End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8937093275488069
+      value: 0.9229934924078091
 ---
 
 <!-- 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: 0.2765
-- Accuracy: 0.8937
+- Loss: 0.2217
+- Accuracy: 0.9230
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 9.846153846153847,
-    "eval_accuracy": 0.8915401301518439,
-    "eval_loss": 0.2895461618900299,
-    "eval_runtime": 86.4574,
-    "eval_samples_per_second": 10.664,
-    "eval_steps_per_second": 0.093,
+    "eval_accuracy": 0.9229934924078091,
+    "eval_loss": 0.22168530523777008,
+    "eval_runtime": 114.2542,
+    "eval_samples_per_second": 8.07,
+    "eval_steps_per_second": 0.07,
     "total_flos": 2.0293244994235208e+18,
-    "train_loss": 0.3741051137447357,
-    "train_runtime": 7958.3559,
-    "train_samples_per_second": 10.415,
-    "train_steps_per_second": 0.02
+    "train_loss": 0.36495583951473237,
+    "train_runtime": 24166.1649,
+    "train_samples_per_second": 3.43,
+    "train_steps_per_second": 0.007
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.846153846153847,
-    "eval_accuracy": 0.8915401301518439,
-    "eval_loss": 0.2895461618900299,
-    "eval_runtime": 86.4574,
-    "eval_samples_per_second": 10.664,
-    "eval_steps_per_second": 0.093
+    "eval_accuracy": 0.9229934924078091,
+    "eval_loss": 0.22168530523777008,
+    "eval_runtime": 114.2542,
+    "eval_samples_per_second": 8.07,
+    "eval_steps_per_second": 0.07
 }

runs/Jul26_19-51-36_affba4c7048c/events.out.tfevents.1722047793.affba4c7048c

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:591ff47fce04a0e54dce90b1763ff50aae8794740f3c9498f299404f15e06107
+size 140

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.846153846153847,
     "total_flos": 2.0293244994235208e+18,
-    "train_loss": 0.3741051137447357,
-    "train_runtime": 7958.3559,
-    "train_samples_per_second": 10.415,
-    "train_steps_per_second": 0.02
+    "train_loss": 0.36495583951473237,
+    "train_runtime": 24166.1649,
+    "train_samples_per_second": 3.43,
+    "train_steps_per_second": 0.007
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8915401301518439,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-MM_Classification/checkpoint-97",
+  "best_metric": 0.8969631236442517,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-MM_Classification/checkpoint-146",
   "epoch": 9.846153846153847,
   "eval_steps": 500,
   "global_step": 160,
@@ -10,214 +10,214 @@
   "log_history": [
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 2.865189790725708,
+      "grad_norm": 2.110320568084717,
       "learning_rate": 3.125e-05,
-      "loss": 1.0635,
+      "loss": 0.9712,
       "step": 10
     },
     {
       "epoch": 0.9846153846153847,
-      "eval_accuracy": 0.6724511930585684,
-      "eval_loss": 0.752371609210968,
-      "eval_runtime": 123.4184,
-      "eval_samples_per_second": 7.471,
-      "eval_steps_per_second": 0.065,
+      "eval_accuracy": 0.7472885032537961,
+      "eval_loss": 0.5833942890167236,
+      "eval_runtime": 119.6965,
+      "eval_samples_per_second": 7.703,
+      "eval_steps_per_second": 0.067,
       "step": 16
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 2.9992876052856445,
+      "grad_norm": 2.4040279388427734,
       "learning_rate": 4.8611111111111115e-05,
-      "loss": 0.689,
+      "loss": 0.6442,
       "step": 20
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 3.645339250564575,
+      "grad_norm": 3.5479025840759277,
       "learning_rate": 4.5138888888888894e-05,
-      "loss": 0.4571,
+      "loss": 0.4751,
       "step": 30
     },
     {
       "epoch": 1.9692307692307693,
-      "eval_accuracy": 0.8741865509761388,
