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: 0.8027522935779816
+      value: 0.7970183486238532
 ---
 
 <!-- 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_sst2
 
-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 SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5706
-- Accuracy: 0.8028
+- Loss: 0.4671
+- Accuracy: 0.7970
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.7970183486238532,
+    "eval_loss": 0.4670983850955963,
+    "eval_runtime": 1.9077,
+    "eval_samples": 872,
+    "eval_samples_per_second": 457.101,
+    "eval_steps_per_second": 3.669,
+    "train_loss": 0.4192447450368441,
+    "train_runtime": 6419.7515,
+    "train_samples": 67349,
+    "train_samples_per_second": 524.545,
+    "train_steps_per_second": 4.105
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.7970183486238532,
+    "eval_loss": 0.4670983850955963,
+    "eval_runtime": 1.9077,
+    "eval_samples": 872,
+    "eval_samples_per_second": 457.101,
+    "eval_steps_per_second": 3.669
+}

logs/events.out.tfevents.1674779510.garda.473409.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35010376fd327a02af7514c9daec952fb451a85888eb1632df236bfe4c39c237
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.4192447450368441,
+    "train_runtime": 6419.7515,
+    "train_samples": 67349,
+    "train_samples_per_second": 524.545,
+    "train_steps_per_second": 4.105
+}

trainer_state.json

@@ -0,0 +1,235 @@
+{
+  "best_metric": 0.4670983850955963,
+  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_sst2/checkpoint-4743",
+  "epoch": 14.0,
+  "global_step": 7378,
+  "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": 0.6635,
+      "step": 527
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5389908256880734,
+      "eval_loss": 0.6994264721870422,
+      "eval_runtime": 1.8725,
+      "eval_samples_per_second": 465.675,
+      "eval_steps_per_second": 3.738,
+      "step": 527
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.5959,
+      "step": 1054
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5665137614678899,
+      "eval_loss": 0.6921247839927673,
+      "eval_runtime": 1.8745,
+      "eval_samples_per_second": 465.202,
+      "eval_steps_per_second": 3.734,
+      "step": 1054
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5684,
+      "step": 1581
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.551605504587156,
+      "eval_loss": 0.7081714868545532,
+      "eval_runtime": 1.8708,
+      "eval_samples_per_second": 466.112,
+      "eval_steps_per_second": 3.742,
+      "step": 1581
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5544,
+      "step": 2108
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5619266055045872,
+      "eval_loss": 0.688310980796814,
+      "eval_runtime": 1.8754,
+      "eval_samples_per_second": 464.97,
+      "eval_steps_per_second": 3.733,
+      "step": 2108
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5471,
+      "step": 2635
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5940366972477065,
+      "eval_loss": 0.6938039064407349,
+      "eval_runtime": 1.8755,
+      "eval_samples_per_second": 464.943,
+      "eval_steps_per_second": 3.732,
+      "step": 2635
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5414,
+      "step": 3162
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5802752293577982,
+      "eval_loss": 0.7045485973358154,
+      "eval_runtime": 1.8702,
+      "eval_samples_per_second": 466.254,
+      "eval_steps_per_second": 3.743,
+      "step": 3162
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5381,
+      "step": 3689
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5653669724770642,
+      "eval_loss": 0.7353540658950806,
+      "eval_runtime": 1.8751,
+      "eval_samples_per_second": 465.044,
+      "eval_steps_per_second": 3.733,
+      "step": 3689
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.5338,
+      "step": 4216
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5825688073394495,
+      "eval_loss": 0.7316306829452515,
+      "eval_runtime": 1.8737,
+      "eval_samples_per_second": 465.401,
+      "eval_steps_per_second": 3.736,
+      "step": 4216
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3529,
+      "step": 4743
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7970183486238532,
+      "eval_loss": 0.4670983850955963,
+      "eval_runtime": 1.8694,
+      "eval_samples_per_second": 466.456,
+      "eval_steps_per_second": 3.744,
+      "step": 4743
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.2415,
+      "step": 5270
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7981651376146789,
+      "eval_loss": 0.4721962809562683,
+      "eval_runtime": 1.8749,
+      "eval_samples_per_second": 465.103,
+      "eval_steps_per_second": 3.734,
+      "step": 5270
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.2075,
+      "step": 5797
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.8061926605504587,
+      "eval_loss": 0.4797086715698242,
+      "eval_runtime": 1.8721,
+      "eval_samples_per_second": 465.778,
+      "eval_steps_per_second": 3.739,
+      "step": 5797
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.1862,
+      "step": 6324
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7993119266055045,
+      "eval_loss": 0.5133755803108215,
+      "eval_runtime": 1.8749,
+      "eval_samples_per_second": 465.095,
+      "eval_steps_per_second": 3.734,
+      "step": 6324
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.1724,
+      "step": 6851
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.7993119266055045,
+      "eval_loss": 0.5256267786026001,
+      "eval_runtime": 1.8734,
+      "eval_samples_per_second": 465.457,
+      "eval_steps_per_second": 3.736,
+      "step": 6851
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.1662,
+      "step": 7378
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.8027522935779816,
+      "eval_loss": 0.5705782771110535,
+      "eval_runtime": 1.8724,
+      "eval_samples_per_second": 465.701,
+      "eval_steps_per_second": 3.738,
+      "step": 7378
+    },
+    {
+      "epoch": 14.0,
+      "step": 7378,
+      "total_flos": 2.789091149296435e+16,
+      "train_loss": 0.4192447450368441,
+      "train_runtime": 6419.7515,
+      "train_samples_per_second": 524.545,
+      "train_steps_per_second": 4.105
+    }
+  ],
+  "max_steps": 26350,
+  "num_train_epochs": 50,
+  "total_flos": 2.789091149296435e+16,
+  "trial_name": null,
+  "trial_params": null
+}