End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 9.8,
-    "eval_accuracy": 0.8715686274509804,
-    "eval_loss": 0.3589998185634613,
-    "eval_runtime": 34.1863,
-    "eval_samples_per_second": 59.673,
-    "eval_steps_per_second": 1.872,
-    "total_flos": 7.501244449451129e+18,
-    "train_loss": 0.7515069055557251,
-    "train_runtime": 2638.4717,
-    "train_samples_per_second": 30.923,
-    "train_steps_per_second": 0.095
+    "epoch": 8.11,
+    "eval_accuracy": 0.0,
+    "eval_loss": NaN,
+    "eval_runtime": 5.2931,
+    "eval_samples_per_second": 55.733,
+    "eval_steps_per_second": 1.889,
+    "total_flos": 3.675135085585367e+17,
+    "train_loss": 0.0,
+    "train_runtime": 249.2359,
+    "train_samples_per_second": 47.345,
+    "train_steps_per_second": 0.12
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.8,
-    "eval_accuracy": 0.8715686274509804,
-    "eval_loss": 0.3589998185634613,
-    "eval_runtime": 34.1863,
-    "eval_samples_per_second": 59.673,
-    "eval_steps_per_second": 1.872
+    "epoch": 8.11,
+    "eval_accuracy": 0.0,
+    "eval_loss": NaN,
+    "eval_runtime": 5.2931,
+    "eval_samples_per_second": 55.733,
+    "eval_steps_per_second": 1.889
 }

runs/Mar10_00-45-05_f2dfbab62cd5/events.out.tfevents.1710031788.f2dfbab62cd5.3507.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:129e2846aa4a442120924ff1dcd5ac816254453c356e816f2b5408868cad1073
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.8,
-    "total_flos": 7.501244449451129e+18,
-    "train_loss": 0.7515069055557251,
-    "train_runtime": 2638.4717,
-    "train_samples_per_second": 30.923,
-    "train_steps_per_second": 0.095
+    "epoch": 8.11,
+    "total_flos": 3.675135085585367e+17,
+    "train_loss": 0.0,
+    "train_runtime": 249.2359,
+    "train_samples_per_second": 47.345,
+    "train_steps_per_second": 0.12
 }

trainer_state.json

@@ -1,268 +1,132 @@
 {
-  "best_metric": 0.8715686274509804,
-  "best_model_checkpoint": "resnet-101-finetuned-CivilEng11k/checkpoint-204",
-  "epoch": 9.803921568627452,
+  "best_metric": 0.0,
+  "best_model_checkpoint": "resnet-101-finetuned-CivilEng11k/checkpoint-3",
+  "epoch": 8.108108108108109,
   "eval_steps": 500,
-  "global_step": 250,
+  "global_step": 30,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.39,
-      "learning_rate": 0.00011999999999999999,
-      "loss": 2.6064,
-      "step": 10
-    },
-    {
-      "epoch": 0.78,
-      "learning_rate": 0.00023999999999999998,
-      "loss": 2.4679,
-      "step": 20
-    },
-    {
-      "epoch": 0.98,
-      "eval_accuracy": 0.5367647058823529,
-      "eval_loss": 1.9448940753936768,
-      "eval_runtime": 41.2734,
-      "eval_samples_per_second": 49.426,
-      "eval_steps_per_second": 1.551,
-      "step": 25
-    },
-    {
-      "epoch": 1.18,
-      "learning_rate": 0.00029333333333333327,
-      "loss": 1.9417,
-      "step": 30
-    },
-    {
-      "epoch": 1.57,
-      "learning_rate": 0.00028,
-      "loss": 1.3729,
-      "step": 40
-    },
-    {
-      "epoch": 1.96,
-      "learning_rate": 0.0002666666666666666,
-      "loss": 1.2247,
-      "step": 50
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.6799019607843138,
-      "eval_loss": 1.1000670194625854,
-      "eval_runtime": 42.9676,
-      "eval_samples_per_second": 47.478,
-      "eval_steps_per_second": 1.489,
-      "step": 51
-    },
-    {
-      "epoch": 2.35,
-      "learning_rate": 0.00025333333333333333,
-      "loss": 1.062,
-      "step": 60
-    },
-    {
-      "epoch": 2.75,
