End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 base_model: gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48
 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
@@ -29,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_w_init_48_ver2_sst2
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the GLUE SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7068
+- Loss: 0.6941
 - Accuracy: 0.5092
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5091743119266054,
+    "eval_loss": 0.6941368579864502,
+    "eval_runtime": 4.4644,
+    "eval_samples": 872,
+    "eval_samples_per_second": 195.323,
+    "eval_steps_per_second": 3.136,
+    "train_loss": 0.6891914381131503,
+    "train_runtime": 7431.7212,
+    "train_samples": 67349,
+    "train_samples_per_second": 135.936,
+    "train_steps_per_second": 2.125
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5091743119266054,
+    "eval_loss": 0.6941368579864502,
+    "eval_runtime": 4.4644,
+    "eval_samples": 872,
+    "eval_samples_per_second": 195.323,
+    "eval_steps_per_second": 3.136
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.6891914381131503,
+    "train_runtime": 7431.7212,
+    "train_samples": 67349,
+    "train_samples_per_second": 135.936,
+    "train_steps_per_second": 2.125
+}

trainer_state.json

@@ -0,0 +1,133 @@
+{
+  "best_metric": 0.6941368579864502,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init_48_ver2_sst2/checkpoint-2106",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 7371,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.6948,
+      "step": 1053
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.694778561592102,
+      "eval_runtime": 4.5684,
+      "eval_samples_per_second": 190.875,
+      "eval_steps_per_second": 3.065,
+      "step": 1053
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 0.6927,
+      "step": 2106
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6941368579864502,
+      "eval_runtime": 4.5828,
+      "eval_samples_per_second": 190.278,
+      "eval_steps_per_second": 3.055,
+      "step": 2106
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6879,
+      "step": 3159
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.7004609704017639,
+      "eval_runtime": 4.5708,
+      "eval_samples_per_second": 190.777,
+      "eval_steps_per_second": 3.063,
+      "step": 3159
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 0.6873,
+      "step": 4212
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.7004290819168091,
+      "eval_runtime": 4.5617,
+      "eval_samples_per_second": 191.157,
+      "eval_steps_per_second": 3.069,
+      "step": 4212
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.6887,
+      "step": 5265
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.7151248455047607,
+      "eval_runtime": 4.571,
+      "eval_samples_per_second": 190.768,
+      "eval_steps_per_second": 3.063,
+      "step": 5265
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 0.6871,
+      "step": 6318
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6974788308143616,
+      "eval_runtime": 4.576,
+      "eval_samples_per_second": 190.561,
+      "eval_steps_per_second": 3.059,
+      "step": 6318
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.6859,
+      "step": 7371
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.7067839503288269,
+      "eval_runtime": 4.581,
+      "eval_samples_per_second": 190.351,
+      "eval_steps_per_second": 3.056,
+      "step": 7371
+    },
+    {
+      "epoch": 7.0,
+      "step": 7371,
+      "total_flos": 6.981918485852979e+16,
+      "train_loss": 0.6891914381131503,
+      "train_runtime": 7431.7212,
+      "train_samples_per_second": 135.936,
+      "train_steps_per_second": 2.125
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 15795,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 6.981918485852979e+16,
+  "trial_name": null,
+  "trial_params": null
+}