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 QNLI
       type: glue
       config: qnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5914332784184514
+      value: 0.615595826468973
 ---
 
 <!-- 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_sa_GLUE_Experiment_logit_kd_qnli
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0290
-- Accuracy: 0.5914
+- Loss: 0.9573
+- Accuracy: 0.6156
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.615595826468973,
+    "eval_loss": 0.9573363661766052,
+    "eval_runtime": 12.7235,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 429.363,
+    "eval_steps_per_second": 3.38,
+    "train_loss": 0.9566188341823108,
+    "train_runtime": 4545.7624,
+    "train_samples": 104743,
+    "train_samples_per_second": 1152.095,
+    "train_steps_per_second": 9.008
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.615595826468973,
+    "eval_loss": 0.9573363661766052,
+    "eval_runtime": 12.7235,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 429.363,
+    "eval_steps_per_second": 3.38
+}

logs/events.out.tfevents.1675035462.serv-3330.1319687.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1c7aa2ebaa713ed066f7006ca28637de72972f066713e87313fd581230b8aa2
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.9566188341823108,
+    "train_runtime": 4545.7624,
+    "train_samples": 104743,
+    "train_samples_per_second": 1152.095,
+    "train_steps_per_second": 9.008
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.9573363661766052,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_qnli/checkpoint-1638",
+  "epoch": 7.0,
+  "global_step": 5733,
+  "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": 1.0984,
+      "step": 819
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6220025626944902,
+      "eval_loss": 0.9626317620277405,
+      "eval_runtime": 12.6845,
+      "eval_samples_per_second": 430.682,
+      "eval_steps_per_second": 3.39,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 1.0171,
+      "step": 1638
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.615595826468973,
+      "eval_loss": 0.9573363661766052,
+      "eval_runtime": 12.7142,
+      "eval_samples_per_second": 429.676,
+      "eval_steps_per_second": 3.382,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.9717,
+      "step": 2457
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6104704374885594,
+      "eval_loss": 0.9651486277580261,
+      "eval_runtime": 12.7327,
+      "eval_samples_per_second": 429.053,
+      "eval_steps_per_second": 3.377,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.9377,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6024162548050521,
+      "eval_loss": 0.9713066220283508,
+      "eval_runtime": 12.6949,
+      "eval_samples_per_second": 430.33,
+      "eval_steps_per_second": 3.387,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.9132,
+      "step": 4095
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5987552626761853,
+      "eval_loss": 0.9811589121818542,
+      "eval_runtime": 12.741,
+      "eval_samples_per_second": 428.772,
+      "eval_steps_per_second": 3.375,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.89,
+      "step": 4914
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5982061138568552,
+      "eval_loss": 1.0107508897781372,
+      "eval_runtime": 12.8729,
+      "eval_samples_per_second": 424.379,
+      "eval_steps_per_second": 3.34,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.8683,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5914332784184514,
+      "eval_loss": 1.0290495157241821,
+      "eval_runtime": 12.6935,
+      "eval_samples_per_second": 430.378,
+      "eval_steps_per_second": 3.388,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "step": 5733,
+      "total_flos": 2.2988974586855424e+16,
+      "train_loss": 0.9566188341823108,
+      "train_runtime": 4545.7624,
+      "train_samples_per_second": 1152.095,
+      "train_steps_per_second": 9.008
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 2.2988974586855424e+16,
+  "trial_name": null,
+  "trial_params": null
+}