Training in progress, step 100

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 21.7,
-    "eval_accuracy": 0.5,
-    "eval_loss": 2.711400032043457,
-    "eval_runtime": 0.2578,
-    "eval_samples_per_second": 62.07,
-    "eval_steps_per_second": 7.759,
-    "train_loss": 0.07943551431099574,
-    "train_runtime": 62.6784,
-    "train_samples_per_second": 18.188,
-    "train_steps_per_second": 4.786
+    "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
 }

config.json

@@ -9,19 +9,25 @@
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
-    "0": "building_site",
-    "1": "factory",
-    "2": "office",
-    "3": "warehouse"
+    "0": "beach",
+    "1": "building_site",
+    "2": "factory",
+    "3": "house",
+    "4": "office",
+    "5": "road",
+    "6": "warehouse"
   },
   "image_size": 224,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "building_site": "0",
-    "factory": "1",
-    "office": "2",
-    "warehouse": "3"
+    "beach": "0",
+    "building_site": "1",
+    "factory": "2",
+    "house": "3",
+    "office": "4",
+    "road": "5",
+    "warehouse": "6"
   },
   "layer_norm_eps": 1e-12,
   "model_type": "vit",

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 21.7,
-    "eval_accuracy": 0.5,
-    "eval_loss": 2.711400032043457,
-    "eval_runtime": 0.2578,
-    "eval_samples_per_second": 62.07,
-    "eval_steps_per_second": 7.759
+    "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
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab6f492fe3b19ac8436ab4019664a0b8b788388a1b11ae542003e7f57d0421b7
-size 343274861
+oid sha256:58bf47a49bd13e419fdb78f05195bdae5d5745b089726bee374462295722010a
+size 343284077

runs/Aug16_13-24-09_surveily-ag-02/events.out.tfevents.1692185120.surveily-ag-02.9722.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59bdd2b2c1018e0bdbc1b372969773efb436e10ea638c88558de855e07f5c99f
+size 411

runs/Aug23_12-32-35_surveily-ag-02/events.out.tfevents.1692786760.surveily-ag-02.18931.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08c81596fa50a2980eaafc7870a405b28b05c7c1ef8452624aa87dbb0b0c4204
+size 16815

runs/Aug23_12-37-35_surveily-ag-02/events.out.tfevents.1692787058.surveily-ag-02.18931.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21db9e6b9154e568276b0f0a9a62eba5451281d04454c21e32efd8f8fd0e9cb6
+size 6090

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 20.0,
-    "train_loss": 0.07943551431099574,
-    "train_runtime": 62.6784,
-    "train_samples_per_second": 18.188,
-    "train_steps_per_second": 4.786
+    "train_loss": 0.015958890169858932,
+    "train_runtime": 31.8203,
+    "train_samples_per_second": 25.141,
+    "train_steps_per_second": 6.285
 }

