End of training

all_results.json

@@ -1,8 +1,13 @@
 {
-    "epoch": 2.95,
-    "total_flos": 1.4272188589307658e+18,
-    "train_loss": 0.7287208272351159,
-    "train_runtime": 993.4233,
-    "train_samples_per_second": 18.817,
-    "train_steps_per_second": 0.145
+    "epoch": 4.92,
+    "eval_accuracy": 0.9845857418111753,
+    "eval_loss": 0.07430984824895859,
+    "eval_runtime": 60.6008,
+    "eval_samples_per_second": 51.385,
+    "eval_steps_per_second": 1.617,
+    "total_flos": 2.378233054601478e+18,
+    "train_loss": 0.6406697725256284,
+    "train_runtime": 1704.9827,
+    "train_samples_per_second": 18.273,
+    "train_steps_per_second": 0.141
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.92,
+    "eval_accuracy": 0.9845857418111753,
+    "eval_loss": 0.07430984824895859,
+    "eval_runtime": 60.6008,
+    "eval_samples_per_second": 51.385,
+    "eval_steps_per_second": 1.617
+}

runs/Mar29_15-56-05_f7c26454068d/events.out.tfevents.1680107179.f7c26454068d.1677.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:009275fe92ccff7d88487c22e47bf27cb3989f7a6624a64017d55e70dce3b947
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.95,
-    "total_flos": 1.4272188589307658e+18,
-    "train_loss": 0.7287208272351159,
-    "train_runtime": 993.4233,
-    "train_samples_per_second": 18.817,
-    "train_steps_per_second": 0.145
+    "epoch": 4.92,
+    "total_flos": 2.378233054601478e+18,
+    "train_loss": 0.6406697725256284,
+    "train_runtime": 1704.9827,
+    "train_samples_per_second": 18.273,
+    "train_steps_per_second": 0.141
 }

