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: 0.9208715596330275
+      value: 0.926605504587156
 ---
 
 <!-- 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_logit_kd_pretrain_sst2
 
-This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the glue dataset.
+This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the GLUE SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2677
-- Accuracy: 0.9209
+- Loss: 0.2364
+- Accuracy: 0.9266
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 23.0,
+    "eval_accuracy": 0.926605504587156,
+    "eval_loss": 0.2364484816789627,
+    "eval_runtime": 1.6442,
+    "eval_samples": 872,
+    "eval_samples_per_second": 530.345,
+    "eval_steps_per_second": 4.257,
+    "train_loss": 0.06443861390230197,
+    "train_runtime": 7079.7355,
+    "train_samples": 67349,
+    "train_samples_per_second": 475.646,
+    "train_steps_per_second": 3.722
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 23.0,
+    "eval_accuracy": 0.926605504587156,
+    "eval_loss": 0.2364484816789627,
+    "eval_runtime": 1.6442,
+    "eval_samples": 872,
+    "eval_samples_per_second": 530.345,
+    "eval_steps_per_second": 4.257
+}

logs/events.out.tfevents.1675095454.serv-3334.1119753.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5123b266180743d5cdbaaca07bb07b6fd2976f20e0780b8b731f2c548c3475bb
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 23.0,
+    "train_loss": 0.06443861390230197,
+    "train_runtime": 7079.7355,
+    "train_samples": 67349,
+    "train_samples_per_second": 475.646,
+    "train_steps_per_second": 3.722
+}

