Training in progress, step 200

all_results.json

@@ -1,20 +1,23 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 88.72549019607843,
+    "eval_accuracy": 0.7083333333333334,
     "eval_average_metrics": 90.18092691622104,
-    "eval_f1": 91.63636363636364,
-    "eval_loss": 0.3694871664047241,
-    "eval_runtime": 1.0502,
-    "eval_samples_per_second": 194.246,
+    "eval_combined_score": 0.7652279521674141,
+    "eval_f1": 0.8221225710014948,
+    "eval_loss": 0.5735756754875183,
+    "eval_runtime": 0.4847,
+    "eval_samples": 408,
+    "eval_samples_per_second": 841.743,
+    "eval_steps_per_second": 14.442,
     "test_accuracy": 89.70588235294117,
     "test_average_metrics": 91.19440459110473,
     "test_f1": 92.6829268292683,
     "test_loss": 0.2836102843284607,
     "test_runtime": 1.0233,
     "test_samples_per_second": 199.347,
-    "train_loss": 0.04842971423397893,
-    "train_runtime": 850.7003,
+    "train_loss": 0.5910721482901737,
+    "train_runtime": 193.8371,
     "train_samples": 3668,
-    "train_samples_per_second": 86.235,
-    "train_steps_per_second": 2.704
+    "train_samples_per_second": 378.462,
+    "train_steps_per_second": 5.984
 }

eval_results.json

@@ -1,9 +1,11 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 88.72549019607843,
-    "eval_average_metrics": 90.18092691622104,
-    "eval_f1": 91.63636363636364,
-    "eval_loss": 0.3694871664047241,
-    "eval_runtime": 1.0502,
-    "eval_samples_per_second": 194.246
+    "eval_accuracy": 0.7083333333333334,
+    "eval_combined_score": 0.7652279521674141,
+    "eval_f1": 0.8221225710014948,
+    "eval_loss": 0.5735756754875183,
+    "eval_runtime": 0.4847,
+    "eval_samples": 408,
+    "eval_samples_per_second": 841.743,
+    "eval_steps_per_second": 14.442
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:636e38716a1496de857a883086df4a681c1d638727ee2bcc063dbc37e14223c9
-size 2834001
+oid sha256:2ea20482f4ef6da5b6daf69ce891f69b21c5ef0fc768a2ea5b9f90c3a92a3d81
+size 41271819

runs/Feb12_11-11-44_node1/events.out.tfevents.1644635578.node1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:427123085ce51a80e53412b2a539079c62df93b916d4fcf8556e90505469741f
-size 5536
+oid sha256:11cc29a257317a9fb9172364c676a9aa44f65c9625e6da973d1a9ed814af1415
+size 6317

runs/Feb12_11-11-44_node1/events.out.tfevents.1644635773.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3136b9d100428a4a7c558b04fb22957cd9b2568b26a29f316c4313decb5df276
+size 467

runs/Feb12_11-26-13_node1/1644636416.8581543/events.out.tfevents.1644636416.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:413116e9ec2f144e058c72fd4402295ffd3f6cb63a706dbc7e2ef54668821cfb
+size 4588

runs/Feb12_11-26-13_node1/events.out.tfevents.1644636416.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47cd20d02aa05775da961c47c7e7c1eae9e1a6f5f482031a893d11b5c7585429
+size 3514

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
-    "train_loss": 0.04842971423397893,
-    "train_runtime": 850.7003,
+    "train_loss": 0.5910721482901737,
+    "train_runtime": 193.8371,
     "train_samples": 3668,
-    "train_samples_per_second": 86.235,
-    "train_steps_per_second": 2.704
+    "train_samples_per_second": 378.462,
+    "train_steps_per_second": 5.984
 }

