Training in progress, step 100

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.8095238095238095,
-    "eval_loss": 0.5754128098487854,
-    "eval_runtime": 1.4452,
-    "eval_samples_per_second": 14.531,
-    "eval_steps_per_second": 2.076,
-    "train_loss": 0.10162481290883281,
-    "train_runtime": 67.7444,
-    "train_samples_per_second": 10.333,
-    "train_steps_per_second": 2.657
+    "eval_accuracy": 0.8181818181818182,
+    "eval_loss": 0.7126299142837524,
+    "eval_runtime": 1.7324,
+    "eval_samples_per_second": 19.048,
+    "eval_steps_per_second": 2.886,
+    "train_loss": 0.1071862239856273,
+    "train_runtime": 140.5527,
+    "train_samples_per_second": 7.826,
+    "train_steps_per_second": 1.992
 }

config.json

@@ -11,15 +11,16 @@
   "id2label": {
     "0": "airport",
     "1": "bathroom",
-    "10": "zoo",
+    "10": "warehouse",
+    "11": "zoo",
     "2": "beach",
     "3": "building_site",
     "4": "factory",
-    "5": "house",
-    "6": "office",
-    "7": "road",
-    "8": "skatepart",
-    "9": "warehouse"
+    "5": "food",
+    "6": "house",
+    "7": "office",
+    "8": "road",
+    "9": "skatepart"
   },
   "image_size": 224,
   "initializer_range": 0.02,
@@ -30,12 +31,13 @@
     "beach": "2",
     "building_site": "3",
     "factory": "4",
-    "house": "5",
-    "office": "6",
-    "road": "7",
-    "skatepart": "8",
-    "warehouse": "9",
-    "zoo": "10"
+    "food": "5",
+    "house": "6",
+    "office": "7",
+    "road": "8",
+    "skatepart": "9",
+    "warehouse": "10",
+    "zoo": "11"
   },
   "layer_norm_eps": 1e-12,
   "model_type": "vit",

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.8095238095238095,
-    "eval_loss": 0.5754128098487854,
-    "eval_runtime": 1.4452,
-    "eval_samples_per_second": 14.531,
-    "eval_steps_per_second": 2.076
+    "eval_accuracy": 0.8181818181818182,
+    "eval_loss": 0.7126299142837524,
+    "eval_runtime": 1.7324,
+    "eval_samples_per_second": 19.048,
+    "eval_steps_per_second": 2.886
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ecf973856cc57db14758c3df45c75f913734fc524dbe970bde54e425009b7607
-size 343296365
+oid sha256:cb7d6d4cc205243a7dd46e5aa1ed9c3eaf175f23b42b9b901fa82aec81c63dfd
+size 343299437

runs/Aug24_10-33-53_surveily-ag-02/events.out.tfevents.1692866292.surveily-ag-02.15198.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd5cd4baca8bae7d76a8e6466d15637a557fcda9a7810bdb6cb232cc1adc3a50
+size 411

runs/Aug24_12-31-14_surveily-ag-02/events.out.tfevents.1692873080.surveily-ag-02.7570.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dd0ba1524f351192fc1468005337fb5398628b521c7f8b603edd34a97a9ea92
+size 6602

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 20.0,
-    "train_loss": 0.10162481290883281,
-    "train_runtime": 67.7444,
-    "train_samples_per_second": 10.333,
-    "train_steps_per_second": 2.657
+    "train_loss": 0.1071862239856273,
+    "train_runtime": 140.5527,
+    "train_samples_per_second": 7.826,
+    "train_steps_per_second": 1.992
 }