-      "eval_loss": 0.3691757023334503,
-      "eval_runtime": 86.3686,
-      "eval_samples_per_second": 10.675,
-      "eval_steps_per_second": 0.093,
+      "eval_accuracy": 0.8644251626898047,
+      "eval_loss": 0.34649166464805603,
+      "eval_runtime": 122.7615,
+      "eval_samples_per_second": 7.51,
+      "eval_steps_per_second": 0.065,
       "step": 32
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 2.6993319988250732,
+      "grad_norm": 2.6174745559692383,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 0.3819,
+      "loss": 0.3677,
       "step": 40
     },
     {
       "epoch": 2.953846153846154,
-      "eval_accuracy": 0.868763557483731,
-      "eval_loss": 0.34995290637016296,
-      "eval_runtime": 86.4691,
-      "eval_samples_per_second": 10.663,
-      "eval_steps_per_second": 0.093,
+      "eval_accuracy": 0.8785249457700651,
+      "eval_loss": 0.31836390495300293,
+      "eval_runtime": 122.4616,
+      "eval_samples_per_second": 7.529,
+      "eval_steps_per_second": 0.065,
       "step": 48
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 2.87962007522583,
+      "grad_norm": 3.0105929374694824,
       "learning_rate": 3.8194444444444444e-05,
-      "loss": 0.3314,
+      "loss": 0.3472,
       "step": 50
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 1.9204134941101074,
+      "grad_norm": 2.121572256088257,
       "learning_rate": 3.472222222222222e-05,
-      "loss": 0.3278,
+      "loss": 0.3052,
       "step": 60
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8796095444685467,
-      "eval_loss": 0.3158332407474518,
-      "eval_runtime": 86.6345,
-      "eval_samples_per_second": 10.642,
-      "eval_steps_per_second": 0.092,
+      "eval_accuracy": 0.8828633405639913,
+      "eval_loss": 0.30051109194755554,
+      "eval_runtime": 122.6551,
+      "eval_samples_per_second": 7.517,
+      "eval_steps_per_second": 0.065,
       "step": 65
     },
     {
       "epoch": 4.3076923076923075,
-      "grad_norm": 2.5817983150482178,
+      "grad_norm": 2.34260892868042,
       "learning_rate": 3.125e-05,
-      "loss": 0.3014,
+      "loss": 0.3105,
       "step": 70
     },
     {
       "epoch": 4.923076923076923,
-      "grad_norm": 4.615570068359375,
+      "grad_norm": 1.8295146226882935,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.2941,
+      "loss": 0.2935,
       "step": 80
     },
     {
       "epoch": 4.984615384615385,
-      "eval_accuracy": 0.8882863340563991,
-      "eval_loss": 0.28860825300216675,
-      "eval_runtime": 86.3289,
-      "eval_samples_per_second": 10.68,
-      "eval_steps_per_second": 0.093,
+      "eval_accuracy": 0.8796095444685467,
+      "eval_loss": 0.3044392466545105,
+      "eval_runtime": 121.9892,
+      "eval_samples_per_second": 7.558,
+      "eval_steps_per_second": 0.066,
       "step": 81
     },
     {
       "epoch": 5.538461538461538,
-      "grad_norm": 2.7304208278656006,
+      "grad_norm": 2.232234239578247,
       "learning_rate": 2.4305555555555558e-05,
-      "loss": 0.2912,
+      "loss": 0.2811,
       "step": 90
     },
     {
       "epoch": 5.969230769230769,
-      "eval_accuracy": 0.8915401301518439,
-      "eval_loss": 0.2895461618900299,
-      "eval_runtime": 86.653,
-      "eval_samples_per_second": 10.64,
-      "eval_steps_per_second": 0.092,
+      "eval_accuracy": 0.89587852494577,
+      "eval_loss": 0.28295043110847473,
+      "eval_runtime": 121.4895,
+      "eval_samples_per_second": 7.589,
+      "eval_steps_per_second": 0.066,
       "step": 97
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 3.1544618606567383,
+      "grad_norm": 2.462690830230713,
       "learning_rate": 2.0833333333333336e-05,
-      "loss": 0.2859,
