Training in progress, step 200

all_results.json

@@ -1,14 +1,20 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.5270758122743683,
-    "eval_loss": 0.7416719794273376,
-    "eval_runtime": 0.4617,
+    "epoch": 20.0,
+    "eval_accuracy": 75.36231884057972,
+    "eval_average_metrics": 75.36231884057972,
+    "eval_loss": 0.2184353768825531,
+    "eval_runtime": 0.505,
     "eval_samples": 277,
-    "eval_samples_per_second": 599.982,
+    "eval_samples_per_second": 273.265,
     "eval_steps_per_second": 6.498,
-    "train_loss": 0.7311154096554487,
-    "train_runtime": 165.3668,
+    "test_accuracy": 69.06474820143885,
+    "test_average_metrics": 69.06474820143885,
+    "test_loss": 0.26203805208206177,
+    "test_runtime": 0.5052,
+    "test_samples_per_second": 275.13,
+    "train_loss": 1.2798988037109376,
+    "train_runtime": 273.704,
     "train_samples": 2490,
-    "train_samples_per_second": 150.574,
-    "train_steps_per_second": 4.717
+    "train_samples_per_second": 181.948,
+    "train_steps_per_second": 1.827
 }

eval_results.json

@@ -1,9 +1,8 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.5270758122743683,
-    "eval_loss": 0.7416719794273376,
-    "eval_runtime": 0.4617,
-    "eval_samples": 277,
-    "eval_samples_per_second": 599.982,
-    "eval_steps_per_second": 6.498
+    "epoch": 20.0,
+    "eval_accuracy": 75.36231884057972,
+    "eval_average_metrics": 75.36231884057972,
+    "eval_loss": 0.2184353768825531,
+    "eval_runtime": 0.505,
+    "eval_samples_per_second": 273.265
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a181fa83a25220bf33cf44d9adf3957ff9e907d54405e714d32d8f0226f5855
+oid sha256:4779d2e8713715239b41a27c9b18d5cdab3da2fa582a864920cbc99d53191448
 size 1084131

runs/Feb01_00-07-44_node1/1643645355.1959577/events.out.tfevents.1643645355.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e515efaad4d9cdeae1ee0df025c81ac44432fd6da84d7a172a360d29c5ce83e4
+size 5008

runs/Feb01_00-07-44_node1/events.out.tfevents.1643645355.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14ccde1a1e9c47fc8a5c725b58f8cabb37d9015922ee898fffbb398f279eef0a
+size 4301

runs/Jan31_20-23-25_node1/events.out.tfevents.1643632055.node1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5605353a378469ef483ecffa7b4bbb04776c2cf3273d8ad500d9e1bc1056247b
-size 4623
+oid sha256:21b275ce845e2942d13e0781632c37b6a26329d9cdaf390b4397fcb6fd5cafae
+size 5456

runs/Jan31_20-23-25_node1/events.out.tfevents.1643632329.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c45e471d452b41652a038c57d839a3eeed2c7888282443217f676f871b148858
+size 684

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "test_accuracy": 69.06474820143885,
+    "test_average_metrics": 69.06474820143885,
+    "test_loss": 0.26203805208206177,
+    "test_runtime": 0.5052,
+    "test_samples_per_second": 275.13
+}

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.7311154096554487,
-    "train_runtime": 165.3668,
+    "epoch": 20.0,
+    "train_loss": 1.2798988037109376,
+    "train_runtime": 273.704,
     "train_samples": 2490,
-    "train_samples_per_second": 150.574,
-    "train_steps_per_second": 4.717
+    "train_samples_per_second": 181.948,
+    "train_steps_per_second": 1.827
 }