trainer_state.json

@@ -1,142 +1,238 @@
 {
-  "best_metric": 0.5754128098487854,
-  "best_model_checkpoint": "./vit-base/checkpoint-100",
+  "best_metric": 0.7126299142837524,
+  "best_model_checkpoint": "./vit-base/checkpoint-200",
   "epoch": 20.0,
-  "global_step": 180,
+  "global_step": 280,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.11,
-      "learning_rate": 0.00018888888888888888,
-      "loss": 1.5669,
+      "epoch": 0.71,
+      "learning_rate": 0.00019357142857142858,
+      "loss": 2.073,
       "step": 10
     },
     {
-      "epoch": 2.22,
-      "learning_rate": 0.00017777777777777779,
-      "loss": 0.2294,
+      "epoch": 1.43,
+      "learning_rate": 0.00018642857142857144,
+      "loss": 0.6674,
       "step": 20
     },
     {
-      "epoch": 3.33,
-      "learning_rate": 0.0001666666666666667,
-      "loss": 0.0267,
+      "epoch": 2.14,
+      "learning_rate": 0.0001792857142857143,
+      "loss": 0.1817,
       "step": 30
     },
     {
-      "epoch": 4.44,
-      "learning_rate": 0.00015555555555555556,
-      "loss": 0.002,
+      "epoch": 2.86,
+      "learning_rate": 0.00017214285714285717,
+      "loss": 0.0578,
       "step": 40
     },
     {
-      "epoch": 5.56,
-      "learning_rate": 0.00014444444444444444,
-      "loss": 0.0008,
+      "epoch": 3.57,
+      "learning_rate": 0.000165,
+      "loss": 0.0041,
       "step": 50
     },
     {
-      "epoch": 6.67,
-      "learning_rate": 0.00013333333333333334,
-      "loss": 0.0005,
+      "epoch": 3.57,
+      "eval_accuracy": 0.7575757575757576,
+      "eval_loss": 0.5183216333389282,
+      "eval_runtime": 1.6943,
+      "eval_samples_per_second": 19.477,
+      "eval_steps_per_second": 2.951,
+      "step": 50
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 0.00015785714285714285,
+      "loss": 0.007,
       "step": 60
     },
     {
-      "epoch": 7.78,
-      "learning_rate": 0.00012222222222222224,
-      "loss": 0.0004,
+      "epoch": 5.0,
+      "learning_rate": 0.0001507142857142857,
+      "loss": 0.0018,
       "step": 70
     },
     {
-      "epoch": 8.89,
-      "learning_rate": 0.00011111111111111112,
-      "loss": 0.0003,
+      "epoch": 5.71,
+      "learning_rate": 0.0001435714285714286,
+      "loss": 0.0009,
       "step": 80
     },
     {
-      "epoch": 10.0,
-      "learning_rate": 0.0001,
-      "loss": 0.0003,
+      "epoch": 6.43,
+      "learning_rate": 0.00013642857142857144,
+      "loss": 0.0007,
       "step": 90
     },
     {
-      "epoch": 11.11,
-      "learning_rate": 8.888888888888889e-05,
-      "loss": 0.0003,
+      "epoch": 7.14,
+      "learning_rate": 0.0001292857142857143,
+      "loss": 0.0006,
       "step": 100
     },
     {
-      "epoch": 11.11,
-      "eval_accuracy": 0.8095238095238095,
-      "eval_loss": 0.5754128098487854,
-      "eval_runtime": 1.0617,
-      "eval_samples_per_second": 19.78,
-      "eval_steps_per_second": 2.826,
+      "epoch": 7.14,
+      "eval_accuracy": 0.8181818181818182,
+      "eval_loss": 0.7554123401641846,
+      "eval_runtime": 1.4254,
+      "eval_samples_per_second": 23.151,
+      "eval_steps_per_second": 3.508,
       "step": 100
     },
     {
-      "epoch": 12.22,
-      "learning_rate": 7.777777777777778e-05,
-      "loss": 0.0002,
+      "epoch": 7.86,
+      "learning_rate": 0.00012214285714285715,
+      "loss": 0.0005,
       "step": 110
     },
     {
-      "epoch": 13.33,
-      "learning_rate": 6.666666666666667e-05,
-      "loss": 0.0002,
+      "epoch": 8.57,
+      "learning_rate": 0.00011499999999999999,
+      "loss": 0.0005,
       "step": 120
     },
     {
-      "epoch": 14.44,
-      "learning_rate": 5.555555555555556e-05,
-      "loss": 0.0002,
+      "epoch": 9.29,
+      "learning_rate": 0.00010785714285714286,
+      "loss": 0.0004,
       "step": 130
