End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.9691358024691358,
-    "eval_loss": 0.08843858540058136,
-    "eval_runtime": 4.5137,
-    "eval_samples_per_second": 358.903,
-    "eval_steps_per_second": 11.299,
-    "total_flos": 1.0874439903456461e+18,
-    "train_loss": 0.41079250579340415,
-    "train_runtime": 248.8888,
-    "train_samples_per_second": 175.741,
-    "train_steps_per_second": 1.374
+    "epoch": 2.9822485207100593,
+    "eval_accuracy": 0.9755555555555555,
+    "eval_loss": 0.07779138535261154,
+    "eval_runtime": 33.0348,
+    "eval_samples_per_second": 163.464,
+    "eval_steps_per_second": 5.116,
+    "total_flos": 1.2017076524313477e+18,
+    "train_loss": 0.406993343716576,
+    "train_runtime": 795.2889,
+    "train_samples_per_second": 61.11,
+    "train_steps_per_second": 0.475
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.9691358024691358,
-    "eval_loss": 0.08843858540058136,
-    "eval_runtime": 4.5137,
-    "eval_samples_per_second": 358.903,
-    "eval_steps_per_second": 11.299
+    "epoch": 2.9822485207100593,
+    "eval_accuracy": 0.9755555555555555,
+    "eval_loss": 0.07779138535261154,
+    "eval_runtime": 33.0348,
+    "eval_samples_per_second": 163.464,
+    "eval_steps_per_second": 5.116
 }

runs/Jul05_18-22-56_65d164acd7e3/events.out.tfevents.1720204611.65d164acd7e3.2983.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:255a6348cd216dcaffd87f0a28ca565355276a1404e133dbfb8bf9eec6a3830e
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 1.0874439903456461e+18,
-    "train_loss": 0.41079250579340415,
-    "train_runtime": 248.8888,
-    "train_samples_per_second": 175.741,
-    "train_steps_per_second": 1.374
+    "epoch": 2.9822485207100593,
+    "total_flos": 1.2017076524313477e+18,
+    "train_loss": 0.406993343716576,
+    "train_runtime": 795.2889,
+    "train_samples_per_second": 61.11,
+    "train_steps_per_second": 0.475
 }

trainer_state.json

@@ -1,290 +1,311 @@
 {
-  "best_metric": 0.9691358024691358,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-342",
-  "epoch": 3.0,
+  "best_metric": 0.9755555555555555,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-378",
+  "epoch": 2.9822485207100593,
   "eval_steps": 500,
-  "global_step": 342,
+  "global_step": 378,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.08771929824561403,
-      "grad_norm": 4.784806251525879,
-      "learning_rate": 1.4285714285714285e-05,
-      "loss": 2.2537,
+      "epoch": 0.07889546351084813,
+      "grad_norm": 4.885411262512207,
+      "learning_rate": 1.3157894736842106e-05,
+      "loss": 2.2985,
       "step": 10
     },
     {
-      "epoch": 0.17543859649122806,
-      "grad_norm": 7.325711727142334,
-      "learning_rate": 2.857142857142857e-05,
-      "loss": 1.9328,
+      "epoch": 0.15779092702169625,
+      "grad_norm": 6.575764179229736,
+      "learning_rate": 2.6315789473684212e-05,
+      "loss": 2.0642,
       "step": 20
     },
     {
-      "epoch": 0.2631578947368421,
-      "grad_norm": 12.205706596374512,
-      "learning_rate": 4.2857142857142856e-05,
-      "loss": 1.3033,
+      "epoch": 0.23668639053254437,
+      "grad_norm": 8.125149726867676,
+      "learning_rate": 3.9473684210526316e-05,
+      "loss": 1.5012,
       "step": 30
     },
     {
-      "epoch": 0.3508771929824561,
-      "grad_norm": 21.546316146850586,
-      "learning_rate": 4.9185667752443e-05,
-      "loss": 0.7435,
+      "epoch": 0.3155818540433925,
+      "grad_norm": 11.912463188171387,
+      "learning_rate": 4.970588235294118e-05,
+      "loss": 0.8692,
       "step": 40
     },
     {
-      "epoch": 0.43859649122807015,
-      "grad_norm": 11.416431427001953,
-      "learning_rate": 4.755700325732899e-05,
-      "loss": 0.5723,
+      "epoch": 0.39447731755424065,
+      "grad_norm": 15.561235427856445,
+      "learning_rate": 4.823529411764706e-05,
+      "loss": 0.6062,
       "step": 50
     },
     {
-      "epoch": 0.5263157894736842,
-      "grad_norm": 13.339532852172852,
-      "learning_rate": 4.592833876221499e-05,
