End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8807785888077859
+      value: 0.8856447688564477
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2599
-- Accuracy: 0.8808
+- Loss: 0.2637
+- Accuracy: 0.8856
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 8.88888888888889,
-    "eval_accuracy": 0.3924731182795699,
-    "eval_loss": 2.05673885345459,
-    "eval_runtime": 21.7695,
-    "eval_samples_per_second": 8.544,
-    "eval_steps_per_second": 0.138,
-    "total_flos": 3.7044103653339955e+17,
-    "train_loss": 2.3554419835408527,
-    "train_runtime": 1715.7637,
-    "train_samples_per_second": 9.733,
-    "train_steps_per_second": 0.035
+    "epoch": 9.836065573770492,
+    "eval_accuracy": 0.8856447688564477,
+    "eval_loss": 0.26367661356925964,
+    "eval_runtime": 46.9001,
+    "eval_samples_per_second": 8.763,
+    "eval_steps_per_second": 0.149,
+    "total_flos": 1.9063087726729052e+18,
+    "train_loss": 0.3090126927693685,
+    "train_runtime": 8089.4797,
+    "train_samples_per_second": 9.637,
+    "train_steps_per_second": 0.037
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.88888888888889,
-    "eval_accuracy": 0.3924731182795699,
-    "eval_loss": 2.05673885345459,
-    "eval_runtime": 21.7695,
-    "eval_samples_per_second": 8.544,
-    "eval_steps_per_second": 0.138
+    "epoch": 9.836065573770492,
+    "eval_accuracy": 0.8856447688564477,
+    "eval_loss": 0.26367661356925964,
+    "eval_runtime": 46.9001,
+    "eval_samples_per_second": 8.763,
+    "eval_steps_per_second": 0.149
 }

runs/Jul04_10-57-49_fc6b70f312e9/events.out.tfevents.1720098814.fc6b70f312e9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e746af3c883d2c7a3c6041506e60c5fce647e40f40c0fb5100a56df368cb103
+size 140

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.88888888888889,
-    "total_flos": 3.7044103653339955e+17,
-    "train_loss": 2.3554419835408527,
-    "train_runtime": 1715.7637,
-    "train_samples_per_second": 9.733,
-    "train_steps_per_second": 0.035
+    "epoch": 9.836065573770492,
+    "total_flos": 1.9063087726729052e+18,
+    "train_loss": 0.3090126927693685,
+    "train_runtime": 8089.4797,
+    "train_samples_per_second": 9.637,
+    "train_steps_per_second": 0.037
 }

trainer_state.json

@@ -1,148 +1,325 @@
 {
-  "best_metric": 0.3924731182795699,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-Mid-NonMidMarket-Classification/checkpoint-47",
-  "epoch": 8.88888888888889,
+  "best_metric": 0.8856447688564477,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-Mid-NonMidMarket-Classification/checkpoint-183",
+  "epoch": 9.836065573770492,
   "eval_steps": 500,
-  "global_step": 60,
+  "global_step": 300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.8888888888888888,
-      "eval_accuracy": 0.13978494623655913,
-      "eval_loss": 2.8975272178649902,
-      "eval_runtime": 21.8981,
-      "eval_samples_per_second": 8.494,
-      "eval_steps_per_second": 0.137,
-      "step": 6
+      "epoch": 0.32786885245901637,
+      "grad_norm": 4.936834812164307,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.9945,
+      "step": 10
     },
     {
-      "epoch": 1.4814814814814814,
-      "grad_norm": 3.0052640438079834,
+      "epoch": 0.6557377049180327,
+      "grad_norm": 2.4745380878448486,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.5619,
+      "step": 20
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "grad_norm": 3.340874433517456,
+      "learning_rate": 5e-05,
+      "loss": 0.4494,
+      "step": 30