trainer_state.json

@@ -1,121 +1,49 @@
 {
-  "best_metric": 0.5270758122743683,
-  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-312",
-  "epoch": 10.0,
-  "global_step": 780,
+  "best_metric": 75.36231884057972,
+  "best_model_checkpoint": "outputs/bitfit/t5-base/rte/checkpoint-400",
+  "epoch": 20.0,
+  "global_step": 500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.755623996257782,
-      "eval_runtime": 0.4771,
-      "eval_samples_per_second": 580.569,
-      "eval_steps_per_second": 6.288,
-      "step": 78
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.822638988494873,
-      "eval_runtime": 0.4598,
-      "eval_samples_per_second": 602.409,
-      "eval_steps_per_second": 6.524,
-      "step": 156
-    },
-    {
-      "epoch": 3.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7435575127601624,
-      "eval_runtime": 0.4678,
-      "eval_samples_per_second": 592.135,
-      "eval_steps_per_second": 6.413,
-      "step": 234
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.7416719794273376,
-      "eval_runtime": 0.4671,
-      "eval_samples_per_second": 592.998,
-      "eval_steps_per_second": 6.422,
-      "step": 312
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.694923996925354,
-      "eval_runtime": 0.4636,
-      "eval_samples_per_second": 597.536,
-      "eval_steps_per_second": 6.472,
-      "step": 390
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7276150584220886,
-      "eval_runtime": 0.469,
-      "eval_samples_per_second": 590.599,
-      "eval_steps_per_second": 6.396,
-      "step": 468
-    },
-    {
-      "epoch": 6.41,
-      "learning_rate": 0.001145975443383356,
-      "loss": 0.7468,
-      "step": 500
-    },
-    {
-      "epoch": 7.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7213642597198486,
-      "eval_runtime": 0.4575,
-      "eval_samples_per_second": 605.474,
-      "eval_steps_per_second": 6.557,
-      "step": 546
-    },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.6917794942855835,
-      "eval_runtime": 0.4708,
-      "eval_samples_per_second": 588.418,
-      "eval_steps_per_second": 6.373,
-      "step": 624
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.693771243095398,
-      "eval_runtime": 0.4588,
-      "eval_samples_per_second": 603.785,
-      "eval_steps_per_second": 6.539,
-      "step": 702
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.6928249597549438,
-      "eval_runtime": 0.4734,
-      "eval_samples_per_second": 585.163,
-      "eval_steps_per_second": 6.338,
-      "step": 780
+      "eval_accuracy": 54.347826086956516,
+      "eval_average_metrics": 54.347826086956516,
+      "eval_loss": 0.3104463517665863,
+      "eval_runtime": 0.6899,
+      "eval_samples_per_second": 200.036,
+      "step": 200
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 75.36231884057972,
+      "eval_average_metrics": 75.36231884057972,
+      "eval_loss": 0.2184353768825531,
+      "eval_runtime": 0.6861,
+      "eval_samples_per_second": 201.146,
+      "step": 400
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0,
+      "loss": 1.2799,
+      "step": 500
     },
     {
-      "epoch": 10.0,
-      "step": 780,
-      "total_flos": 5862398439168000.0,
-      "train_loss": 0.7311154096554487,
-      "train_runtime": 165.3668,
-      "train_samples_per_second": 150.574,
-      "train_steps_per_second": 4.717
+      "epoch": 20.0,
+      "step": 500,
+      "total_flos": 7590919206297600.0,
+      "train_loss": 1.2798988037109376,
+      "train_runtime": 273.704,
+      "train_samples_per_second": 181.948,
+      "train_steps_per_second": 1.827
     }
   ],
-  "max_steps": 780,
-  "num_train_epochs": 10,
-  "total_flos": 5862398439168000.0,
+  "max_steps": 500,
+  "num_train_epochs": 20,
+  "total_flos": 7590919206297600.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7379d0fb530a33275f9a29436cd89bd9b522f8fde4c16696f3d8aa86486a818
+oid sha256:cb2732fc0f205ecb2ff6a4dc01793c5bad05e44f957c3092d344cc75969c345d
 size 3183