End of training

all_results.json

@@ -1,8 +1,13 @@
 {
-    "epoch": 3.0,
-    "total_flos": 1.232080345041961e+17,
-    "train_loss": 2.0565889065082255,
-    "train_runtime": 193.8332,
-    "train_samples_per_second": 25.553,
-    "train_steps_per_second": 0.201
+    "epoch": 15.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.01327650435268879,
+    "eval_runtime": 4.8947,
+    "eval_samples_per_second": 37.592,
+    "eval_steps_per_second": 1.226,
+    "total_flos": 6.160401725209805e+17,
+    "train_loss": 0.1391575378485215,
+    "train_runtime": 1009.255,
+    "train_samples_per_second": 24.538,
+    "train_steps_per_second": 0.193
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.01327650435268879,
+    "eval_runtime": 4.8947,
+    "eval_samples_per_second": 37.592,
+    "eval_steps_per_second": 1.226
+}

runs/Sep15_16-08-50_0e9448f70cc0/events.out.tfevents.1694795267.0e9448f70cc0.5168.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:727e4e90d3fcb55d9b625fe7acb0712069ce68a06cb2de14d45356f1433f565b
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 1.232080345041961e+17,
-    "train_loss": 2.0565889065082255,
-    "train_runtime": 193.8332,
-    "train_samples_per_second": 25.553,
-    "train_steps_per_second": 0.201
+    "epoch": 15.0,
+    "total_flos": 6.160401725209805e+17,
+    "train_loss": 0.1391575378485215,
+    "train_runtime": 1009.255,
+    "train_samples_per_second": 24.538,
+    "train_steps_per_second": 0.193
 }