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "eval_accuracy": 0.8223844282238443,
+      "eval_loss": 0.495254784822464,
+      "eval_runtime": 47.0869,
+      "eval_samples_per_second": 8.729,
+      "eval_steps_per_second": 0.149,
+      "step": 30
+    },
+    {
+      "epoch": 1.3114754098360657,
+      "grad_norm": 2.102787733078003,
+      "learning_rate": 4.814814814814815e-05,
+      "loss": 0.3852,
+      "step": 40
+    },
+    {
+      "epoch": 1.639344262295082,
+      "grad_norm": 1.9069348573684692,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 2.9658,
-      "step": 10
+      "loss": 0.3536,
+      "step": 50
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "grad_norm": 2.154897451400757,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 0.3369,
+      "step": 60
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8832116788321168,
+      "eval_loss": 0.3206270635128021,
+      "eval_runtime": 46.9296,
+      "eval_samples_per_second": 8.758,
+      "eval_steps_per_second": 0.149,
+      "step": 61
+    },
+    {
+      "epoch": 2.2950819672131146,
+      "grad_norm": 2.829212188720703,
+      "learning_rate": 4.259259259259259e-05,
+      "loss": 0.3171,
+      "step": 70
+    },
+    {
+      "epoch": 2.6229508196721314,
+      "grad_norm": 2.4426212310791016,
+      "learning_rate": 4.074074074074074e-05,
+      "loss": 0.3019,
+      "step": 80
+    },
+    {
+      "epoch": 2.9508196721311473,
+      "grad_norm": 2.601443290710449,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 0.3173,
+      "step": 90
     },
     {
-      "epoch": 1.925925925925926,
-      "eval_accuracy": 0.22043010752688172,
-      "eval_loss": 2.6865546703338623,
-      "eval_runtime": 21.7609,
-      "eval_samples_per_second": 8.547,
-      "eval_steps_per_second": 0.138,
-      "step": 13
+      "epoch": 2.9836065573770494,
+      "eval_accuracy": 0.8734793187347932,
+      "eval_loss": 0.3022870123386383,
+      "eval_runtime": 47.009,
+      "eval_samples_per_second": 8.743,
+      "eval_steps_per_second": 0.149,
+      "step": 91
     },
     {
-      "epoch": 2.962962962962963,
-      "grad_norm": 4.745502948760986,
+      "epoch": 3.278688524590164,
+      "grad_norm": 1.8936508893966675,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 2.6529,
-      "step": 20
+      "loss": 0.3082,
+      "step": 100
     },
     {
-      "epoch": 2.962962962962963,
-      "eval_accuracy": 0.3010752688172043,
-      "eval_loss": 2.4369595050811768,
-      "eval_runtime": 21.5886,
-      "eval_samples_per_second": 8.616,
-      "eval_steps_per_second": 0.139,
-      "step": 20
+      "epoch": 3.6065573770491803,
+      "grad_norm": 2.6445701122283936,
+      "learning_rate": 3.518518518518519e-05,
+      "loss": 0.2837,
+      "step": 110
+    },
+    {
+      "epoch": 3.9344262295081966,
+      "grad_norm": 4.797309875488281,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.2865,
+      "step": 120
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.34946236559139787,
-      "eval_loss": 2.2516093254089355,
-      "eval_runtime": 21.6045,
-      "eval_samples_per_second": 8.609,
-      "eval_steps_per_second": 0.139,
-      "step": 27
+      "eval_accuracy": 0.8734793187347932,
+      "eval_loss": 0.2900189161300659,
+      "eval_runtime": 46.7107,
+      "eval_samples_per_second": 8.799,
+      "eval_steps_per_second": 0.15,
+      "step": 122
+    },
+    {
+      "epoch": 4.262295081967213,
+      "grad_norm": 2.120328664779663,
+      "learning_rate": 3.148148148148148e-05,
