Training in progress, step 100

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.625,
-    "eval_loss": 2.153944969177246,
-    "eval_runtime": 0.2984,
-    "eval_samples_per_second": 53.624,
-    "eval_steps_per_second": 6.703,
-    "train_loss": 0.015958890169858932,
-    "train_runtime": 31.8203,
-    "train_samples_per_second": 25.141,
-    "train_steps_per_second": 6.285
+    "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
 }

config.json

@@ -9,25 +9,33 @@
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
-    "0": "beach",
-    "1": "building_site",
-    "2": "factory",
-    "3": "house",
-    "4": "office",
-    "5": "road",
-    "6": "warehouse"
+    "0": "airport",
+    "1": "bathroom",
+    "10": "zoo",
+    "2": "beach",
+    "3": "building_site",
+    "4": "factory",
+    "5": "house",
+    "6": "office",
+    "7": "road",
+    "8": "skatepart",
+    "9": "warehouse"
   },
   "image_size": 224,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "beach": "0",
-    "building_site": "1",
-    "factory": "2",
-    "house": "3",
-    "office": "4",
-    "road": "5",
-    "warehouse": "6"
+    "airport": "0",
+    "bathroom": "1",
+    "beach": "2",
+    "building_site": "3",
+    "factory": "4",
+    "house": "5",
+    "office": "6",
+    "road": "7",
+    "skatepart": "8",
+    "warehouse": "9",
+    "zoo": "10"
   },
   "layer_norm_eps": 1e-12,
   "model_type": "vit",

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.625,
-    "eval_loss": 2.153944969177246,
-    "eval_runtime": 0.2984,
-    "eval_samples_per_second": 53.624,
-    "eval_steps_per_second": 6.703
+    "eval_accuracy": 0.8095238095238095,
+    "eval_loss": 0.5754128098487854,
+    "eval_runtime": 1.4452,
+    "eval_samples_per_second": 14.531,
+    "eval_steps_per_second": 2.076
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:58bf47a49bd13e419fdb78f05195bdae5d5745b089726bee374462295722010a
-size 343284077
+oid sha256:e40848dbe7fa03668b4520e956ead4cd1c9e51ee07832684daa99460d02ac9aa
+size 343296365

runs/Aug23_12-37-35_surveily-ag-02/events.out.tfevents.1692787144.surveily-ag-02.18931.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8935b5b676800f2df1cd92eb7795f4af20c1e0d621b17bf3e8960d835c81715
+size 411

runs/Aug24_10-33-53_surveily-ag-02/events.out.tfevents.1692866126.surveily-ag-02.15198.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21db484db82fa46ed748c199c76e3c4b3e140967f864dac908fe140592e84763
+size 6566

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 20.0,
-    "train_loss": 0.015958890169858932,
-    "train_runtime": 31.8203,
-    "train_samples_per_second": 25.141,
-    "train_steps_per_second": 6.285
+    "train_loss": 0.10162481290883281,
+    "train_runtime": 67.7444,
+    "train_samples_per_second": 10.333,
+    "train_steps_per_second": 2.657
 }

