End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.3544574630667346
+      value: 0.3522172497965826
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_no_pretrain_mnli
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: nan
-- Accuracy: 0.3545
+- Loss: 1.0982
+- Accuracy: 0.3522
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0981897115707397,
+    "eval_loss_mm": 1.0982112884521484,
+    "eval_runtime": 17.0982,
+    "eval_runtime_mm": 17.1583,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 574.038,
+    "eval_samples_per_second_mm": 573.017,
+    "eval_steps_per_second": 4.503,
+    "eval_steps_per_second_mm": 4.488,
+    "train_loss": 0.7333283836765944,
+    "train_runtime": 17892.0454,
+    "train_samples": 392702,
+    "train_samples_per_second": 1097.421,
+    "train_steps_per_second": 8.574
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0981897115707397,
+    "eval_loss_mm": 1.0982112884521484,
+    "eval_runtime": 17.0982,
+    "eval_runtime_mm": 17.1583,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 574.038,
+    "eval_samples_per_second_mm": 573.017,
+    "eval_steps_per_second": 4.503,
+    "eval_steps_per_second_mm": 4.488
+}

logs/events.out.tfevents.1685377873.serv-9225.2840720.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd80d2f2c4bbac87150d6dbe8c740380698a2453bf5b1efea671d5df274247b2
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.7333283836765944,
+    "train_runtime": 17892.0454,
+    "train_samples": 392702,
+    "train_samples_per_second": 1097.421,
+    "train_steps_per_second": 8.574
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 1.0981897115707397,
+  "best_model_checkpoint": "hBERTv1_no_pretrain_mnli/checkpoint-12272",
+  "epoch": 9.0,
+  "global_step": 27612,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.000490006518904824,
+      "loss": 1.1069,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0985444784164429,
+      "eval_runtime": 17.187,
+      "eval_samples_per_second": 571.071,
+      "eval_steps_per_second": 4.48,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.000480006518904824,
+      "loss": 1.0987,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0985891819000244,
+      "eval_runtime": 17.1415,
+      "eval_samples_per_second": 572.587,
+      "eval_steps_per_second": 4.492,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.000470006518904824,
+      "loss": 1.0986,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0984275341033936,
+      "eval_runtime": 17.1414,
+      "eval_samples_per_second": 572.589,
+      "eval_steps_per_second": 4.492,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.000460009778357236,
+      "loss": 1.0986,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0981897115707397,
+      "eval_runtime": 17.1872,
+      "eval_samples_per_second": 571.065,
+      "eval_steps_per_second": 4.48,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045001629726206,
+      "loss": 1.0986,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0984472036361694,
+      "eval_runtime": 17.1683,
+      "eval_samples_per_second": 571.695,
+      "eval_steps_per_second": 4.485,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044056714471968713,
+      "loss": 1.0985,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": NaN,
+      "eval_runtime": 17.1388,
+      "eval_samples_per_second": 572.677,
+      "eval_steps_per_second": 4.493,
+      "step": 18408
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0004305671447196871,
+      "loss": 0.0,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": NaN,
+      "eval_runtime": 17.1475,
+      "eval_samples_per_second": 572.388,
+      "eval_steps_per_second": 4.49,
+      "step": 21476
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.0004205671447196871,
+      "loss": 0.0,
+      "step": 24544
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": NaN,
+      "eval_runtime": 17.2161,
+      "eval_samples_per_second": 570.106,
+      "eval_steps_per_second": 4.473,
+      "step": 24544
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.0004105671447196871,
+      "loss": 0.0,
+      "step": 27612
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": NaN,
+      "eval_runtime": 17.1444,
+      "eval_samples_per_second": 572.491,
+      "eval_steps_per_second": 4.491,
+      "step": 27612
+    },
+    {
+      "epoch": 9.0,
+      "step": 27612,
+      "total_flos": 5.2342526347680154e+17,
+      "train_loss": 0.7333283836765944,
+      "train_runtime": 17892.0454,
+      "train_samples_per_second": 1097.421,
+      "train_steps_per_second": 8.574
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 5.2342526347680154e+17,
+  "trial_name": null,
+  "trial_params": null
+}