End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -14,7 +16,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -22,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7034313725490197
+      value: 0.6838235294117647
     - name: F1
       type: f1
-      value: 0.8032520325203252
+      value: 0.8122270742358079
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,12 +35,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mobilebert_sa_GLUE_Experiment_mrpc
 
-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 MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9829
-- Accuracy: 0.7034
-- F1: 0.8033
-- Combined Score: 0.7533
+- Loss: 0.6145
+- Accuracy: 0.6838
+- F1: 0.8122
+- Combined Score: 0.7480
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 16.0,
-    "eval_accuracy": 0.6617647058823529,
-    "eval_combined_score": 0.7139332003988036,
-    "eval_f1": 0.7661016949152543,
-    "eval_loss": 0.6120356917381287,
-    "eval_runtime": 0.4331,
+    "epoch": 10.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 0.614494800567627,
+    "eval_runtime": 0.7302,
     "eval_samples": 408,
-    "eval_samples_per_second": 942.046,
-    "eval_steps_per_second": 4.618,
-    "train_loss": 0.555461339155833,
-    "train_runtime": 276.9705,
+    "eval_samples_per_second": 558.746,
+    "eval_steps_per_second": 5.478,
+    "train_loss": 0.5173478488264412,
+    "train_runtime": 338.752,
     "train_samples": 3668,
-    "train_samples_per_second": 662.164,
-    "train_steps_per_second": 2.708
+    "train_samples_per_second": 541.399,
+    "train_steps_per_second": 4.28
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 16.0,
-    "eval_accuracy": 0.6617647058823529,
-    "eval_combined_score": 0.7139332003988036,
-    "eval_f1": 0.7661016949152543,
-    "eval_loss": 0.6120356917381287,
-    "eval_runtime": 0.4331,
+    "epoch": 10.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 0.614494800567627,
+    "eval_runtime": 0.7302,
     "eval_samples": 408,
-    "eval_samples_per_second": 942.046,
-    "eval_steps_per_second": 4.618
+    "eval_samples_per_second": 558.746,
+    "eval_steps_per_second": 5.478
 }

logs/events.out.tfevents.1674619910.garda.1933772.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57035a7fbfa5c7c7cf6f89b1ff21360913b2ea3bb30d5bbcea8efb074d9bc888
+size 467

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 16.0,
-    "train_loss": 0.555461339155833,
-    "train_runtime": 276.9705,
+    "epoch": 10.0,
+    "train_loss": 0.5173478488264412,
+    "train_runtime": 338.752,
     "train_samples": 3668,
-    "train_samples_per_second": 662.164,
-    "train_steps_per_second": 2.708
+    "train_samples_per_second": 541.399,
+    "train_steps_per_second": 4.28
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.6120356917381287,
-  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_mrpc/checkpoint-165",
-  "epoch": 16.0,
-  "global_step": 240,
+  "best_metric": 0.614494800567627,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_mrpc/checkpoint-145",
+  "epoch": 10.0,
+  "global_step": 290,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -10,288 +10,186 @@
     {
       "epoch": 1.0,
       "learning_rate": 4.9e-05,
-      "loss": 0.6462,
-      "step": 15
+      "loss": 0.6377,
+      "step": 29
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6247462034225464,
-      "eval_runtime": 0.416,
-      "eval_samples_per_second": 980.655,
-      "eval_steps_per_second": 4.807,
-      "step": 15
+      "eval_loss": 0.6240395903587341,
+      "eval_runtime": 0.7115,
+      "eval_samples_per_second": 573.413,
+      "eval_steps_per_second": 5.622,
+      "step": 29
     },
     {
       "epoch": 2.0,
       "learning_rate": 4.8e-05,
-      "loss": 0.6292,
-      "step": 30
+      "loss": 0.6309,
+      "step": 58
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6235859990119934,
-      "eval_runtime": 0.4143,
-      "eval_samples_per_second": 984.679,
-      "eval_steps_per_second": 4.827,
-      "step": 30
+      "eval_loss": 0.6235603094100952,
+      "eval_runtime": 0.7114,
+      "eval_samples_per_second": 573.506,
+      "eval_steps_per_second": 5.623,
+      "step": 58
     },
     {
       "epoch": 3.0,
       "learning_rate": 4.7e-05,