trainer_state.json

@@ -1,163 +1,142 @@
 {
-  "best_metric": 2.153944969177246,
-  "best_model_checkpoint": "./vit-base/checkpoint-200",
+  "best_metric": 0.5754128098487854,
+  "best_model_checkpoint": "./vit-base/checkpoint-100",
   "epoch": 20.0,
-  "global_step": 200,
+  "global_step": 180,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "learning_rate": 0.00019,
-      "loss": 0.0007,
+      "epoch": 1.11,
+      "learning_rate": 0.00018888888888888888,
+      "loss": 1.5669,
       "step": 10
     },
     {
-      "epoch": 2.0,
-      "learning_rate": 0.00018,
-      "loss": 0.1073,
+      "epoch": 2.22,
+      "learning_rate": 0.00017777777777777779,
+      "loss": 0.2294,
       "step": 20
     },
     {
-      "epoch": 3.0,
-      "learning_rate": 0.00017,
-      "loss": 0.205,
+      "epoch": 3.33,
+      "learning_rate": 0.0001666666666666667,
+      "loss": 0.0267,
       "step": 30
     },
     {
-      "epoch": 4.0,
-      "learning_rate": 0.00016,
-      "loss": 0.0034,
+      "epoch": 4.44,
+      "learning_rate": 0.00015555555555555556,
+      "loss": 0.002,
       "step": 40
     },
     {
-      "epoch": 5.0,
-      "learning_rate": 0.00015000000000000001,
-      "loss": 0.0011,
+      "epoch": 5.56,
+      "learning_rate": 0.00014444444444444444,
+      "loss": 0.0008,
       "step": 50
     },
     {
-      "epoch": 6.0,
-      "learning_rate": 0.00014,
-      "loss": 0.001,
+      "epoch": 6.67,
+      "learning_rate": 0.00013333333333333334,
+      "loss": 0.0005,
       "step": 60
     },
     {
-      "epoch": 7.0,
-      "learning_rate": 0.00013000000000000002,
-      "loss": 0.0003,
+      "epoch": 7.78,
+      "learning_rate": 0.00012222222222222224,
+      "loss": 0.0004,
       "step": 70
     },
     {
-      "epoch": 8.0,
-      "learning_rate": 0.00012,
-      "loss": 0.0,
+      "epoch": 8.89,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 0.0003,
       "step": 80
     },
     {
-      "epoch": 9.0,
-      "learning_rate": 0.00011000000000000002,
-      "loss": 0.0,
+      "epoch": 10.0,
+      "learning_rate": 0.0001,
+      "loss": 0.0003,
       "step": 90
     },
     {
-      "epoch": 10.0,
-      "learning_rate": 0.0001,
-      "loss": 0.0,
+      "epoch": 11.11,
+      "learning_rate": 8.888888888888889e-05,
+      "loss": 0.0003,
       "step": 100
     },
     {
-      "epoch": 10.0,
-      "eval_accuracy": 0.625,
-      "eval_loss": 2.155715227127075,
-      "eval_runtime": 0.2927,
-      "eval_samples_per_second": 54.664,
-      "eval_steps_per_second": 6.833,
+      "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,
       "step": 100
     },
     {
-      "epoch": 11.0,
-      "learning_rate": 9e-05,
-      "loss": 0.0,
+      "epoch": 12.22,
+      "learning_rate": 7.777777777777778e-05,
+      "loss": 0.0002,
       "step": 110
     },
     {
-      "epoch": 12.0,
-      "learning_rate": 8e-05,
-      "loss": 0.0,
+      "epoch": 13.33,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 0.0002,
       "step": 120
     },
     {
-      "epoch": 13.0,
-      "learning_rate": 7e-05,
-      "loss": 0.0,
+      "epoch": 14.44,
+      "learning_rate": 5.555555555555556e-05,
+      "loss": 0.0002,
       "step": 130
     },
     {
-      "epoch": 14.0,
-      "learning_rate": 6e-05,
-      "loss": 0.0,
+      "epoch": 15.56,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 0.0002,
       "step": 140
     },
     {
-      "epoch": 15.0,
-      "learning_rate": 5e-05,
-      "loss": 0.0,
+      "epoch": 16.67,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0002,
       "step": 150
     },
     {
-      "epoch": 16.0,
-      "learning_rate": 4e-05,
-      "loss": 0.0,
+      "epoch": 17.78,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.0002,
       "step": 160
     },
     {
-      "epoch": 17.0,
-      "learning_rate": 3e-05,
-      "loss": 0.0,
+      "epoch": 18.89,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.0002,
       "step": 170
     },
-    {
-      "epoch": 18.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0,
-      "step": 180
-    },
-    {
-      "epoch": 19.0,
-      "learning_rate": 1e-05,
-      "loss": 0.0,
-      "step": 190
-    },
     {
       "epoch": 20.0,
       "learning_rate": 0.0,
-      "loss": 0.0,
-      "step": 200
-    },
-    {
-      "epoch": 20.0,
-      "eval_accuracy": 0.625,
-      "eval_loss": 2.153944969177246,
-      "eval_runtime": 0.2762,
-      "eval_samples_per_second": 57.924,
-      "eval_steps_per_second": 7.241,
-      "step": 200
+      "loss": 0.0002,
+      "step": 180
     },
     {
       "epoch": 20.0,
-      "step": 200,
-      "total_flos": 6.19947029495808e+16,
-      "train_loss": 0.015958890169858932,
-      "train_runtime": 31.8203,
-      "train_samples_per_second": 25.141,
-      "train_steps_per_second": 6.285
+      "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
     }
   ],
-  "max_steps": 200,
+  "max_steps": 180,
   "num_train_epochs": 20,
-  "total_flos": 6.19947029495808e+16,
+  "total_flos": 5.42468236068864e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70798a76542aeb79707236cf9ca15785b7ac1ccbc99482451f5b0e5a5849715f
+oid sha256:e5a7444871ccfb3c61a6a37bb117b483095c240103b17ea642d8889548d76bc4
 size 3963