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 MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7026999490575649
+      value: 0.7034174125305126
 ---
 
 <!-- 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. -->
 
 # sa_BERT_48_mnli
 
-This model is a fine-tuned version of [gokuls/bert_base_48](https://huggingface.co/gokuls/bert_base_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_base_48](https://huggingface.co/gokuls/bert_base_48) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0196
-- Accuracy: 0.7027
+- Loss: 0.7082
+- Accuracy: 0.7034
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.7075904228222109,
+    "eval_accuracy_mm": 0.7034174125305126,
+    "eval_loss": 0.7072979807853699,
+    "eval_loss_mm": 0.7082261443138123,
+    "eval_runtime": 20.4606,
+    "eval_runtime_mm": 20.6513,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 479.702,
+    "eval_samples_per_second_mm": 476.095,
+    "eval_steps_per_second": 5.034,
+    "eval_steps_per_second_mm": 4.988,
+    "train_loss": 0.5405747011563323,
+    "train_runtime": 22664.7689,
+    "train_samples": 392702,
+    "train_samples_per_second": 866.327,
+    "train_steps_per_second": 9.025
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.7075904228222109,
+    "eval_accuracy_mm": 0.7034174125305126,
+    "eval_loss": 0.7072979807853699,
+    "eval_loss_mm": 0.7082261443138123,
+    "eval_runtime": 20.4606,
+    "eval_runtime_mm": 20.6513,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 479.702,
+    "eval_samples_per_second_mm": 476.095,
+    "eval_steps_per_second": 5.034,
+    "eval_steps_per_second_mm": 4.988
+}

logs/events.out.tfevents.1686919073.serv-3318.1430967.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77c4713958056b4cc9c4b3e19635eff30190997d7f12a032caa338efe0ffbcc7
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.5405747011563323,
+    "train_runtime": 22664.7689,
+    "train_samples": 392702,
+    "train_samples_per_second": 866.327,
+    "train_steps_per_second": 9.025
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 0.7072979807853699,
+  "best_model_checkpoint": "sa_BERT_48_mnli/checkpoint-16364",
+  "epoch": 9.0,
+  "global_step": 36819,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.9145,
+      "step": 4091
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6535914416709119,
+      "eval_loss": 0.8005583882331848,
+      "eval_runtime": 20.7416,
+      "eval_samples_per_second": 473.203,
+      "eval_steps_per_second": 4.966,
+      "step": 4091
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7442,
+      "step": 8182
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6902699949057565,
+      "eval_loss": 0.7245084047317505,
+      "eval_runtime": 20.6753,
+      "eval_samples_per_second": 474.721,
+      "eval_steps_per_second": 4.982,
+      "step": 8182
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6631,
+      "step": 12273
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6979113601630158,
+      "eval_loss": 0.7323198914527893,
+      "eval_runtime": 20.6762,
+      "eval_samples_per_second": 474.702,
+      "eval_steps_per_second": 4.982,
+      "step": 12273
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.5942,
+      "step": 16364
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7075904228222109,
+      "eval_loss": 0.7072979807853699,
+      "eval_runtime": 20.589,
+      "eval_samples_per_second": 476.712,
+      "eval_steps_per_second": 5.003,
+      "step": 16364
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.5241,
+      "step": 20455
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7015792154865003,
+      "eval_loss": 0.7474916577339172,
+      "eval_runtime": 20.6918,
+      "eval_samples_per_second": 474.342,
+      "eval_steps_per_second": 4.978,
+      "step": 20455
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.4526,
+      "step": 24546
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7088130412633724,
+      "eval_loss": 0.8377429842948914,
+      "eval_runtime": 20.7171,
+      "eval_samples_per_second": 473.764,
+      "eval_steps_per_second": 4.972,
+      "step": 24546
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.3842,
+      "step": 28637
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6955680081507896,
+      "eval_loss": 0.8736042380332947,
+      "eval_runtime": 20.7409,
+      "eval_samples_per_second": 473.219,
+      "eval_steps_per_second": 4.966,
+      "step": 28637
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.3213,
+      "step": 32728
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6945491594498217,
+      "eval_loss": 0.9334085583686829,
+      "eval_runtime": 20.6303,
+      "eval_samples_per_second": 475.756,
+      "eval_steps_per_second": 4.993,
+      "step": 32728
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.2669,
+      "step": 36819
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7026999490575649,
+      "eval_loss": 1.0196205377578735,
+      "eval_runtime": 20.722,
+      "eval_samples_per_second": 473.652,
+      "eval_steps_per_second": 4.971,
+      "step": 36819
+    },
+    {
+      "epoch": 9.0,
+      "step": 36819,
+      "total_flos": 4.6496322950057165e+17,
+      "train_loss": 0.5405747011563323,
+      "train_runtime": 22664.7689,
+      "train_samples_per_second": 866.327,
+      "train_steps_per_second": 9.025
+    }
+  ],
+  "max_steps": 204550,
+  "num_train_epochs": 50,
+  "total_flos": 4.6496322950057165e+17,
+  "trial_name": null,
+  "trial_params": null
+}