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 SST2
       type: glue
       config: sst2
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8061926605504587
+      value: 0.7901376146788991
 ---
 
 <!-- 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_sst2
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the GLUE SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6409
-- Accuracy: 0.8062
+- Loss: 0.4525
+- Accuracy: 0.7901
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.7901376146788991,
+    "eval_loss": 0.4525200426578522,
+    "eval_runtime": 1.0996,
+    "eval_samples": 872,
+    "eval_samples_per_second": 793.011,
+    "eval_steps_per_second": 3.638,
+    "train_loss": 0.32838042577107746,
+    "train_runtime": 2121.3212,
+    "train_samples": 67349,
+    "train_samples_per_second": 1587.431,
+    "train_steps_per_second": 6.223
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.7901376146788991,
+    "eval_loss": 0.4525200426578522,
+    "eval_runtime": 1.0996,
+    "eval_samples": 872,
+    "eval_samples_per_second": 793.011,
+    "eval_steps_per_second": 3.638
+}

logs/events.out.tfevents.1677698969.serv-9220.1178704.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e32ec27bc18f9fdcdb5fc02547c33263f3b21913dfb72ef2c0bfefb2dc13e9c
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.32838042577107746,
+    "train_runtime": 2121.3212,
+    "train_samples": 67349,
+    "train_samples_per_second": 1587.431,
+    "train_steps_per_second": 6.223
+}

trainer_state.json

@@ -0,0 +1,145 @@
+{
+  "best_metric": 0.4525200426578522,
+  "best_model_checkpoint": "hBERTv1_sst2/checkpoint-792",
+  "epoch": 8.0,
+  "global_step": 2112,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.900757575757576e-05,
+      "loss": 0.6905,
+      "step": 264
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5252293577981652,
+      "eval_loss": 0.6919494271278381,
+      "eval_runtime": 1.1023,
+      "eval_samples_per_second": 791.06,
+      "eval_steps_per_second": 3.629,
+      "step": 264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.801136363636364e-05,
+      "loss": 0.6609,
+      "step": 528
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6915137614678899,
+      "eval_loss": 0.6088164448738098,
+      "eval_runtime": 1.0979,
+      "eval_samples_per_second": 794.271,
+      "eval_steps_per_second": 3.643,
+      "step": 528
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7011363636363636e-05,
+      "loss": 0.4152,
+      "step": 792
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7901376146788991,
+      "eval_loss": 0.4525200426578522,
+      "eval_runtime": 1.0982,
+      "eval_samples_per_second": 794.011,
+      "eval_steps_per_second": 3.642,
+      "step": 792
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.601136363636364e-05,
+      "loss": 0.2611,
+      "step": 1056
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8096330275229358,
+      "eval_loss": 0.46269044280052185,
+      "eval_runtime": 1.0972,
+      "eval_samples_per_second": 794.783,
+      "eval_steps_per_second": 3.646,
+      "step": 1056
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.501136363636364e-05,
+      "loss": 0.1953,
+      "step": 1320
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8073394495412844,
+      "eval_loss": 0.48936158418655396,
+      "eval_runtime": 1.0982,
+      "eval_samples_per_second": 794.035,
+      "eval_steps_per_second": 3.642,
+      "step": 1320
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.401136363636364e-05,
+      "loss": 0.1588,
+      "step": 1584
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.801605504587156,
+      "eval_loss": 0.6001835465431213,
+      "eval_runtime": 1.0997,
+      "eval_samples_per_second": 792.972,
+      "eval_steps_per_second": 3.637,
+      "step": 1584
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.301136363636364e-05,
+      "loss": 0.1336,
+      "step": 1848
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8061926605504587,
+      "eval_loss": 0.646708607673645,
+      "eval_runtime": 1.0961,
+      "eval_samples_per_second": 795.513,
+      "eval_steps_per_second": 3.649,
+      "step": 1848
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.201136363636364e-05,
+      "loss": 0.1117,
+      "step": 2112
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8061926605504587,
+      "eval_loss": 0.6409061551094055,
+      "eval_runtime": 1.0981,
+      "eval_samples_per_second": 794.077,
+      "eval_steps_per_second": 3.643,
+      "step": 2112
+    },
+    {
+      "epoch": 8.0,
+      "step": 2112,
+      "total_flos": 6.805537600779059e+16,
+      "train_loss": 0.32838042577107746,
+      "train_runtime": 2121.3212,
+      "train_samples_per_second": 1587.431,
+      "train_steps_per_second": 6.223
+    }
+  ],
+  "max_steps": 13200,
+  "num_train_epochs": 50,
+  "total_flos": 6.805537600779059e+16,
+  "trial_name": null,
+  "trial_params": null
+}