995722,
+      "train_runtime": 15224.9542,
+      "train_samples_per_second": 1289.666,
+      "train_steps_per_second": 10.076
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 9.374910732304384e+16,
+  "trial_name": null,
+  "trial_params": null
+}