+      "loss": 0.2838,
+      "step": 130
+    },
+    {
+      "epoch": 4.590163934426229,
+      "grad_norm": 2.334688663482666,
+      "learning_rate": 2.962962962962963e-05,
+      "loss": 0.2709,
+      "step": 140
     },
     {
-      "epoch": 4.444444444444445,
-      "grad_norm": 6.134488105773926,
+      "epoch": 4.918032786885246,
+      "grad_norm": 2.006906747817993,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 2.3311,
-      "step": 30
+      "loss": 0.2612,
+      "step": 150
+    },
+    {
+      "epoch": 4.983606557377049,
+      "eval_accuracy": 0.8710462287104623,
+      "eval_loss": 0.28622761368751526,
+      "eval_runtime": 47.1214,
+      "eval_samples_per_second": 8.722,
+      "eval_steps_per_second": 0.149,
+      "step": 152
+    },
+    {
+      "epoch": 5.245901639344262,
+      "grad_norm": 2.6484014987945557,
+      "learning_rate": 2.5925925925925925e-05,
+      "loss": 0.2525,
+      "step": 160
+    },
+    {
+      "epoch": 5.573770491803279,
+      "grad_norm": 2.379870891571045,
+      "learning_rate": 2.4074074074074074e-05,
+      "loss": 0.2548,
+      "step": 170
+    },
+    {
+      "epoch": 5.901639344262295,
+      "grad_norm": 2.97287917137146,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.2412,
+      "step": 180
     },
     {
-      "epoch": 4.888888888888889,
-      "eval_accuracy": 0.3709677419354839,
-      "eval_loss": 2.1685309410095215,
-      "eval_runtime": 21.6954,
-      "eval_samples_per_second": 8.573,
-      "eval_steps_per_second": 0.138,
-      "step": 33
+      "epoch": 6.0,
+      "eval_accuracy": 0.8856447688564477,
+      "eval_loss": 0.26367661356925964,
+      "eval_runtime": 46.9018,
+      "eval_samples_per_second": 8.763,
+      "eval_steps_per_second": 0.149,
+      "step": 183
     },
     {
-      "epoch": 5.925925925925926,
-      "grad_norm": 7.691503524780273,
+      "epoch": 6.229508196721311,
+      "grad_norm": 2.9260308742523193,
+      "learning_rate": 2.037037037037037e-05,
+      "loss": 0.246,
+      "step": 190
+    },
+    {
+      "epoch": 6.557377049180328,
+      "grad_norm": 3.2257614135742188,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 2.1441,
-      "step": 40
+      "loss": 0.2434,
+      "step": 200
     },
     {
-      "epoch": 5.925925925925926,
-      "eval_accuracy": 0.3655913978494624,
-      "eval_loss": 2.0986852645874023,
-      "eval_runtime": 21.7432,
-      "eval_samples_per_second": 8.554,
-      "eval_steps_per_second": 0.138,
-      "step": 40
+      "epoch": 6.885245901639344,
+      "grad_norm": 2.4753129482269287,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.2533,
+      "step": 210
     },
     {
-      "epoch": 6.962962962962963,
-      "eval_accuracy": 0.3924731182795699,
-      "eval_loss": 2.05673885345459,
-      "eval_runtime": 21.4068,
-      "eval_samples_per_second": 8.689,
-      "eval_steps_per_second": 0.14,
-      "step": 47
+      "epoch": 6.983606557377049,
+      "eval_accuracy": 0.878345498783455,
+      "eval_loss": 0.27688971161842346,
+      "eval_runtime": 46.8021,
+      "eval_samples_per_second": 8.782,
+      "eval_steps_per_second": 0.15,
+      "step": 213
     },
     {
-      "epoch": 7.407407407407407,
-      "grad_norm": 6.981929302215576,
-      "learning_rate": 9.259259259259259e-06,
-      "loss": 2.0507,
-      "step": 50
+      "epoch": 7.213114754098361,
+      "grad_norm": 2.772411584854126,
+      "learning_rate": 1.4814814814814815e-05,
+      "loss": 0.2399,
+      "step": 220
+    },
+    {
