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 QNLI
       type: glue
       config: qnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.589602782354018
+      value: 0.6093721398498994
 ---
 
 <!-- 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_qnli
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7055
-- Accuracy: 0.5896
+- Loss: 0.6487
+- Accuracy: 0.6094
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.6086399414241259,
-    "eval_loss": 0.652438759803772,
-    "eval_runtime": 5.4307,
+    "eval_accuracy": 0.6093721398498994,
+    "eval_loss": 0.6487034559249878,
+    "eval_runtime": 8.8318,
     "eval_samples": 5463,
-    "eval_samples_per_second": 1005.942,
-    "eval_steps_per_second": 4.051,
-    "train_loss": 0.6110314930772948,
-    "train_runtime": 2217.6487,
+    "eval_samples_per_second": 618.561,
+    "eval_steps_per_second": 4.869,
+    "train_loss": 0.603873611681752,
+    "train_runtime": 4016.5926,
     "train_samples": 104743,
-    "train_samples_per_second": 2361.578,
-    "train_steps_per_second": 9.244
+    "train_samples_per_second": 1303.879,
+    "train_steps_per_second": 10.195
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.6086399414241259,
-    "eval_loss": 0.652438759803772,
-    "eval_runtime": 5.4307,
+    "eval_accuracy": 0.6093721398498994,
+    "eval_loss": 0.6487034559249878,
+    "eval_runtime": 8.8318,
     "eval_samples": 5463,
-    "eval_samples_per_second": 1005.942,
-    "eval_steps_per_second": 4.051
+    "eval_samples_per_second": 618.561,
+    "eval_steps_per_second": 4.869
 }

logs/events.out.tfevents.1674624040.garda.1933772.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22f9ee41b991603e11e8e06c11c3ea53371f180b56206617bb3f628c3f8ac999
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.0,
-    "train_loss": 0.6110314930772948,
-    "train_runtime": 2217.6487,
+    "train_loss": 0.603873611681752,
+    "train_runtime": 4016.5926,
     "train_samples": 104743,
-    "train_samples_per_second": 2361.578,
-    "train_steps_per_second": 9.244
+    "train_samples_per_second": 1303.879,
+    "train_steps_per_second": 10.195
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.652438759803772,
-  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_qnli/checkpoint-820",
+  "best_metric": 0.6487034559249878,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_qnli/checkpoint-1638",
   "epoch": 7.0,
-  "global_step": 2870,
+  "global_step": 5733,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -10,121 +10,121 @@
     {
       "epoch": 1.0,
       "learning_rate": 4.9e-05,
-      "loss": 0.6824,
-      "step": 410
+      "loss": 0.6754,
+      "step": 819
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5996705107084019,
-      "eval_loss": 0.6577625870704651,
-      "eval_runtime": 5.6422,
-      "eval_samples_per_second": 968.24,
-      "eval_steps_per_second": 3.899,
-      "step": 410
+      "eval_accuracy": 0.6177924217462932,
+      "eval_loss": 0.6491163372993469,
+      "eval_runtime": 8.8255,
+      "eval_samples_per_second": 619.001,
+      "eval_steps_per_second": 4.872,
+      "step": 819
     },
     {
       "epoch": 2.0,
       "learning_rate": 4.8e-05,
-      "loss": 0.6441,
-      "step": 820
+      "loss": 0.6369,
+      "step": 1638
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6086399414241259,
-      "eval_loss": 0.652438759803772,
-      "eval_runtime": 5.5355,
-      "eval_samples_per_second": 986.904,
-      "eval_steps_per_second": 3.974,
-      "step": 820
+      "eval_accuracy": 0.6093721398498994,
+      "eval_loss": 0.6487034559249878,
+      "eval_runtime": 8.8229,
+      "eval_samples_per_second": 619.186,
+      "eval_steps_per_second": 4.874,
+      "step": 1638
     },
     {
       "epoch": 3.0,
-      "learning_rate": 4.7e-05,
-      "loss": 0.6202,
-      "step": 1230
