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 QQP
       type: glue
       config: qqp
       split: validation
@@ -22,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7844422458570368
+      value: 0.7784071234232006
     - name: F1
       type: f1
-      value: 0.7043256997455472
+      value: 0.6885884111369878
 ---
 
 <!-- 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_qqp_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 QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5092
-- Accuracy: 0.7844
-- F1: 0.7043
-- Combined Score: 0.7444
+- Loss: 0.4700
+- Accuracy: 0.7784
+- F1: 0.6886
+- Combined Score: 0.7335
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.7784071234232006,
+    "eval_combined_score": 0.7334977672800942,
+    "eval_f1": 0.6885884111369878,
+    "eval_loss": 0.4699857532978058,
+    "eval_runtime": 89.3651,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 452.414,
+    "eval_steps_per_second": 3.536,
+    "train_loss": 0.4099273109145315,
+    "train_runtime": 29349.5289,
+    "train_samples": 363846,
+    "train_samples_per_second": 619.85,
+    "train_steps_per_second": 4.843
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.7784071234232006,
+    "eval_combined_score": 0.7334977672800942,
+    "eval_f1": 0.6885884111369878,
+    "eval_loss": 0.4699857532978058,
+    "eval_runtime": 89.3651,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 452.414,
+    "eval_steps_per_second": 3.536
+}

logs/events.out.tfevents.1674681450.garda.227848.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2631ad58e7f255377a417f87fdf03e06d0a479485b5a849f5668797cc19626da
+size 475

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 0.4099273109145315,
+    "train_runtime": 29349.5289,
+    "train_samples": 363846,
+    "train_samples_per_second": 619.85,
+    "train_steps_per_second": 4.843
+}

trainer_state.json

@@ -0,0 +1,229 @@
+{
+  "best_metric": 0.4699857532978058,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_qqp_128/checkpoint-19901",
+  "epoch": 12.0,
+  "global_step": 34116,
+  "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": 0.5294,
+      "step": 2843
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7512490724709374,
+      "eval_combined_score": 0.7074078968884644,
+      "eval_f1": 0.6635667213059914,
+      "eval_loss": 0.5076429843902588,
+      "eval_runtime": 89.4131,
+      "eval_samples_per_second": 452.171,
+      "eval_steps_per_second": 3.534,
+      "step": 2843
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.4791,
+      "step": 5686
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7613405886717783,
+      "eval_combined_score": 0.6991138272452333,
+      "eval_f1": 0.6368870658186881,
+      "eval_loss": 0.4889410436153412,
+      "eval_runtime": 89.507,
+      "eval_samples_per_second": 451.696,
+      "eval_steps_per_second": 3.53,
+      "step": 5686
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.4622,
+      "step": 8529
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7656690576304724,
+      "eval_combined_score": 0.7065937370896614,
+      "eval_f1": 0.6475184165488505,
+      "eval_loss": 0.48212409019470215,
+      "eval_runtime": 89.4828,
+      "eval_samples_per_second": 451.819,
+      "eval_steps_per_second": 3.531,
+      "step": 8529
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.4463,
+      "step": 11372
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7693544397724462,
+      "eval_combined_score": 0.72116678678084,
+      "eval_f1": 0.6729791337892337,
+      "eval_loss": 0.48311617970466614,
+      "eval_runtime": 89.5837,
+      "eval_samples_per_second": 451.31,
+      "eval_steps_per_second": 3.527,
+      "step": 11372
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4288,
+      "step": 14215
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7751916893395993,
+      "eval_combined_score": 0.7268025813927237,
+      "eval_f1": 0.678413473445848,
+      "eval_loss": 0.4724300801753998,
+      "eval_runtime": 89.3424,
+      "eval_samples_per_second": 452.529,
+      "eval_steps_per_second": 3.537,
+      "step": 14215
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4129,
+      "step": 17058
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.77491961414791,
+      "eval_combined_score": 0.7321060785844712,
+      "eval_f1": 0.6892925430210324,
+      "eval_loss": 0.4805930554866791,
+      "eval_runtime": 89.5371,
+      "eval_samples_per_second": 451.545,
+      "eval_steps_per_second": 3.529,
+      "step": 17058
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3969,
+      "step": 19901
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7784071234232006,
+      "eval_combined_score": 0.7334977672800942,
+      "eval_f1": 0.6885884111369878,
+      "eval_loss": 0.4699857532978058,
+      "eval_runtime": 89.4631,
+      "eval_samples_per_second": 451.918,
+      "eval_steps_per_second": 3.532,
+      "step": 19901
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.3813,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7790007420232501,
+      "eval_combined_score": 0.7376014216389413,
+      "eval_f1": 0.6962021012546326,
+      "eval_loss": 0.48019975423812866,
+      "eval_runtime": 89.437,
+      "eval_samples_per_second": 452.05,
+      "eval_steps_per_second": 3.533,
+      "step": 22744
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3664,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7804847885233738,
+      "eval_combined_score": 0.7378400176432165,
+      "eval_f1": 0.6951952467630593,
+      "eval_loss": 0.4764688313007355,
+      "eval_runtime": 89.5287,
+      "eval_samples_per_second": 451.587,
+      "eval_steps_per_second": 3.53,
+      "step": 25587
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.352,
+      "step": 28430
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7767994063814,
+      "eval_combined_score": 0.7427103263744261,
+      "eval_f1": 0.7086212463674524,
+      "eval_loss": 0.49650800228118896,
+      "eval_runtime": 89.4364,
+      "eval_samples_per_second": 452.053,
+      "eval_steps_per_second": 3.533,
+      "step": 28430
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.3381,
+      "step": 31273
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.7844917140737077,
+      "eval_combined_score": 0.7402554861754647,
+      "eval_f1": 0.6960192582772216,
+      "eval_loss": 0.48946356773376465,
+      "eval_runtime": 89.3745,
+      "eval_samples_per_second": 452.366,
+      "eval_steps_per_second": 3.536,
+      "step": 31273
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.3258,
+      "step": 34116
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7844422458570368,
+      "eval_combined_score": 0.744383972801292,
+      "eval_f1": 0.7043256997455472,
+      "eval_loss": 0.5092117190361023,
+      "eval_runtime": 89.3611,
+      "eval_samples_per_second": 452.434,
+      "eval_steps_per_second": 3.536,
+      "step": 34116
+    },
+    {
+      "epoch": 12.0,
+      "step": 34116,
+      "total_flos": 1.0406546414921318e+17,
+      "train_loss": 0.4099273109145315,
+      "train_runtime": 29349.5289,
+      "train_samples_per_second": 619.85,
+      "train_steps_per_second": 4.843
+    }
+  ],
+  "max_steps": 142150,
+  "num_train_epochs": 50,
+  "total_flos": 1.0406546414921318e+17,
+  "trial_name": null,
+  "trial_params": null
+}