trainer_state.json

@@ -1,159 +1,92 @@
 {
-  "best_metric": 90.18092691622104,
-  "best_model_checkpoint": "outputs/bitfit/t5-base/mrpc/checkpoint-2200",
+  "best_metric": 0.7083333333333334,
+  "best_model_checkpoint": "outputs/prefix/roberta-base/mrpc/checkpoint-600",
   "epoch": 20.0,
-  "global_step": 2300,
+  "global_step": 1160,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.74,
-      "eval_accuracy": 86.76470588235294,
-      "eval_average_metrics": 88.54364326375712,
-      "eval_f1": 90.3225806451613,
-      "eval_loss": 0.1595430225133896,
-      "eval_runtime": 1.0042,
-      "eval_samples_per_second": 203.156,
+      "epoch": 3.45,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7631208171891511,
+      "eval_f1": 0.8203592814371258,
+      "eval_loss": 0.602976381778717,
+      "eval_runtime": 1.1443,
+      "eval_samples_per_second": 356.557,
+      "eval_steps_per_second": 6.117,
       "step": 200
     },
     {
-      "epoch": 3.48,
-      "eval_accuracy": 87.25490196078431,
-      "eval_average_metrics": 89.1753961858716,
-      "eval_f1": 91.0958904109589,
-      "eval_loss": 0.19655147194862366,
-      "eval_runtime": 0.9978,
-      "eval_samples_per_second": 204.44,
+      "epoch": 6.9,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7620860566448802,
+      "eval_f1": 0.8207407407407408,
+      "eval_loss": 0.58584064245224,
+      "eval_runtime": 0.5538,
+      "eval_samples_per_second": 736.744,
+      "eval_steps_per_second": 12.64,
       "step": 400
     },
     {
-      "epoch": 4.35,
-      "learning_rate": 0.00023478260869565215,
-      "loss": 0.1466,
+      "epoch": 8.62,
+      "learning_rate": 0.0001706896551724138,
+      "loss": 0.6074,
       "step": 500
     },
     {
-      "epoch": 5.22,
-      "eval_accuracy": 88.23529411764706,
-      "eval_average_metrics": 89.83193277310924,
-      "eval_f1": 91.42857142857143,
-      "eval_loss": 0.20800147950649261,
-      "eval_runtime": 1.03,
-      "eval_samples_per_second": 198.057,
+      "epoch": 10.34,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7652279521674141,
+      "eval_f1": 0.8221225710014948,
+      "eval_loss": 0.5735756754875183,
+      "eval_runtime": 0.6085,
+      "eval_samples_per_second": 670.475,
+      "eval_steps_per_second": 11.503,
       "step": 600
     },
     {
-      "epoch": 6.96,
-      "eval_accuracy": 87.25490196078431,
-      "eval_average_metrics": 88.91730605285592,
-      "eval_f1": 90.57971014492753,
-      "eval_loss": 0.25488224625587463,
-      "eval_runtime": 0.9975,
-      "eval_samples_per_second": 204.517,
+      "epoch": 13.79,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7612822034643454,
+      "eval_f1": 0.8191330343796712,
+      "eval_loss": 0.5688421726226807,
+      "eval_runtime": 0.6104,
+      "eval_samples_per_second": 668.411,
+      "eval_steps_per_second": 11.468,
       "step": 800
     },
     {
-      "epoch": 8.7,
-      "learning_rate": 0.00016956521739130433,
-      "loss": 0.0433,
+      "epoch": 17.24,
+      "learning_rate": 4.137931034482758e-05,
+      "loss": 0.5801,
       "step": 1000
     },
     {
-      "epoch": 8.7,
-      "eval_accuracy": 87.74509803921569,
-      "eval_average_metrics": 89.48658410732715,
-      "eval_f1": 91.2280701754386,
-      "eval_loss": 0.27675318717956543,
-      "eval_runtime": 1.004,
-      "eval_samples_per_second": 203.193,
+      "epoch": 17.24,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7649612693653174,
+      "eval_f1": 0.8215892053973014,
+      "eval_loss": 0.5650697350502014,
+      "eval_runtime": 0.7692,
+      "eval_samples_per_second": 530.398,
+      "eval_steps_per_second": 9.1,
       "step": 1000
     },
-    {
-      "epoch": 10.43,
-      "eval_accuracy": 88.23529411764706,
-      "eval_average_metrics": 89.83193277310924,
-      "eval_f1": 91.42857142857143,
-      "eval_loss": 0.3537355363368988,
-      "eval_runtime": 1.0419,
-      "eval_samples_per_second": 195.799,
-      "step": 1200
-    },
-    {
-      "epoch": 12.17,
-      "eval_accuracy": 86.76470588235294,
-      "eval_average_metrics": 88.43729799612153,
-      "eval_f1": 90.10989010989012,
-      "eval_loss": 0.4248361885547638,
-      "eval_runtime": 0.9784,
-      "eval_samples_per_second": 208.512,
-      "step": 1400
-    },
-    {
-      "epoch": 13.04,
-      "learning_rate": 0.00010434782608695651,
-      "loss": 0.0174,
-      "step": 1500
-    },
-    {
-      "epoch": 13.91,
-      "eval_accuracy": 87.25490196078431,
-      "eval_average_metrics": 88.9511919875864,
-      "eval_f1": 90.64748201438849,
-      "eval_loss": 0.35347217321395874,
-      "eval_runtime": 1.0071,
-      "eval_samples_per_second": 202.569,
-      "step": 1600
-    },
-    {
-      "epoch": 15.65,
-      "eval_accuracy": 88.23529411764706,
-      "eval_average_metrics": 89.7380850150279,
-      "eval_f1": 91.24087591240875,
-      "eval_loss": 0.37131425738334656,
-      "eval_runtime": 0.9929,
-      "eval_samples_per_second": 205.451,
-      "step": 1800
-    },
-    {
-      "epoch": 17.39,
-      "learning_rate": 3.913043478260869e-05,
-      "loss": 0.0119,
-      "step": 2000
-    },
-    {
-      "epoch": 17.39,
-      "eval_accuracy": 88.23529411764706,
-      "eval_average_metrics": 89.769820971867,
-      "eval_f1": 91.30434782608695,
-      "eval_loss": 0.3558318316936493,
-      "eval_runtime": 1.0675,
-      "eval_samples_per_second": 191.099,
-      "step": 2000
-    },
-    {
-      "epoch": 19.13,
-      "eval_accuracy": 88.72549019607843,
-      "eval_average_metrics": 90.18092691622104,
-      "eval_f1": 91.63636363636364,
-      "eval_loss": 0.3694871664047241,
-      "eval_runtime": 1.0401,
-      "eval_samples_per_second": 196.137,
-      "step": 2200
-    },
     {
       "epoch": 20.0,
-      "step": 2300,
-      "total_flos": 9229438723811328.0,
-      "train_loss": 0.04842971423397893,
-      "train_runtime": 850.7003,
-      "train_samples_per_second": 86.235,
-      "train_steps_per_second": 2.704
+      "step": 1160,
+      "total_flos": 4831687561666560.0,
+      "train_loss": 0.5910721482901737,
+      "train_runtime": 193.8371,
+      "train_samples_per_second": 378.462,
+      "train_steps_per_second": 5.984
     }
   ],
-  "max_steps": 2300,
+  "max_steps": 1160,
   "num_train_epochs": 20,
-  "total_flos": 9229438723811328.0,
+  "total_flos": 4831687561666560.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6dd2861778bc36ab1ce0455b0ed185f4a6bd9fdabe4ba7aafabdd1fe22a2899c
+oid sha256:b636208b801856553c9b1677f10515a2d30edac604dfc118b64fa760241111b4
 size 2991