-      "loss": 0.4688,
+      "epoch": 0.47337278106508873,
+      "grad_norm": 22.308109283447266,
+      "learning_rate": 4.6764705882352944e-05,
+      "loss": 0.4878,
       "step": 60
     },
     {
-      "epoch": 0.6140350877192983,
-      "grad_norm": 18.226308822631836,
-      "learning_rate": 4.429967426710098e-05,
-      "loss": 0.4296,
+      "epoch": 0.5522682445759369,
+      "grad_norm": 16.376893997192383,
+      "learning_rate": 4.5294117647058826e-05,
+      "loss": 0.4598,
       "step": 70
     },
     {
-      "epoch": 0.7017543859649122,
-      "grad_norm": 24.142580032348633,
-      "learning_rate": 4.2671009771986977e-05,
-      "loss": 0.3557,
+      "epoch": 0.631163708086785,
+      "grad_norm": 20.709991455078125,
+      "learning_rate": 4.382352941176471e-05,
+      "loss": 0.3735,
       "step": 80
     },
     {
-      "epoch": 0.7894736842105263,
-      "grad_norm": 18.747182846069336,
-      "learning_rate": 4.104234527687297e-05,
-      "loss": 0.3364,
+      "epoch": 0.7100591715976331,
+      "grad_norm": 19.054445266723633,
+      "learning_rate": 4.235294117647059e-05,
+      "loss": 0.4128,
       "step": 90
     },
     {
-      "epoch": 0.8771929824561403,
-      "grad_norm": 9.604598999023438,
-      "learning_rate": 3.941368078175896e-05,
-      "loss": 0.379,
+      "epoch": 0.7889546351084813,
+      "grad_norm": 10.85267162322998,
+      "learning_rate": 4.0882352941176474e-05,
+      "loss": 0.3343,
       "step": 100
     },
     {
-      "epoch": 0.9649122807017544,
-      "grad_norm": 9.6405668258667,
-      "learning_rate": 3.778501628664495e-05,
-      "loss": 0.2967,
+      "epoch": 0.8678500986193294,
+      "grad_norm": 12.394430160522461,
+      "learning_rate": 3.9411764705882356e-05,
+      "loss": 0.3172,
       "step": 110
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.9271604938271605,
-      "eval_loss": 0.20734398066997528,
-      "eval_runtime": 4.5343,
-      "eval_samples_per_second": 357.28,
-      "eval_steps_per_second": 11.248,
-      "step": 114
+      "epoch": 0.9467455621301775,
+      "grad_norm": 13.280746459960938,
+      "learning_rate": 3.794117647058824e-05,
+      "loss": 0.2875,
+      "step": 120
     },
     {
-      "epoch": 1.0526315789473684,
-      "grad_norm": 10.606453895568848,
-      "learning_rate": 3.615635179153095e-05,
-      "loss": 0.2506,
-      "step": 120
+      "epoch": 0.9940828402366864,
+      "eval_accuracy": 0.9516666666666667,
+      "eval_loss": 0.15403828024864197,
+      "eval_runtime": 32.6659,
+      "eval_samples_per_second": 165.31,
+      "eval_steps_per_second": 5.174,
+      "step": 126
     },
     {
-      "epoch": 1.1403508771929824,
-      "grad_norm": 13.219770431518555,
-      "learning_rate": 3.452768729641694e-05,
-      "loss": 0.2646,
+      "epoch": 1.0256410256410255,
+      "grad_norm": 11.388362884521484,
+      "learning_rate": 3.6470588235294114e-05,
+      "loss": 0.2601,
       "step": 130
     },
     {
-      "epoch": 1.2280701754385965,
-      "grad_norm": 14.424461364746094,
-      "learning_rate": 3.289902280130293e-05,
-      "loss": 0.3161,
+      "epoch": 1.1045364891518739,
+      "grad_norm": 21.39406394958496,
+      "learning_rate": 3.5e-05,
+      "loss": 0.287,
       "step": 140
     },
     {
-      "epoch": 1.3157894736842106,
-      "grad_norm": 16.468141555786133,
-      "learning_rate": 3.127035830618892e-05,
-      "loss": 0.2296,
+      "epoch": 1.183431952662722,
+      "grad_norm": 14.232524871826172,
+      "learning_rate": 3.352941176470588e-05,
+      "loss": 0.2604,
       "step": 150
     },
     {
-      "epoch": 1.4035087719298245,
-      "grad_norm": 18.474872589111328,
-      "learning_rate": 2.9641693811074923e-05,
-      "loss": 0.252,
+      "epoch": 1.26232741617357,
+      "grad_norm": 9.439920425415039,
+      "learning_rate": 3.205882352941177e-05,
+      "loss": 0.2563,
       "step": 160
     },
     {
-      "epoch": 1.4912280701754386,
-      "grad_norm": 8.512451171875,