+      "learning_rate": 4.7001221001221e-05,
+      "loss": 0.6125,
+      "step": 2457
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6071755445725792,
-      "eval_loss": 0.6553815603256226,
-      "eval_runtime": 5.4718,
-      "eval_samples_per_second": 998.389,
-      "eval_steps_per_second": 4.021,
-      "step": 1230
+      "eval_accuracy": 0.6088229910305692,
+      "eval_loss": 0.6555132269859314,
+      "eval_runtime": 8.8047,
+      "eval_samples_per_second": 620.464,
+      "eval_steps_per_second": 4.884,
+      "step": 2457
     },
     {
       "epoch": 4.0,
-      "learning_rate": 4.600000000000001e-05,
-      "loss": 0.6009,
-      "step": 1640
+      "learning_rate": 4.6001221001221e-05,
+      "loss": 0.5942,
+      "step": 3276
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6051619989017024,
-      "eval_loss": 0.6619159579277039,
-      "eval_runtime": 5.4488,
-      "eval_samples_per_second": 1002.613,
-      "eval_steps_per_second": 4.038,
-      "step": 1640
+      "eval_accuracy": 0.6027823540179389,
+      "eval_loss": 0.6647323369979858,
+      "eval_runtime": 8.7921,
+      "eval_samples_per_second": 621.356,
+      "eval_steps_per_second": 4.891,
+      "step": 3276
     },
     {
       "epoch": 5.0,
-      "learning_rate": 4.5e-05,
-      "loss": 0.587,
-      "step": 2050
+      "learning_rate": 4.5001221001221004e-05,
+      "loss": 0.5805,
+      "step": 4095
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.598572213069742,
-      "eval_loss": 0.668364405632019,
-      "eval_runtime": 5.5979,
-      "eval_samples_per_second": 975.905,
-      "eval_steps_per_second": 3.93,
-      "step": 2050
+      "eval_accuracy": 0.5934468240893283,
+      "eval_loss": 0.6735221147537231,
+      "eval_runtime": 8.6724,
+      "eval_samples_per_second": 629.928,
+      "eval_steps_per_second": 4.958,
+      "step": 4095
     },
     {
       "epoch": 6.0,
-      "learning_rate": 4.400243902439024e-05,
-      "loss": 0.5755,
-      "step": 2460
+      "learning_rate": 4.400244200244201e-05,
+      "loss": 0.5689,
+      "step": 4914
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.5978400146439685,
-      "eval_loss": 0.680844783782959,
-      "eval_runtime": 5.501,
-      "eval_samples_per_second": 993.091,
-      "eval_steps_per_second": 3.999,
-      "step": 2460
+      "eval_loss": 0.6893225312232971,
+      "eval_runtime": 8.8418,
+      "eval_samples_per_second": 617.861,
+      "eval_steps_per_second": 4.863,
+      "step": 4914
     },
     {
       "epoch": 7.0,
-      "learning_rate": 4.3002439024390246e-05,
-      "loss": 0.5671,
-      "step": 2870
+      "learning_rate": 4.3002442002442004e-05,
+      "loss": 0.5587,
+      "step": 5733
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.5844773933736043,
-      "eval_loss": 0.7068149447441101,
-      "eval_runtime": 5.4741,
-      "eval_samples_per_second": 997.977,
-      "eval_steps_per_second": 4.019,
-      "step": 2870
+      "eval_accuracy": 0.589602782354018,
+      "eval_loss": 0.7054734230041504,
+      "eval_runtime": 8.8696,
+      "eval_samples_per_second": 615.924,
+      "eval_steps_per_second": 4.848,
+      "step": 5733
     },
     {
       "epoch": 7.0,
-      "step": 2870,
-      "total_flos": 2.328477811133645e+16,
-      "train_loss": 0.6110314930772948,
-      "train_runtime": 2217.6487,
-      "train_samples_per_second": 2361.578,
-      "train_steps_per_second": 9.244
+      "step": 5733,
+      "total_flos": 2.2988974586855424e+16,
+      "train_loss": 0.603873611681752,
+      "train_runtime": 4016.5926,
+      "train_samples_per_second": 1303.879,
+      "train_steps_per_second": 10.195
     }
   ],
-  "max_steps": 20500,
+  "max_steps": 40950,
   "num_train_epochs": 50,
-  "total_flos": 2.328477811133645e+16,
+  "total_flos": 2.2988974586855424e+16,
   "trial_name": null,
   "trial_params": null
 }