End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.6666666666666665,
-    "eval_accuracy": 0.46875,
+    "epoch": 2.9411764705882355,
+    "eval_accuracy": 0.30833333333333335,
     "eval_loss": NaN,
-    "eval_runtime": 0.3392,
-    "eval_samples_per_second": 283.038,
-    "eval_steps_per_second": 8.845,
-    "total_flos": 2.326039595266867e+16,
+    "eval_runtime": 1.611,
+    "eval_samples_per_second": 223.468,
+    "eval_steps_per_second": 7.449,
+    "total_flos": 9.643372488710554e+16,
     "train_loss": 0.0,
-    "train_runtime": 17.8085,
-    "train_samples_per_second": 145.549,
-    "train_steps_per_second": 1.011
+    "train_runtime": 96.2766,
+    "train_samples_per_second": 100.959,
+    "train_steps_per_second": 0.779
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.6666666666666665,
-    "eval_accuracy": 0.46875,
+    "epoch": 2.9411764705882355,
+    "eval_accuracy": 0.30833333333333335,
     "eval_loss": NaN,
-    "eval_runtime": 0.3392,
-    "eval_samples_per_second": 283.038,
-    "eval_steps_per_second": 8.845
+    "eval_runtime": 1.611,
+    "eval_samples_per_second": 223.468,
+    "eval_steps_per_second": 7.449
 }

runs/Jun06_04-32-49_46a18104521a/events.out.tfevents.1717648477.46a18104521a.1419.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29da9b2c62281e71e206cffc2d09b5e58910cc1885112eb069bd399382d19664
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.6666666666666665,
-    "total_flos": 2.326039595266867e+16,
+    "epoch": 2.9411764705882355,
+    "total_flos": 9.643372488710554e+16,
     "train_loss": 0.0,
-    "train_runtime": 17.8085,
-    "train_samples_per_second": 145.549,
-    "train_steps_per_second": 1.011
+    "train_runtime": 96.2766,
+    "train_samples_per_second": 100.959,
+    "train_steps_per_second": 0.779
 }

trainer_state.json

@@ -1,59 +1,101 @@
 {
-  "best_metric": 0.46875,
-  "best_model_checkpoint": "resnet-18-please-work/checkpoint-6",
-  "epoch": 2.6666666666666665,
+  "best_metric": 0.30833333333333335,
+  "best_model_checkpoint": "resnet-18-please-work/checkpoint-25",
+  "epoch": 2.9411764705882355,
   "eval_steps": 500,
-  "global_step": 18,
+  "global_step": 75,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.8888888888888888,
-      "eval_accuracy": 0.46875,
+      "epoch": 0.39215686274509803,
+      "grad_norm": NaN,
+      "learning_rate": 4.850746268656717e-05,
+      "loss": 0.0,
+      "step": 10
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": NaN,
+      "learning_rate": 4.104477611940299e-05,
+      "loss": 0.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "eval_accuracy": 0.30833333333333335,
       "eval_loss": NaN,
-      "eval_runtime": 0.5279,
-      "eval_samples_per_second": 181.87,
-      "eval_steps_per_second": 5.683,
-      "step": 6
+      "eval_runtime": 2.1345,
+      "eval_samples_per_second": 168.656,
+      "eval_steps_per_second": 5.622,
+      "step": 25
     },
     {
-      "epoch": 1.4814814814814814,
+      "epoch": 1.1764705882352942,
       "grad_norm": NaN,
-      "learning_rate": 2.5e-05,
+      "learning_rate": 3.358208955223881e-05,
       "loss": 0.0,
-      "step": 10
+      "step": 30
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": NaN,
+      "learning_rate": 2.6119402985074626e-05,
+      "loss": 0.0,
+      "step": 40
     },
     {
-      "epoch": 1.925925925925926,
-      "eval_accuracy": 0.46875,
+      "epoch": 1.9607843137254903,
+      "grad_norm": NaN,
+      "learning_rate": 1.865671641791045e-05,
+      "loss": 0.0,
+      "step": 50
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.30833333333333335,
       "eval_loss": NaN,
-      "eval_runtime": 0.3091,
-      "eval_samples_per_second": 310.537,
-      "eval_steps_per_second": 9.704,
-      "step": 13
+      "eval_runtime": 1.8559,
+      "eval_samples_per_second": 193.978,
+      "eval_steps_per_second": 6.466,
+      "step": 51
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": NaN,
+      "learning_rate": 1.119402985074627e-05,
+      "loss": 0.0,
+      "step": 60
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": NaN,
+      "learning_rate": 3.7313432835820893e-06,
+      "loss": 0.0,
+      "step": 70
     },
     {
-      "epoch": 2.6666666666666665,
-      "eval_accuracy": 0.46875,
+      "epoch": 2.9411764705882355,
+      "eval_accuracy": 0.30833333333333335,
       "eval_loss": NaN,
-      "eval_runtime": 0.2963,
-      "eval_samples_per_second": 323.973,
-      "eval_steps_per_second": 10.124,
-      "step": 18
+      "eval_runtime": 1.403,
+      "eval_samples_per_second": 256.584,
+      "eval_steps_per_second": 8.553,
+      "step": 75
     },
     {
-      "epoch": 2.6666666666666665,
-      "step": 18,
-      "total_flos": 2.326039595266867e+16,
+      "epoch": 2.9411764705882355,
+      "step": 75,
+      "total_flos": 9.643372488710554e+16,
       "train_loss": 0.0,
-      "train_runtime": 17.8085,
-      "train_samples_per_second": 145.549,
-      "train_steps_per_second": 1.011
+      "train_runtime": 96.2766,
+      "train_samples_per_second": 100.959,
+      "train_steps_per_second": 0.779
     }
   ],
   "logging_steps": 10,
-  "max_steps": 18,
+  "max_steps": 75,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -69,7 +111,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.326039595266867e+16,
+  "total_flos": 9.643372488710554e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null