-      "learning_rate": 2.8013029315960915e-05,
-      "loss": 0.2526,
+      "epoch": 1.3412228796844181,
+      "grad_norm": 22.91378402709961,
+      "learning_rate": 3.058823529411765e-05,
+      "loss": 0.2376,
       "step": 170
     },
     {
-      "epoch": 1.5789473684210527,
-      "grad_norm": 10.639461517333984,
-      "learning_rate": 2.6384364820846906e-05,
-      "loss": 0.2218,
+      "epoch": 1.4201183431952662,
+      "grad_norm": 26.201648712158203,
+      "learning_rate": 2.9117647058823534e-05,
+      "loss": 0.2354,
       "step": 180
     },
     {
-      "epoch": 1.6666666666666665,
-      "grad_norm": 10.84925651550293,
-      "learning_rate": 2.47557003257329e-05,
-      "loss": 0.2246,
+      "epoch": 1.4990138067061145,
+      "grad_norm": 11.116905212402344,
+      "learning_rate": 2.7647058823529416e-05,
+      "loss": 0.2424,
       "step": 190
     },
     {
-      "epoch": 1.7543859649122808,
-      "grad_norm": 7.055878162384033,
-      "learning_rate": 2.3127035830618892e-05,
-      "loss": 0.2181,
+      "epoch": 1.5779092702169626,
+      "grad_norm": 9.413314819335938,
+      "learning_rate": 2.6176470588235295e-05,
+      "loss": 0.2572,
       "step": 200
     },
     {
-      "epoch": 1.8421052631578947,
-      "grad_norm": 10.151970863342285,
-      "learning_rate": 2.149837133550489e-05,
-      "loss": 0.2336,
+      "epoch": 1.6568047337278107,
+      "grad_norm": 11.392881393432617,
+      "learning_rate": 2.4705882352941178e-05,
+      "loss": 0.227,
       "step": 210
     },
     {
-      "epoch": 1.9298245614035088,
-      "grad_norm": 10.90556812286377,
-      "learning_rate": 1.986970684039088e-05,
-      "loss": 0.2175,
+      "epoch": 1.7357001972386588,
+      "grad_norm": 8.600625038146973,
+      "learning_rate": 2.323529411764706e-05,
+      "loss": 0.2078,
       "step": 220
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.9617283950617284,
-      "eval_loss": 0.1142377108335495,
-      "eval_runtime": 4.4649,
-      "eval_samples_per_second": 362.828,
-      "eval_steps_per_second": 11.422,
-      "step": 228
-    },
-    {
-      "epoch": 2.017543859649123,
-      "grad_norm": 7.411689281463623,
-      "learning_rate": 1.8241042345276872e-05,
-      "loss": 0.2493,
+      "epoch": 1.8145956607495068,
+      "grad_norm": 11.280962944030762,
+      "learning_rate": 2.1764705882352943e-05,
+      "loss": 0.2259,
       "step": 230
     },
     {
-      "epoch": 2.1052631578947367,
-      "grad_norm": 6.981323719024658,
-      "learning_rate": 1.6612377850162867e-05,
-      "loss": 0.1832,
+      "epoch": 1.893491124260355,
+      "grad_norm": 5.870893955230713,
+      "learning_rate": 2.0294117647058825e-05,
+      "loss": 0.1989,
       "step": 240
     },
     {
-      "epoch": 2.192982456140351,
-      "grad_norm": 17.770326614379883,
-      "learning_rate": 1.4983713355048862e-05,
-      "loss": 0.1862,
+      "epoch": 1.972386587771203,
+      "grad_norm": 12.228656768798828,
+      "learning_rate": 1.8823529411764708e-05,
+      "loss": 0.2201,
       "step": 250
     },
     {
-      "epoch": 2.280701754385965,
-      "grad_norm": 14.688613891601562,
-      "learning_rate": 1.3355048859934855e-05,
-      "loss": 0.1725,
+      "epoch": 1.9960552268244576,
+      "eval_accuracy": 0.975,
+      "eval_loss": 0.08536241203546524,
+      "eval_runtime": 32.8385,
+      "eval_samples_per_second": 164.441,
+      "eval_steps_per_second": 5.146,
+      "step": 253
+    },
+    {
+      "epoch": 2.051282051282051,
+      "grad_norm": 7.734664440155029,
+      "learning_rate": 1.735294117647059e-05,
+      "loss": 0.1785,
       "step": 260
     },
     {
-      "epoch": 2.3684210526315788,
-      "grad_norm": 9.546713829040527,
-      "learning_rate": 1.1726384364820847e-05,
-      "loss": 0.2187,
+      "epoch": 2.1301775147928996,
+      "grad_norm": 13.498882293701172,
+      "learning_rate": 1.588235294117647e-05,
+      "loss": 0.2043,
       "step": 270
     },
     {
-      "epoch": 2.456140350877193,
-      "grad_norm": 13.742387771606445,