trainer_state.json

@@ -1,136 +1,214 @@
 {
-  "best_metric": 0.9605009633911368,
-  "best_model_checkpoint": "clip-vit-large-patch14-finetuned-fruits-360_vitlarge/checkpoint-144",
-  "epoch": 2.953846153846154,
-  "global_step": 144,
+  "best_metric": 0.9845857418111753,
+  "best_model_checkpoint": "clip-vit-large-patch14-finetuned-fruits-360_vitlarge/checkpoint-240",
+  "epoch": 4.923076923076923,
+  "global_step": 240,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.21,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 1.991,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 2.7421,
       "step": 10
     },
     {
       "epoch": 0.41,
-      "learning_rate": 4.8062015503875976e-05,
-      "loss": 1.4175,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 1.7164,
       "step": 20
     },
     {
       "epoch": 0.62,
-      "learning_rate": 4.418604651162791e-05,
-      "loss": 1.0701,
+      "learning_rate": 4.8611111111111115e-05,
+      "loss": 1.3248,
       "step": 30
     },
     {
       "epoch": 0.82,
-      "learning_rate": 4.0310077519379843e-05,
-      "loss": 0.8631,
+      "learning_rate": 4.62962962962963e-05,
+      "loss": 1.0674,
       "step": 40
     },
     {
       "epoch": 0.98,
-      "eval_accuracy": 0.852922286448298,
-      "eval_loss": 0.4619566798210144,
-      "eval_runtime": 62.7979,
-      "eval_samples_per_second": 49.588,
-      "eval_steps_per_second": 1.561,
+      "eval_accuracy": 0.7475915221579962,
+      "eval_loss": 0.6957976222038269,
+      "eval_runtime": 62.2188,
+      "eval_samples_per_second": 50.049,
+      "eval_steps_per_second": 1.575,
       "step": 48
     },
     {
       "epoch": 1.03,
-      "learning_rate": 3.6434108527131784e-05,
-      "loss": 0.6585,
+      "learning_rate": 4.3981481481481486e-05,
+      "loss": 0.8297,
       "step": 50
     },
     {
       "epoch": 1.23,
-      "learning_rate": 3.2558139534883724e-05,
-      "loss": 0.6017,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.718,
       "step": 60
     },
     {
       "epoch": 1.44,
-      "learning_rate": 2.868217054263566e-05,
-      "loss": 0.6272,
+      "learning_rate": 3.935185185185186e-05,
+      "loss": 0.7231,
       "step": 70
     },
     {
       "epoch": 1.64,
-      "learning_rate": 2.48062015503876e-05,
-      "loss": 0.6237,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.6447,
       "step": 80
     },
     {
       "epoch": 1.85,
-      "learning_rate": 2.0930232558139536e-05,
-      "loss": 0.485,
+      "learning_rate": 3.472222222222222e-05,
+      "loss": 0.5475,
       "step": 90
     },
     {
       "epoch": 1.99,
-      "eval_accuracy": 0.9100834938985228,
-      "eval_loss": 0.2783240079879761,
-      "eval_runtime": 62.6573,
-      "eval_samples_per_second": 49.699,
-      "eval_steps_per_second": 1.564,
+      "eval_accuracy": 0.8542068079640334,
+      "eval_loss": 0.44836971163749695,
+      "eval_runtime": 64.139,
+      "eval_samples_per_second": 48.551,
+      "eval_steps_per_second": 1.528,
       "step": 97
     },
     {
       "epoch": 2.05,
-      "learning_rate": 1.7054263565891473e-05,
-      "loss": 0.4887,
+      "learning_rate": 3.240740740740741e-05,
+      "loss": 0.6017,
       "step": 100
     },
     {
       "epoch": 2.26,
-      "learning_rate": 1.3178294573643413e-05,
-      "loss": 0.44,
+      "learning_rate": 3.0092592592592593e-05,
+      "loss": 0.5081,
       "step": 110
     },
     {
       "epoch": 2.46,
-      "learning_rate": 9.302325581395349e-06,
-      "loss": 0.4065,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.4969,
       "step": 120
     },
     {
       "epoch": 2.67,
-      "learning_rate": 5.426356589147287e-06,
-      "loss": 0.3612,
+      "learning_rate": 2.5462962962962965e-05,
+      "loss": 0.4153,
       "step": 130
     },
     {
       "epoch": 2.87,
-      "learning_rate": 1.550387596899225e-06,
-      "loss": 0.3246,
+      "learning_rate": 2.314814814814815e-05,
+      "loss": 0.4065,
       "step": 140
     },
     {
-      "epoch": 2.95,
-      "eval_accuracy": 0.9605009633911368,
-      "eval_loss": 0.17322921752929688,
-      "eval_runtime": 62.2033,
-      "eval_samples_per_second": 50.062,
-      "eval_steps_per_second": 1.575,
-      "step": 144
+      "epoch": 2.99,
+      "eval_accuracy": 0.9274245343609505,
+      "eval_loss": 0.22493986785411835,
+      "eval_runtime": 63.0661,
+      "eval_samples_per_second": 49.377,
+      "eval_steps_per_second": 1.554,
+      "step": 146
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 0.3714,
+      "step": 150
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.3319,
+      "step": 160
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 1.6203703703703704e-05,
+      "loss": 0.3135,
+      "step": 170
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.2645,
+      "step": 180
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 1.1574074074074075e-05,
+      "loss": 0.2386,
+      "step": 190
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9723827874116892,
+      "eval_loss": 0.11541179567575455,
+      "eval_runtime": 63.0456,
+      "eval_samples_per_second": 49.393,
+      "eval_steps_per_second": 1.554,
+      "step": 195
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 9.259259259259259e-06,
+      "loss": 0.2559,
+      "step": 200
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 6.944444444444445e-06,
+      "loss": 0.2543,
+      "step": 210
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 4.6296296296296296e-06,
+      "loss": 0.2274,
+      "step": 220
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 2.3148148148148148e-06,
+      "loss": 0.1793,
+      "step": 230
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 0.0,
+      "loss": 0.197,
+      "step": 240
+    },
+    {
+      "epoch": 4.92,
+      "eval_accuracy": 0.9845857418111753,
+      "eval_loss": 0.07430984824895859,
+      "eval_runtime": 62.3565,
+      "eval_samples_per_second": 49.939,
+      "eval_steps_per_second": 1.572,
+      "step": 240
     },
     {
-      "epoch": 2.95,
-      "step": 144,
-      "total_flos": 1.4272188589307658e+18,
-      "train_loss": 0.7287208272351159,
-      "train_runtime": 993.4233,
-      "train_samples_per_second": 18.817,
-      "train_steps_per_second": 0.145
+      "epoch": 4.92,
+      "step": 240,
+      "total_flos": 2.378233054601478e+18,
+      "train_loss": 0.6406697725256284,
+      "train_runtime": 1704.9827,
+      "train_samples_per_second": 18.273,
+      "train_steps_per_second": 0.141
     }
   ],
-  "max_steps": 144,
-  "num_train_epochs": 3,
-  "total_flos": 1.4272188589307658e+18,
+  "max_steps": 240,
+  "num_train_epochs": 5,
+  "total_flos": 2.378233054601478e+18,
   "trial_name": null,
   "trial_params": null
 }