+      "loss": 0.2815,
       "step": 100
     },
     {
       "epoch": 6.769230769230769,
-      "grad_norm": 3.098551034927368,
+      "grad_norm": 3.978851556777954,
       "learning_rate": 1.736111111111111e-05,
-      "loss": 0.2575,
+      "loss": 0.2659,
       "step": 110
     },
     {
       "epoch": 6.953846153846154,
-      "eval_accuracy": 0.8839479392624728,
-      "eval_loss": 0.2800944447517395,
-      "eval_runtime": 86.242,
-      "eval_samples_per_second": 10.691,
-      "eval_steps_per_second": 0.093,
+      "eval_accuracy": 0.89587852494577,
+      "eval_loss": 0.2783919870853424,
+      "eval_runtime": 120.5663,
+      "eval_samples_per_second": 7.647,
+      "eval_steps_per_second": 0.066,
       "step": 113
     },
     {
       "epoch": 7.384615384615385,
-      "grad_norm": 2.2395167350769043,
+      "grad_norm": 4.286837577819824,
       "learning_rate": 1.388888888888889e-05,
-      "loss": 0.272,
+      "loss": 0.2649,
       "step": 120
     },
     {
       "epoch": 8.0,
-      "grad_norm": 6.406205177307129,
+      "grad_norm": 4.650554656982422,
       "learning_rate": 1.0416666666666668e-05,
-      "loss": 0.2604,
+      "loss": 0.2584,
       "step": 130
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.886117136659436,
-      "eval_loss": 0.28469106554985046,
-      "eval_runtime": 86.3518,
-      "eval_samples_per_second": 10.677,
-      "eval_steps_per_second": 0.093,
+      "eval_accuracy": 0.8937093275488069,
+      "eval_loss": 0.282360702753067,
+      "eval_runtime": 119.2752,
+      "eval_samples_per_second": 7.73,
+      "eval_steps_per_second": 0.067,
       "step": 130
     },
     {
       "epoch": 8.615384615384615,
-      "grad_norm": 3.8146533966064453,
+      "grad_norm": 4.211384296417236,
       "learning_rate": 6.944444444444445e-06,
-      "loss": 0.2519,
+      "loss": 0.2623,
       "step": 140
     },
     {
       "epoch": 8.984615384615385,
-      "eval_accuracy": 0.8872017353579176,
-      "eval_loss": 0.2804240584373474,
-      "eval_runtime": 86.6439,
-      "eval_samples_per_second": 10.641,
-      "eval_steps_per_second": 0.092,
+      "eval_accuracy": 0.8969631236442517,
+      "eval_loss": 0.27903664112091064,
+      "eval_runtime": 125.6315,
+      "eval_samples_per_second": 7.339,
+      "eval_steps_per_second": 0.064,
       "step": 146
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 2.1230437755584717,
+      "grad_norm": 2.3360867500305176,
       "learning_rate": 3.4722222222222224e-06,
-      "loss": 0.2614,
+      "loss": 0.2617,
       "step": 150
     },
     {
       "epoch": 9.846153846153847,
-      "grad_norm": 1.9836598634719849,
+      "grad_norm": 2.65038800239563,
       "learning_rate": 0.0,
-      "loss": 0.2592,
+      "loss": 0.2487,
       "step": 160
     },
     {
       "epoch": 9.846153846153847,
-      "eval_accuracy": 0.8872017353579176,
-      "eval_loss": 0.27946537733078003,
-      "eval_runtime": 86.4591,
-      "eval_samples_per_second": 10.664,
-      "eval_steps_per_second": 0.093,
+      "eval_accuracy": 0.8937093275488069,
+      "eval_loss": 0.276502400636673,
+      "eval_runtime": 129.1467,
+      "eval_samples_per_second": 7.139,
+      "eval_steps_per_second": 0.062,
       "step": 160
     },
     {
       "epoch": 9.846153846153847,
       "step": 160,
       "total_flos": 2.0293244994235208e+18,
-      "train_loss": 0.3741051137447357,
-      "train_runtime": 7958.3559,
-      "train_samples_per_second": 10.415,
-      "train_steps_per_second": 0.02
+      "train_loss": 0.36495583951473237,
+      "train_runtime": 24166.1649,
+      "train_samples_per_second": 3.43,
+      "train_steps_per_second": 0.007
     }
   ],
   "logging_steps": 10,