     },
     {
-      "epoch": 15.56,
-      "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.0002,
+      "epoch": 10.0,
+      "learning_rate": 0.00010071428571428571,
+      "loss": 0.0004,
       "step": 140
     },
     {
-      "epoch": 16.67,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0002,
+      "epoch": 10.71,
+      "learning_rate": 9.357142857142858e-05,
+      "loss": 0.0004,
       "step": 150
     },
     {
-      "epoch": 17.78,
-      "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.0002,
+      "epoch": 10.71,
+      "eval_accuracy": 0.8181818181818182,
+      "eval_loss": 0.7197360396385193,
+      "eval_runtime": 1.4885,
+      "eval_samples_per_second": 22.17,
+      "eval_steps_per_second": 3.359,
+      "step": 150
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 8.642857142857143e-05,
+      "loss": 0.0004,
       "step": 160
     },
     {
-      "epoch": 18.89,
-      "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.0002,
+      "epoch": 12.14,
+      "learning_rate": 7.928571428571429e-05,
+      "loss": 0.0003,
       "step": 170
     },
     {
-      "epoch": 20.0,
-      "learning_rate": 0.0,
-      "loss": 0.0002,
+      "epoch": 12.86,
+      "learning_rate": 7.214285714285714e-05,
+      "loss": 0.0003,
       "step": 180
     },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 0.0003,
+      "step": 190
+    },
+    {
+      "epoch": 14.29,
+      "learning_rate": 5.785714285714287e-05,
+      "loss": 0.0003,
+      "step": 200
+    },
+    {
+      "epoch": 14.29,
+      "eval_accuracy": 0.8181818181818182,
+      "eval_loss": 0.7126299142837524,
+      "eval_runtime": 1.5758,
+      "eval_samples_per_second": 20.942,
+      "eval_steps_per_second": 3.173,
+      "step": 200
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 5.0714285714285716e-05,
+      "loss": 0.0003,
+      "step": 210
+    },
+    {
+      "epoch": 15.71,
+      "learning_rate": 4.3571428571428576e-05,
+      "loss": 0.0003,
+      "step": 220
+    },
+    {
+      "epoch": 16.43,
+      "learning_rate": 3.642857142857143e-05,
+      "loss": 0.0003,
+      "step": 230
+    },
+    {
+      "epoch": 17.14,
+      "learning_rate": 2.9285714285714288e-05,
+      "loss": 0.0003,
+      "step": 240
+    },
+    {
+      "epoch": 17.86,
+      "learning_rate": 2.214285714285714e-05,
+      "loss": 0.0003,
+      "step": 250
+    },
+    {
+      "epoch": 17.86,
+      "eval_accuracy": 0.8181818181818182,
+      "eval_loss": 0.7098276615142822,
+      "eval_runtime": 1.5439,
+      "eval_samples_per_second": 21.374,
+      "eval_steps_per_second": 3.239,
+      "step": 250
+    },
+    {
+      "epoch": 18.57,
+      "learning_rate": 1.5e-05,
+      "loss": 0.0003,
+      "step": 260
+    },
+    {
+      "epoch": 19.29,
+      "learning_rate": 7.857142857142858e-06,
+      "loss": 0.0003,
+      "step": 270
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7.142857142857143e-07,
+      "loss": 0.0003,
+      "step": 280
+    },
     {
       "epoch": 20.0,
-      "step": 180,
-      "total_flos": 5.42468236068864e+16,
-      "train_loss": 0.10162481290883281,
-      "train_runtime": 67.7444,
-      "train_samples_per_second": 10.333,
-      "train_steps_per_second": 2.657
+      "step": 280,
+      "total_flos": 8.5248064484352e+16,
+      "train_loss": 0.1071862239856273,
+      "train_runtime": 140.5527,
+      "train_samples_per_second": 7.826,
+      "train_steps_per_second": 1.992
     }
   ],
-  "max_steps": 180,
+  "max_steps": 280,
   "num_train_epochs": 20,
-  "total_flos": 5.42468236068864e+16,
+  "total_flos": 8.5248064484352e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5a7444871ccfb3c61a6a37bb117b483095c240103b17ea642d8889548d76bc4
+oid sha256:c49aa2c0a4b308580ecc8e9ea52b2bab03884419dd31d6c0fc3bf7ba3fc18f2b
 size 3963