-      "loss": 0.6312,
-      "step": 45
+      "loss": 0.6306,
+      "step": 87
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6266204118728638,
-      "eval_runtime": 0.4139,
-      "eval_samples_per_second": 985.824,
-      "eval_steps_per_second": 4.832,
-      "step": 45
+      "eval_loss": 0.6232509016990662,
+      "eval_runtime": 0.716,
+      "eval_samples_per_second": 569.871,
+      "eval_steps_per_second": 5.587,
+      "step": 87
     },
     {
       "epoch": 4.0,
       "learning_rate": 4.600000000000001e-05,
       "loss": 0.6291,
-      "step": 60
+      "step": 116
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6229846477508545,
-      "eval_runtime": 0.4134,
-      "eval_samples_per_second": 987.054,
-      "eval_steps_per_second": 4.839,
-      "step": 60
+      "eval_loss": 0.6225568056106567,
+      "eval_runtime": 0.7145,
+      "eval_samples_per_second": 571.018,
+      "eval_steps_per_second": 5.598,
+      "step": 116
     },
     {
       "epoch": 5.0,
       "learning_rate": 4.5e-05,
-      "loss": 0.6306,
-      "step": 75
+      "loss": 0.6222,
+      "step": 145
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.622560977935791,
-      "eval_runtime": 0.4291,
-      "eval_samples_per_second": 950.747,
-      "eval_steps_per_second": 4.661,
-      "step": 75
+      "eval_loss": 0.614494800567627,
+      "eval_runtime": 0.7135,
+      "eval_samples_per_second": 571.84,
+      "eval_steps_per_second": 5.606,
+      "step": 145
     },
     {
       "epoch": 6.0,
       "learning_rate": 4.4000000000000006e-05,
-      "loss": 0.6265,
-      "step": 90
+      "loss": 0.5736,
+      "step": 174
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6228667497634888,
-      "eval_runtime": 0.4149,
-      "eval_samples_per_second": 983.438,
-      "eval_steps_per_second": 4.821,
-      "step": 90
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7474496555378909,
+      "eval_f1": 0.793918918918919,
+      "eval_loss": 0.620771586894989,
+      "eval_runtime": 0.7131,
+      "eval_samples_per_second": 572.122,
+      "eval_steps_per_second": 5.609,
+      "step": 174
     },
     {
       "epoch": 7.0,
       "learning_rate": 4.3e-05,
-      "loss": 0.6287,
-      "step": 105
+      "loss": 0.488,
+      "step": 203
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6208460330963135,
-      "eval_runtime": 0.5134,
-      "eval_samples_per_second": 794.775,
-      "eval_steps_per_second": 3.896,
-      "step": 105
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7365844485942525,
+      "eval_f1": 0.7795414462081129,
+      "eval_loss": 0.6414217948913574,
+      "eval_runtime": 0.7129,
+      "eval_samples_per_second": 572.297,
+      "eval_steps_per_second": 5.611,
+      "step": 203
     },
     {
       "epoch": 8.0,
       "learning_rate": 4.2e-05,
-      "loss": 0.6215,
-      "step": 120
+      "loss": 0.3939,
+      "step": 232
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.621772289276123,
-      "eval_runtime": 0.4654,
-      "eval_samples_per_second": 876.685,
-      "eval_steps_per_second": 4.297,
-      "step": 120
+      "eval_accuracy": 0.7279411764705882,
+      "eval_combined_score": 0.7700619587936697,
+      "eval_f1": 0.8121827411167513,
+      "eval_loss": 0.7659339308738708,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 569.346,
+      "eval_steps_per_second": 5.582,
+      "step": 232
     },
     {
       "epoch": 9.0,
-      "learning_rate": 4.1e-05,
-      "loss": 0.6117,
-      "step": 135
+      "learning_rate": 4.103448275862069e-05,
+      "loss": 0.3038,
+      "step": 261
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.7009803921568627,
-      "eval_combined_score": 0.7591728308089702,
-      "eval_f1": 0.8173652694610778,
-      "eval_loss": 0.6204394698143005,
-      "eval_runtime": 0.4684,
-      "eval_samples_per_second": 871.136,
-      "eval_steps_per_second": 4.27,
-      "step": 135
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7554933665008292,
+      "eval_f1": 0.8026533996683249,
+      "eval_loss": 0.8875442147254944,
+      "eval_runtime": 0.7168,
+      "eval_samples_per_second": 569.19,
+      "eval_steps_per_second": 5.58,
+      "step": 261
     },
     {