trainer_state.json

@@ -1,232 +1,163 @@
 {
-  "best_metric": 0.9943184852600098,
-  "best_model_checkpoint": "./vit-base/checkpoint-100",
+  "best_metric": 2.153944969177246,
+  "best_model_checkpoint": "./vit-base/checkpoint-200",
   "epoch": 20.0,
-  "global_step": 300,
+  "global_step": 200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.67,
-      "learning_rate": 0.000194,
-      "loss": 1.4506,
+      "epoch": 1.0,
+      "learning_rate": 0.00019,
+      "loss": 0.0007,
       "step": 10
     },
     {
-      "epoch": 1.33,
-      "learning_rate": 0.000188,
-      "loss": 0.7681,
+      "epoch": 2.0,
+      "learning_rate": 0.00018,
+      "loss": 0.1073,
       "step": 20
     },
     {
-      "epoch": 2.0,
-      "learning_rate": 0.00018133333333333334,
-      "loss": 0.1499,
+      "epoch": 3.0,
+      "learning_rate": 0.00017,
+      "loss": 0.205,
       "step": 30
     },
     {
-      "epoch": 2.67,
-      "learning_rate": 0.00017466666666666667,
-      "loss": 0.0079,
+      "epoch": 4.0,
+      "learning_rate": 0.00016,
+      "loss": 0.0034,
       "step": 40
     },
     {
-      "epoch": 3.33,
-      "learning_rate": 0.000168,
-      "loss": 0.0022,
+      "epoch": 5.0,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.0011,
       "step": 50
     },
     {
-      "epoch": 4.0,
-      "learning_rate": 0.00016133333333333334,
-      "loss": 0.0006,
+      "epoch": 6.0,
+      "learning_rate": 0.00014,
+      "loss": 0.001,
       "step": 60
     },
     {
-      "epoch": 4.67,
-      "learning_rate": 0.00015466666666666667,
-      "loss": 0.0004,
+      "epoch": 7.0,
+      "learning_rate": 0.00013000000000000002,
+      "loss": 0.0003,
       "step": 70
     },
     {
-      "epoch": 5.33,
-      "learning_rate": 0.000148,
-      "loss": 0.0003,
+      "epoch": 8.0,
+      "learning_rate": 0.00012,
+      "loss": 0.0,
       "step": 80
     },
     {
-      "epoch": 6.0,
-      "learning_rate": 0.00014133333333333334,
-      "loss": 0.0002,
+      "epoch": 9.0,
+      "learning_rate": 0.00011000000000000002,
+      "loss": 0.0,
       "step": 90
     },
     {
-      "epoch": 6.67,
-      "learning_rate": 0.00013466666666666667,
-      "loss": 0.0002,
+      "epoch": 10.0,
+      "learning_rate": 0.0001,
+      "loss": 0.0,
       "step": 100
     },
     {
-      "epoch": 6.67,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.9943184852600098,
-      "eval_runtime": 0.2625,
-      "eval_samples_per_second": 60.953,
-      "eval_steps_per_second": 7.619,
+      "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,
       "step": 100
     },
     {
-      "epoch": 7.33,
-      "learning_rate": 0.00012800000000000002,
-      "loss": 0.0002,
+      "epoch": 11.0,
+      "learning_rate": 9e-05,
+      "loss": 0.0,
       "step": 110
     },
     {
-      "epoch": 8.0,
-      "learning_rate": 0.00012133333333333335,
-      "loss": 0.0002,
+      "epoch": 12.0,
+      "learning_rate": 8e-05,
+      "loss": 0.0,
       "step": 120
     },
     {
-      "epoch": 8.67,
-      "learning_rate": 0.00011466666666666667,
-      "loss": 0.0002,
+      "epoch": 13.0,
+      "learning_rate": 7e-05,
+      "loss": 0.0,
       "step": 130
     },
     {
-      "epoch": 9.33,
-      "learning_rate": 0.00010800000000000001,
-      "loss": 0.0001,
+      "epoch": 14.0,
+      "learning_rate": 6e-05,
+      "loss": 0.0,
       "step": 140
     },
     {
-      "epoch": 10.0,
-      "learning_rate": 0.00010133333333333335,
-      "loss": 0.0002,
+      "epoch": 15.0,
+      "learning_rate": 5e-05,
+      "loss": 0.0,
       "step": 150
     },
     {
-      "epoch": 10.67,
-      "learning_rate": 9.466666666666667e-05,
-      "loss": 0.0001,
+      "epoch": 16.0,
+      "learning_rate": 4e-05,
+      "loss": 0.0,
       "step": 160
     },
     {
-      "epoch": 11.33,
-      "learning_rate": 8.800000000000001e-05,
-      "loss": 0.0001,
+      "epoch": 17.0,
+      "learning_rate": 3e-05,
+      "loss": 0.0,
       "step": 170
     },
     {
-      "epoch": 12.0,
-      "learning_rate": 8.133333333333334e-05,
-      "loss": 0.0001,
+      "epoch": 18.0,
+      "learning_rate": 2e-05,
+      "loss": 0.0,
       "step": 180
     },
     {
-      "epoch": 12.67,
-      "learning_rate": 7.466666666666667e-05,
-      "loss": 0.0001,
+      "epoch": 19.0,
+      "learning_rate": 1e-05,
+      "loss": 0.0,
       "step": 190
     },
-    {
-      "epoch": 13.33,
-      "learning_rate": 6.800000000000001e-05,
-      "loss": 0.0001,
-      "step": 200
-    },
-    {
-      "epoch": 13.33,
-      "eval_accuracy": 0.75,
-      "eval_loss": 1.0497348308563232,
-      "eval_runtime": 0.2471,
-      "eval_samples_per_second": 64.739,
-      "eval_steps_per_second": 8.092,
-      "step": 200
-    },
-    {
-      "epoch": 14.0,
-      "learning_rate": 6.133333333333334e-05,
-      "loss": 0.0001,
-      "step": 210
-    },
-    {
-      "epoch": 14.67,
-      "learning_rate": 5.466666666666666e-05,
-      "loss": 0.0001,
-      "step": 220
-    },
-    {
-      "epoch": 15.33,
-      "learning_rate": 4.8e-05,
-      "loss": 0.0001,
-      "step": 230
-    },
-    {
-      "epoch": 16.0,
-      "learning_rate": 4.133333333333333e-05,
-      "loss": 0.0001,
-      "step": 240
-    },
-    {
-      "epoch": 16.67,
-      "learning_rate": 3.466666666666667e-05,
-      "loss": 0.0001,
-      "step": 250
-    },
-    {
-      "epoch": 17.33,
-      "learning_rate": 2.8000000000000003e-05,
-      "loss": 0.0001,
-      "step": 260
-    },
-    {
-      "epoch": 18.0,
-      "learning_rate": 2.1333333333333335e-05,
-      "loss": 0.0001,
-      "step": 270
-    },
-    {
-      "epoch": 18.67,
-      "learning_rate": 1.4666666666666668e-05,
-      "loss": 0.0001,
-      "step": 280
-    },
-    {
-      "epoch": 19.33,
-      "learning_rate": 8.000000000000001e-06,
-      "loss": 0.0001,
-      "step": 290
-    },
     {
       "epoch": 20.0,
-      "learning_rate": 1.3333333333333334e-06,
-      "loss": 0.0001,
-      "step": 300
+      "learning_rate": 0.0,
+      "loss": 0.0,
+      "step": 200
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.75,
-      "eval_loss": 1.0663217306137085,
-      "eval_runtime": 0.2864,
-      "eval_samples_per_second": 55.866,
-      "eval_steps_per_second": 6.983,
-      "step": 300
+      "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
     },
     {
       "epoch": 20.0,
-      "step": 300,
-      "total_flos": 8.834324347441152e+16,
-      "train_loss": 0.07943551431099574,
-      "train_runtime": 62.6784,
-      "train_samples_per_second": 18.188,
-      "train_steps_per_second": 4.786
+      "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
     }
   ],
-  "max_steps": 300,
+  "max_steps": 200,
   "num_train_epochs": 20,
-  "total_flos": 8.834324347441152e+16,
+  "total_flos": 6.19947029495808e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65c747ee9f5afbf1e6c18868d1fdc2542bba84912749af8294f8466b3600f253
+oid sha256:70798a76542aeb79707236cf9ca15785b7ac1ccbc99482451f5b0e5a5849715f
 size 3963