trainer_state.json

@@ -0,0 +1,370 @@
+{
+  "best_metric": 0.2364484816789627,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_sst2/checkpoint-9486",
+  "epoch": 23.0,
+  "global_step": 12121,
+  "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": 0.4176,
+      "step": 527
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9197247706422018,
+      "eval_loss": 0.29781466722488403,
+      "eval_runtime": 1.6438,
+      "eval_samples_per_second": 530.48,
+      "eval_steps_per_second": 4.258,
+      "step": 527
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.1807,
+      "step": 1054
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9174311926605505,
+      "eval_loss": 0.2951326370239258,
+      "eval_runtime": 1.6349,
+      "eval_samples_per_second": 533.377,
+      "eval_steps_per_second": 4.282,
+      "step": 1054
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.1163,
+      "step": 1581
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9185779816513762,
+      "eval_loss": 0.2749492824077606,
+      "eval_runtime": 1.6378,
+      "eval_samples_per_second": 532.414,
+      "eval_steps_per_second": 4.274,
+      "step": 1581
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.0862,
+      "step": 2108
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.908256880733945,
+      "eval_loss": 0.2988210916519165,
+      "eval_runtime": 1.6403,
+      "eval_samples_per_second": 531.618,
+      "eval_steps_per_second": 4.268,
+      "step": 2108
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.0695,
+      "step": 2635
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9174311926605505,
+      "eval_loss": 0.2760469913482666,
+      "eval_runtime": 1.6372,
+      "eval_samples_per_second": 532.603,
+      "eval_steps_per_second": 4.275,
+      "step": 2635
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.0598,
+      "step": 3162
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9151376146788991,
+      "eval_loss": 0.26954537630081177,
+      "eval_runtime": 1.636,
+      "eval_samples_per_second": 532.996,
+      "eval_steps_per_second": 4.279,
+      "step": 3162
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.0525,
+      "step": 3689
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9254587155963303,
+      "eval_loss": 0.27225229144096375,
+      "eval_runtime": 1.633,
+      "eval_samples_per_second": 533.988,
+      "eval_steps_per_second": 4.287,
+      "step": 3689
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0464,
+      "step": 4216
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9243119266055045,
+      "eval_loss": 0.24295173585414886,
+      "eval_runtime": 1.6291,
+      "eval_samples_per_second": 535.278,
+      "eval_steps_per_second": 4.297,
+      "step": 4216
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.0422,
+      "step": 4743
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9243119266055045,
+      "eval_loss": 0.2813539206981659,
+      "eval_runtime": 1.6396,
+      "eval_samples_per_second": 531.844,
+      "eval_steps_per_second": 4.269,
+      "step": 4743
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.0395,
+      "step": 5270
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9162844036697247,
+      "eval_loss": 0.24639670550823212,
+      "eval_runtime": 1.6343,
+      "eval_samples_per_second": 533.555,
+      "eval_steps_per_second": 4.283,
+      "step": 5270
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.0357,
+      "step": 5797
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.9197247706422018,
+      "eval_loss": 0.23897552490234375,
+      "eval_runtime": 1.6396,
+      "eval_samples_per_second": 531.822,
+      "eval_steps_per_second": 4.269,
+      "step": 5797
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.0341,
+      "step": 6324
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.9197247706422018,
+      "eval_loss": 0.271278440952301,
+      "eval_runtime": 1.6863,
+      "eval_samples_per_second": 517.109,
+      "eval_steps_per_second": 4.151,
+      "step": 6324
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.0328,
+      "step": 6851
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.9220183486238532,
+      "eval_loss": 0.2685054540634155,
+      "eval_runtime": 1.6377,
+      "eval_samples_per_second": 532.44,
+      "eval_steps_per_second": 4.274,
+      "step": 6851
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.0315,
+      "step": 7378
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.9185779816513762,
+      "eval_loss": 0.2584623694419861,
+      "eval_runtime": 1.6363,
+      "eval_samples_per_second": 532.924,
+      "eval_steps_per_second": 4.278,
+      "step": 7378
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 3.5e-05,
+      "loss": 0.0296,
+      "step": 7905
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.9220183486238532,
+      "eval_loss": 0.23671826720237732,
+      "eval_runtime": 1.6376,
+      "eval_samples_per_second": 532.47,
+      "eval_steps_per_second": 4.274,
+      "step": 7905
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.0283,
+      "step": 8432
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.9185779816513762,
+      "eval_loss": 0.25598371028900146,
+      "eval_runtime": 1.6346,
+      "eval_samples_per_second": 533.461,
+      "eval_steps_per_second": 4.282,
+      "step": 8432
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.3e-05,
+      "loss": 0.0277,
+      "step": 8959
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.9174311926605505,
+      "eval_loss": 0.2635485529899597,
+      "eval_runtime": 1.6357,
+      "eval_samples_per_second": 533.119,
+      "eval_steps_per_second": 4.28,
+      "step": 8959
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.0269,
+      "step": 9486
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.926605504587156,
+      "eval_loss": 0.2364484816789627,
+      "eval_runtime": 1.6377,
+      "eval_samples_per_second": 532.449,
+      "eval_steps_per_second": 4.274,
+      "step": 9486
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 3.1e-05,
+      "loss": 0.026,
+      "step": 10013
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.9208715596330275,
+      "eval_loss": 0.2748643755912781,
+      "eval_runtime": 1.6351,
+      "eval_samples_per_second": 533.307,
+      "eval_steps_per_second": 4.281,
+      "step": 10013
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 3e-05,
+      "loss": 0.0252,
+      "step": 10540
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.9174311926605505,
+      "eval_loss": 0.2506615221500397,
+      "eval_runtime": 1.633,
+      "eval_samples_per_second": 533.976,
+      "eval_steps_per_second": 4.287,
+      "step": 10540
+    },
+    {
+      "epoch": 21.0,
+      "learning_rate": 2.9e-05,
+      "loss": 0.0248,
+      "step": 11067
+    },
+    {
+      "epoch": 21.0,
+      "eval_accuracy": 0.9162844036697247,
+      "eval_loss": 0.2768906354904175,
+      "eval_runtime": 1.6344,
+      "eval_samples_per_second": 533.527,
+      "eval_steps_per_second": 4.283,
+      "step": 11067
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.0248,
+      "step": 11594
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.9220183486238532,
+      "eval_loss": 0.25428006052970886,
+      "eval_runtime": 1.6413,
+      "eval_samples_per_second": 531.296,
+      "eval_steps_per_second": 4.265,
+      "step": 11594
+    },
+    {
+      "epoch": 23.0,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.024,
+      "step": 12121
+    },
+    {
+      "epoch": 23.0,
+      "eval_accuracy": 0.9208715596330275,
+      "eval_loss": 0.2676636278629303,
+      "eval_runtime": 1.6366,
+      "eval_samples_per_second": 532.827,
+      "eval_steps_per_second": 4.277,
+      "step": 12121
+    },
+    {
+      "epoch": 23.0,
+      "step": 12121,
+      "total_flos": 4.919353360882074e+16,
+      "train_loss": 0.06443861390230197,
+      "train_runtime": 7079.7355,
+      "train_samples_per_second": 475.646,
+      "train_steps_per_second": 3.722
+    }
+  ],
+  "max_steps": 26350,
+  "num_train_epochs": 50,
+  "total_flos": 4.919353360882074e+16,
+  "trial_name": null,
+  "trial_params": null
+}