       "epoch": 10.0,
-      "learning_rate": 4e-05,
-      "loss": 0.592,
-      "step": 150
+      "learning_rate": 4.003448275862069e-05,
+      "loss": 0.2636,
+      "step": 290
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.7009803921568627,
-      "eval_combined_score": 0.7588986044868398,
-      "eval_f1": 0.8168168168168168,
-      "eval_loss": 0.6202014088630676,
-      "eval_runtime": 0.4641,
-      "eval_samples_per_second": 879.144,
-      "eval_steps_per_second": 4.31,
-      "step": 150
-    },
-    {
-      "epoch": 11.0,
-      "learning_rate": 3.9000000000000006e-05,
-      "loss": 0.5599,
-      "step": 165
-    },
-    {
-      "epoch": 11.0,
-      "eval_accuracy": 0.6617647058823529,
-      "eval_combined_score": 0.7139332003988036,
-      "eval_f1": 0.7661016949152543,
-      "eval_loss": 0.6120356917381287,
-      "eval_runtime": 0.4633,
-      "eval_samples_per_second": 880.609,
-      "eval_steps_per_second": 4.317,
-      "step": 165
-    },
-    {
-      "epoch": 12.0,
-      "learning_rate": 3.8e-05,
-      "loss": 0.5079,
-      "step": 180
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7533417025346725,
+      "eval_f1": 0.8032520325203252,
+      "eval_loss": 0.9829146265983582,
+      "eval_runtime": 0.7178,
+      "eval_samples_per_second": 568.4,
+      "eval_steps_per_second": 5.573,
+      "step": 290
     },
     {
-      "epoch": 12.0,
-      "eval_accuracy": 0.696078431372549,
-      "eval_combined_score": 0.7453902090637579,
-      "eval_f1": 0.7947019867549668,
-      "eval_loss": 0.6299110651016235,
-      "eval_runtime": 0.4656,
-      "eval_samples_per_second": 876.31,
-      "eval_steps_per_second": 4.296,
-      "step": 180
-    },
-    {
-      "epoch": 13.0,
-      "learning_rate": 3.7e-05,
-      "loss": 0.4459,
-      "step": 195
-    },
-    {
-      "epoch": 13.0,
-      "eval_accuracy": 0.7009803921568627,
-      "eval_combined_score": 0.7501612487100103,
-      "eval_f1": 0.799342105263158,
-      "eval_loss": 0.767034649848938,
-      "eval_runtime": 0.4718,
-      "eval_samples_per_second": 864.746,
-      "eval_steps_per_second": 4.239,
-      "step": 195
-    },
-    {
-      "epoch": 14.0,
-      "learning_rate": 3.606666666666667e-05,
-      "loss": 0.4334,
-      "step": 210
-    },
-    {
-      "epoch": 14.0,
-      "eval_accuracy": 0.7083333333333334,
-      "eval_combined_score": 0.7592703349282297,
-      "eval_f1": 0.810207336523126,
-      "eval_loss": 0.8543146848678589,
-      "eval_runtime": 0.4957,
-      "eval_samples_per_second": 823.012,
-      "eval_steps_per_second": 4.034,
-      "step": 210
-    },
-    {
-      "epoch": 15.0,
-      "learning_rate": 3.513333333333334e-05,
-      "loss": 0.3764,
-      "step": 225
-    },
-    {
-      "epoch": 15.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_combined_score": 0.7131919905771495,
-      "eval_f1": 0.7597173144876325,
-      "eval_loss": 0.7447585463523865,
-      "eval_runtime": 0.502,
-      "eval_samples_per_second": 812.78,
-      "eval_steps_per_second": 3.984,
-      "step": 225
-    },
-    {
-      "epoch": 16.0,
-      "learning_rate": 3.4133333333333334e-05,
-      "loss": 0.3172,
-      "step": 240
-    },
-    {
-      "epoch": 16.0,
-      "eval_accuracy": 0.7058823529411765,
-      "eval_combined_score": 0.7529411764705882,
-      "eval_f1": 0.7999999999999999,
-      "eval_loss": 0.8614088296890259,
-      "eval_runtime": 0.5042,
-      "eval_samples_per_second": 809.272,
-      "eval_steps_per_second": 3.967,
-      "step": 240
-    },
-    {
-      "epoch": 16.0,
-      "step": 240,
-      "total_flos": 1863795957825536.0,
-      "train_loss": 0.555461339155833,
-      "train_runtime": 276.9705,
-      "train_samples_per_second": 662.164,
-      "train_steps_per_second": 2.708
+      "epoch": 10.0,
+      "step": 290,
+      "total_flos": 1150074298040320.0,
+      "train_loss": 0.5173478488264412,
+      "train_runtime": 338.752,
+      "train_samples_per_second": 541.399,
+      "train_steps_per_second": 4.28
     }
   ],
-  "max_steps": 750,
+  "max_steps": 1450,
   "num_train_epochs": 50,
-  "total_flos": 1863795957825536.0,
+  "total_flos": 1150074298040320.0,
   "trial_name": null,
   "trial_params": null
 }