trainer_state.json

@@ -1,70 +1,274 @@
 {
-  "best_metric": 0.8641304347826086,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-39",
-  "epoch": 3.0,
-  "global_step": 39,
+  "best_metric": 1.0,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-91",
+  "epoch": 15.0,
+  "global_step": 195,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.77,
-      "learning_rate": 4.1428571428571437e-05,
-      "loss": 3.2618,
+      "learning_rate": 2.5e-05,
+      "loss": 0.8235,
       "step": 10
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.483695652173913,
-      "eval_loss": 2.5475988388061523,
-      "eval_runtime": 5.3667,
-      "eval_samples_per_second": 34.285,
-      "eval_steps_per_second": 1.118,
+      "eval_accuracy": 0.9239130434782609,
+      "eval_loss": 0.6033865213394165,
+      "eval_runtime": 4.2784,
+      "eval_samples_per_second": 43.007,
+      "eval_steps_per_second": 1.402,
       "step": 13
     },
     {
       "epoch": 1.54,
-      "learning_rate": 2.714285714285714e-05,
-      "loss": 2.4064,
+      "learning_rate": 5e-05,
+      "loss": 0.5091,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7445652173913043,
-      "eval_loss": 1.3322842121124268,
-      "eval_runtime": 5.3179,
-      "eval_samples_per_second": 34.6,
-      "eval_steps_per_second": 1.128,
+      "eval_accuracy": 0.9728260869565217,
+      "eval_loss": 0.1870061606168747,
+      "eval_runtime": 5.3886,
+      "eval_samples_per_second": 34.146,
+      "eval_steps_per_second": 1.113,
       "step": 26
     },
     {
       "epoch": 2.31,
-      "learning_rate": 1.2857142857142857e-05,
-      "loss": 1.4499,
+      "learning_rate": 4.714285714285714e-05,
+      "loss": 0.273,
       "step": 30
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8641304347826086,
-      "eval_loss": 0.9030408263206482,
-      "eval_runtime": 4.6219,
-      "eval_samples_per_second": 39.81,
-      "eval_steps_per_second": 1.298,
+      "eval_accuracy": 0.9945652173913043,
+      "eval_loss": 0.08950551599264145,
+      "eval_runtime": 4.1088,
+      "eval_samples_per_second": 44.782,
+      "eval_steps_per_second": 1.46,
       "step": 39
     },
     {
-      "epoch": 3.0,
-      "step": 39,
-      "total_flos": 1.232080345041961e+17,
-      "train_loss": 2.0565889065082255,
-      "train_runtime": 193.8332,
-      "train_samples_per_second": 25.553,
-      "train_steps_per_second": 0.201
+      "epoch": 3.08,
+      "learning_rate": 4.428571428571428e-05,
+      "loss": 0.1833,
+      "step": 40
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 4.1428571428571437e-05,
+      "loss": 0.1401,
+      "step": 50
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9945652173913043,
+      "eval_loss": 0.054347552359104156,
+      "eval_runtime": 4.17,
+      "eval_samples_per_second": 44.125,
+      "eval_steps_per_second": 1.439,
+      "step": 52
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.857142857142858e-05,
+      "loss": 0.0936,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9891304347826086,
+      "eval_loss": 0.04841860011219978,
+      "eval_runtime": 4.1427,
+      "eval_samples_per_second": 44.416,
+      "eval_steps_per_second": 1.448,
+      "step": 65
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.091,
+      "step": 70
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9891304347826086,
+      "eval_loss": 0.04977855086326599,
+      "eval_runtime": 4.8364,
+      "eval_samples_per_second": 38.045,
+      "eval_steps_per_second": 1.241,
+      "step": 78
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 3.285714285714286e-05,
+      "loss": 0.0664,
+      "step": 80
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 3e-05,
+      "loss": 0.0603,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.01327650435268879,
+      "eval_runtime": 5.4357,
+      "eval_samples_per_second": 33.85,
+      "eval_steps_per_second": 1.104,
+      "step": 91
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 2.714285714285714e-05,
+      "loss": 0.0421,
+      "step": 100
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9945652173913043,
+      "eval_loss": 0.019574709236621857,
+      "eval_runtime": 5.3751,
+      "eval_samples_per_second": 34.232,
+      "eval_steps_per_second": 1.116,
+      "step": 104
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 2.4285714285714288e-05,
+      "loss": 0.0557,
+      "step": 110
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9945652173913043,
+      "eval_loss": 0.01723380759358406,
+      "eval_runtime": 5.3966,
+      "eval_samples_per_second": 34.096,
+      "eval_steps_per_second": 1.112,
+      "step": 117
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 2.1428571428571428e-05,
+      "loss": 0.0646,
+      "step": 120
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.8571428571428572e-05,
+      "loss": 0.0552,
+      "step": 130
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.010338729247450829,
+      "eval_runtime": 5.2174,
+      "eval_samples_per_second": 35.267,
+      "eval_steps_per_second": 1.15,
+      "step": 130
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 1.5714285714285715e-05,
+      "loss": 0.045,
+      "step": 140
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.00823963712900877,
+      "eval_runtime": 4.9659,
+      "eval_samples_per_second": 37.052,
+      "eval_steps_per_second": 1.208,
+      "step": 143
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 1.2857142857142857e-05,
+      "loss": 0.0355,
+      "step": 150
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.007135531399399042,
+      "eval_runtime": 5.3254,
+      "eval_samples_per_second": 34.551,
+      "eval_steps_per_second": 1.127,
+      "step": 156
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 1e-05,
+      "loss": 0.0491,
+      "step": 160
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.008664855733513832,
+      "eval_runtime": 4.2336,
+      "eval_samples_per_second": 43.462,
+      "eval_steps_per_second": 1.417,
+      "step": 169
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 7.142857142857143e-06,
+      "loss": 0.0395,
+      "step": 170
+    },
+    {
+      "epoch": 13.85,
+      "learning_rate": 4.285714285714286e-06,
+      "loss": 0.0384,
+      "step": 180
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.006527730729430914,
+      "eval_runtime": 4.2794,
+      "eval_samples_per_second": 42.997,
+      "eval_steps_per_second": 1.402,
+      "step": 182
+    },
+    {
+      "epoch": 14.62,
+      "learning_rate": 1.4285714285714286e-06,
+      "loss": 0.0324,
+      "step": 190
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.006102540995925665,
+      "eval_runtime": 4.1839,
+      "eval_samples_per_second": 43.978,
+      "eval_steps_per_second": 1.434,
+      "step": 195
+    },
+    {
+      "epoch": 15.0,
+      "step": 195,
+      "total_flos": 6.160401725209805e+17,
+      "train_loss": 0.1391575378485215,
+      "train_runtime": 1009.255,
+      "train_samples_per_second": 24.538,
+      "train_steps_per_second": 0.193
     }
   ],
-  "max_steps": 39,
-  "num_train_epochs": 3,
-  "total_flos": 1.232080345041961e+17,
+  "max_steps": 195,
+  "num_train_epochs": 15,
+  "total_flos": 6.160401725209805e+17,
   "trial_name": null,
   "trial_params": null
 }