+      "epoch": 7.540983606557377,
+      "grad_norm": 2.4606740474700928,
+      "learning_rate": 1.2962962962962962e-05,
+      "loss": 0.2454,
+      "step": 230
+    },
+    {
+      "epoch": 7.868852459016393,
+      "grad_norm": 2.5486361980438232,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.2413,
+      "step": 240
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.3870967741935484,
-      "eval_loss": 2.041614055633545,
-      "eval_runtime": 21.6481,
-      "eval_samples_per_second": 8.592,
-      "eval_steps_per_second": 0.139,
-      "step": 54
+      "eval_accuracy": 0.8832116788321168,
+      "eval_loss": 0.26485782861709595,
+      "eval_runtime": 47.0631,
+      "eval_samples_per_second": 8.733,
+      "eval_steps_per_second": 0.149,
+      "step": 244
+    },
+    {
+      "epoch": 8.19672131147541,
+      "grad_norm": 4.631506443023682,
+      "learning_rate": 9.259259259259259e-06,
+      "loss": 0.2199,
+      "step": 250
+    },
+    {
+      "epoch": 8.524590163934427,
+      "grad_norm": 2.3710556030273438,
+      "learning_rate": 7.4074074074074075e-06,
+      "loss": 0.2372,
+      "step": 260
+    },
+    {
+      "epoch": 8.852459016393443,
+      "grad_norm": 4.1779608726501465,
+      "learning_rate": 5.555555555555556e-06,
+      "loss": 0.2139,
+      "step": 270
     },
     {
-      "epoch": 8.88888888888889,
-      "grad_norm": 5.651075839996338,
+      "epoch": 8.98360655737705,
+      "eval_accuracy": 0.878345498783455,
+      "eval_loss": 0.2632748484611511,
+      "eval_runtime": 46.7653,
+      "eval_samples_per_second": 8.789,
+      "eval_steps_per_second": 0.15,
+      "step": 274
+    },
+    {
+      "epoch": 9.180327868852459,
+      "grad_norm": 2.4798853397369385,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 0.2365,
+      "step": 280
+    },
+    {
+      "epoch": 9.508196721311476,
+      "grad_norm": 3.2969157695770264,
+      "learning_rate": 1.8518518518518519e-06,
+      "loss": 0.2242,
+      "step": 290
+    },
+    {
+      "epoch": 9.836065573770492,
+      "grad_norm": 2.2613918781280518,
       "learning_rate": 0.0,
-      "loss": 1.988,
-      "step": 60
+      "loss": 0.2087,
+      "step": 300
     },
     {
-      "epoch": 8.88888888888889,
-      "eval_accuracy": 0.3763440860215054,
-      "eval_loss": 2.0368075370788574,
-      "eval_runtime": 21.6178,
-      "eval_samples_per_second": 8.604,
-      "eval_steps_per_second": 0.139,
-      "step": 60
+      "epoch": 9.836065573770492,
+      "eval_accuracy": 0.8807785888077859,
+      "eval_loss": 0.2599342465400696,
+      "eval_runtime": 47.0372,
+      "eval_samples_per_second": 8.738,
+      "eval_steps_per_second": 0.149,
+      "step": 300
     },
     {
-      "epoch": 8.88888888888889,
-      "step": 60,
-      "total_flos": 3.7044103653339955e+17,
-      "train_loss": 2.3554419835408527,
-      "train_runtime": 1715.7637,
-      "train_samples_per_second": 9.733,
-      "train_steps_per_second": 0.035
+      "epoch": 9.836065573770492,
+      "step": 300,
+      "total_flos": 1.9063087726729052e+18,
+      "train_loss": 0.3090126927693685,
+      "train_runtime": 8089.4797,
+      "train_samples_per_second": 9.637,
+      "train_steps_per_second": 0.037
     }
   ],
   "logging_steps": 10,
-  "max_steps": 60,
+  "max_steps": 300,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
@@ -158,7 +335,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.7044103653339955e+17,
+  "total_flos": 1.9063087726729052e+18,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null