-      "learning_rate": 1.009771986970684e-05,
-      "loss": 0.211,
+      "epoch": 2.2090729783037477,
+      "grad_norm": 11.542675971984863,
+      "learning_rate": 1.4411764705882352e-05,
+      "loss": 0.1842,
       "step": 280
     },
     {
-      "epoch": 2.543859649122807,
-      "grad_norm": 10.309969902038574,
-      "learning_rate": 8.469055374592833e-06,
-      "loss": 0.1776,
+      "epoch": 2.287968441814596,
+      "grad_norm": 10.036293029785156,
+      "learning_rate": 1.2941176470588238e-05,
+      "loss": 0.1996,
       "step": 290
     },
     {
-      "epoch": 2.6315789473684212,
-      "grad_norm": 11.343367576599121,
-      "learning_rate": 6.840390879478828e-06,
-      "loss": 0.1828,
+      "epoch": 2.366863905325444,
+      "grad_norm": 8.317439079284668,
+      "learning_rate": 1.1470588235294118e-05,
+      "loss": 0.1573,
       "step": 300
     },
     {
-      "epoch": 2.719298245614035,
-      "grad_norm": 12.15509033203125,
-      "learning_rate": 5.211726384364822e-06,
-      "loss": 0.1565,
+      "epoch": 2.445759368836292,
+      "grad_norm": 6.859732627868652,
+      "learning_rate": 1e-05,
+      "loss": 0.1835,
       "step": 310
     },
     {
-      "epoch": 2.807017543859649,
-      "grad_norm": 13.258170127868652,
-      "learning_rate": 3.5830618892508147e-06,
-      "loss": 0.1601,
+      "epoch": 2.52465483234714,
+      "grad_norm": 13.328524589538574,
+      "learning_rate": 8.529411764705883e-06,
+      "loss": 0.187,
       "step": 320
     },
     {
-      "epoch": 2.8947368421052633,
-      "grad_norm": 13.760292053222656,
-      "learning_rate": 1.9543973941368076e-06,
-      "loss": 0.1777,
+      "epoch": 2.603550295857988,
+      "grad_norm": 7.502381801605225,
+      "learning_rate": 7.058823529411765e-06,
+      "loss": 0.1502,
       "step": 330
     },
     {
-      "epoch": 2.982456140350877,
-      "grad_norm": 10.578232765197754,
-      "learning_rate": 3.2573289902280136e-07,
-      "loss": 0.1881,
+      "epoch": 2.6824457593688362,
+      "grad_norm": 14.761018753051758,
+      "learning_rate": 5.588235294117647e-06,
+      "loss": 0.193,
       "step": 340
     },
     {
-      "epoch": 3.0,
-      "eval_accuracy": 0.9691358024691358,
-      "eval_loss": 0.08843858540058136,
-      "eval_runtime": 4.4533,
-      "eval_samples_per_second": 363.776,
-      "eval_steps_per_second": 11.452,
-      "step": 342
+      "epoch": 2.7613412228796843,
+      "grad_norm": 10.346915245056152,
+      "learning_rate": 4.11764705882353e-06,
+      "loss": 0.1617,
+      "step": 350
+    },
+    {
+      "epoch": 2.8402366863905324,
+      "grad_norm": 12.84180736541748,
+      "learning_rate": 2.647058823529412e-06,
+      "loss": 0.1645,
+      "step": 360
+    },
+    {
+      "epoch": 2.9191321499013805,
+      "grad_norm": 7.301783561706543,
+      "learning_rate": 1.1764705882352942e-06,
+      "loss": 0.1714,
+      "step": 370
+    },
+    {
+      "epoch": 2.9822485207100593,
+      "eval_accuracy": 0.9755555555555555,
+      "eval_loss": 0.07779138535261154,
+      "eval_runtime": 33.1571,
+      "eval_samples_per_second": 162.861,
+      "eval_steps_per_second": 5.097,
+      "step": 378
     },
     {
-      "epoch": 3.0,
-      "step": 342,
-      "total_flos": 1.0874439903456461e+18,
-      "train_loss": 0.41079250579340415,
-      "train_runtime": 248.8888,
-      "train_samples_per_second": 175.741,
-      "train_steps_per_second": 1.374
+      "epoch": 2.9822485207100593,
+      "step": 378,
+      "total_flos": 1.2017076524313477e+18,
+      "train_loss": 0.406993343716576,
+      "train_runtime": 795.2889,
+      "train_samples_per_second": 61.11,
+      "train_steps_per_second": 0.475
     }
   ],
   "logging_steps": 10,
-  "max_steps": 342,
+  "max_steps": 378,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -300,7 +321,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.0874439903456461e+18,
+  "total_flos": 1.2017076524313477e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null