-      "learning_rate": 0.00023999999999999998,
-      "loss": 0.8434,
-      "step": 70
-    },
-    {
-      "epoch": 2.98,
-      "eval_accuracy": 0.7887254901960784,
-      "eval_loss": 0.5851422548294067,
-      "eval_runtime": 40.7115,
-      "eval_samples_per_second": 50.109,
-      "eval_steps_per_second": 1.572,
-      "step": 76
-    },
-    {
-      "epoch": 3.14,
-      "learning_rate": 0.00022666666666666663,
-      "loss": 0.6665,
-      "step": 80
-    },
-    {
-      "epoch": 3.53,
-      "learning_rate": 0.00021333333333333333,
-      "loss": 0.5464,
-      "step": 90
-    },
-    {
-      "epoch": 3.92,
-      "learning_rate": 0.00019999999999999998,
-      "loss": 0.491,
-      "step": 100
+      "epoch": 0.81,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.4471,
+      "eval_samples_per_second": 54.157,
+      "eval_steps_per_second": 1.836,
+      "step": 3
     },
     {
-      "epoch": 4.0,
-      "eval_accuracy": 0.8338235294117647,
-      "eval_loss": 0.43349361419677734,
-      "eval_runtime": 41.9829,
-      "eval_samples_per_second": 48.591,
-      "eval_steps_per_second": 1.524,
-      "step": 102
+      "epoch": 1.89,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.2566,
+      "eval_samples_per_second": 56.12,
+      "eval_steps_per_second": 1.902,
+      "step": 7
     },
     {
-      "epoch": 4.31,
-      "learning_rate": 0.00018666666666666666,
-      "loss": 0.4475,
-      "step": 110
-    },
-    {
-      "epoch": 4.71,
-      "learning_rate": 0.0001733333333333333,
-      "loss": 0.4512,
-      "step": 120
-    },
-    {
-      "epoch": 4.98,
-      "eval_accuracy": 0.8401960784313726,
-      "eval_loss": 0.40097689628601074,
-      "eval_runtime": 40.6062,
-      "eval_samples_per_second": 50.239,
-      "eval_steps_per_second": 1.576,
-      "step": 127
-    },
-    {
-      "epoch": 5.1,
-      "learning_rate": 0.00015999999999999999,
-      "loss": 0.4392,
-      "step": 130
-    },
-    {
-      "epoch": 5.49,
-      "learning_rate": 0.00014666666666666664,
-      "loss": 0.4147,
-      "step": 140
-    },
-    {
-      "epoch": 5.88,
-      "learning_rate": 0.0001333333333333333,
-      "loss": 0.3891,
-      "step": 150
+      "epoch": 2.7,
+      "grad_norm": NaN,
+      "learning_rate": 0.00022222222222222218,
+      "loss": 0.0,
+      "step": 10
     },
     {
-      "epoch": 6.0,
-      "eval_accuracy": 0.8544117647058823,
-      "eval_loss": 0.38171347975730896,
-      "eval_runtime": 43.7502,
-      "eval_samples_per_second": 46.628,
-      "eval_steps_per_second": 1.463,
-      "step": 153
+      "epoch": 2.97,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.2368,
+      "eval_samples_per_second": 56.332,
+      "eval_steps_per_second": 1.91,
+      "step": 11
     },
     {
-      "epoch": 6.27,
-      "learning_rate": 0.00011999999999999999,
-      "loss": 0.3902,
-      "step": 160
+      "epoch": 3.78,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.1779,
+      "eval_samples_per_second": 56.973,
+      "eval_steps_per_second": 1.931,
+      "step": 14
     },
     {
-      "epoch": 6.67,
-      "learning_rate": 0.00010666666666666667,
-      "loss": 0.3533,
-      "step": 170
+      "epoch": 4.86,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.1256,
+      "eval_samples_per_second": 57.555,
+      "eval_steps_per_second": 1.951,
+      "step": 18
     },
     {
-      "epoch": 6.98,
-      "eval_accuracy": 0.8602941176470589,
-      "eval_loss": 0.3679962158203125,
-      "eval_runtime": 38.6694,
-      "eval_samples_per_second": 52.755,
-      "eval_steps_per_second": 1.655,
-      "step": 178
+      "epoch": 5.41,
+      "grad_norm": NaN,
+      "learning_rate": 0.00011111111111111109,
+      "loss": 0.0,
+      "step": 20
     },
     {
-      "epoch": 7.06,
-      "learning_rate": 9.333333333333333e-05,
-      "loss": 0.3553,
-      "step": 180
+      "epoch": 5.95,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.1976,
+      "eval_samples_per_second": 56.757,
+      "eval_steps_per_second": 1.924,
+      "step": 22
     },
     {
-      "epoch": 7.45,
-      "learning_rate": 7.999999999999999e-05,
-      "loss": 0.352,
-      "step": 190
+      "epoch": 6.76,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.1726,
+      "eval_samples_per_second": 57.032,
+      "eval_steps_per_second": 1.933,
+      "step": 25
     },
     {
       "epoch": 7.84,
-      "learning_rate": 6.666666666666666e-05,
-      "loss": 0.3407,
-      "step": 200
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.0794,
+      "eval_samples_per_second": 58.078,
+      "eval_steps_per_second": 1.969,
+      "step": 29
     },
     {
-      "epoch": 8.0,
-      "eval_accuracy": 0.8715686274509804,
-      "eval_loss": 0.3589998185634613,
-      "eval_runtime": 31.761,
-      "eval_samples_per_second": 64.23,
-      "eval_steps_per_second": 2.015,
-      "step": 204
-    },
-    {
-      "epoch": 8.24,
-      "learning_rate": 5.333333333333333e-05,
-      "loss": 0.351,
-      "step": 210
-    },
-    {
-      "epoch": 8.63,
-      "learning_rate": 3.9999999999999996e-05,
-      "loss": 0.3244,
-      "step": 220
-    },
-    {
-      "epoch": 8.98,
-      "eval_accuracy": 0.8661764705882353,
-      "eval_loss": 0.36056193709373474,
-      "eval_runtime": 34.9574,
-      "eval_samples_per_second": 58.357,
-      "eval_steps_per_second": 1.831,
-      "step": 229
-    },
-    {
-      "epoch": 9.02,
-      "learning_rate": 2.6666666666666667e-05,
-      "loss": 0.3148,
-      "step": 230
-    },
-    {
-      "epoch": 9.41,
-      "learning_rate": 1.3333333333333333e-05,
-      "loss": 0.3156,
-      "step": 240
-    },
-    {
-      "epoch": 9.8,
+      "epoch": 8.11,
+      "grad_norm": NaN,
       "learning_rate": 0.0,
-      "loss": 0.3258,
-      "step": 250
+      "loss": 0.0,
+      "step": 30
     },
     {
-      "epoch": 9.8,
-      "eval_accuracy": 0.8691176470588236,
-      "eval_loss": 0.3549394905567169,
-      "eval_runtime": 38.198,
-      "eval_samples_per_second": 53.406,
-      "eval_steps_per_second": 1.675,
-      "step": 250
+      "epoch": 8.11,
+      "eval_accuracy": 0.0,
+      "eval_loss": NaN,
+      "eval_runtime": 5.2939,
+      "eval_samples_per_second": 55.725,
+      "eval_steps_per_second": 1.889,
+      "step": 30
     },
     {
-      "epoch": 9.8,
-      "step": 250,
-      "total_flos": 7.501244449451129e+18,
-      "train_loss": 0.7515069055557251,
-      "train_runtime": 2638.4717,
-      "train_samples_per_second": 30.923,
-      "train_steps_per_second": 0.095
+      "epoch": 8.11,
+      "step": 30,
+      "total_flos": 3.675135085585367e+17,
+      "train_loss": 0.0,
+      "train_runtime": 249.2359,
+      "train_samples_per_second": 47.345,
+      "train_steps_per_second": 0.12
     }
   ],
   "logging_steps": 10,
-  "max_steps": 250,
+  "max_steps": 30,
+  "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 7.501244449451129e+18,
+  "total_flos": 3.675135085585367e+17,
+  "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null
 }