End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8732394366197183
+      value: 0.8802816901408451
 ---
 
 <!-- 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 [Melo1512/vit-msn-small-beta-fia-manually-enhanced-HSV_test_2](https://huggingface.co/Melo1512/vit-msn-small-beta-fia-manually-enhanced-HSV_test_2) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5343
-- Accuracy: 0.8732
+- Loss: 0.5013
+- Accuracy: 0.8803
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 28.571428571428573,
-    "eval_accuracy": 0.8873239436619719,
-    "eval_loss": 0.5122641324996948,
-    "eval_runtime": 0.6466,
-    "eval_samples_per_second": 219.608,
-    "eval_steps_per_second": 4.64,
+    "eval_accuracy": 0.8802816901408451,
+    "eval_loss": 0.5013329386711121,
+    "eval_runtime": 0.6221,
+    "eval_samples_per_second": 228.251,
+    "eval_steps_per_second": 4.822,
     "total_flos": 2.3842598606630093e+17,
-    "train_loss": 0.4234132957458496,
-    "train_runtime": 127.7102,
-    "train_samples_per_second": 166.784,
-    "train_steps_per_second": 0.392
+    "train_loss": 0.4329641246795654,
+    "train_runtime": 127.8111,
+    "train_samples_per_second": 166.652,
+    "train_steps_per_second": 0.391
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 28.571428571428573,
-    "eval_accuracy": 0.8873239436619719,
-    "eval_loss": 0.5122641324996948,
-    "eval_runtime": 0.6466,
-    "eval_samples_per_second": 219.608,
-    "eval_steps_per_second": 4.64
+    "eval_accuracy": 0.8802816901408451,
+    "eval_loss": 0.5013329386711121,
+    "eval_runtime": 0.6221,
+    "eval_samples_per_second": 228.251,
+    "eval_steps_per_second": 4.822
 }

runs/Jan27_17-20-31_db1093ce036b/events.out.tfevents.1737998729.db1093ce036b.224.9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8332d9f0d9802a77fa0c303af28b8ac86cc7cf33a4549c2cad27e24b29a453de
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 28.571428571428573,
     "total_flos": 2.3842598606630093e+17,
-    "train_loss": 0.4234132957458496,
-    "train_runtime": 127.7102,
-    "train_samples_per_second": 166.784,
-    "train_steps_per_second": 0.392
+    "train_loss": 0.4329641246795654,
+    "train_runtime": 127.8111,
+    "train_samples_per_second": 166.652,
+    "train_steps_per_second": 0.391
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.5122641324996948,
-  "best_model_checkpoint": "vit-msn-small-beta-fia-manually-enhanced-HSV_test_3/checkpoint-1",
+  "best_metric": 0.5013329386711121,
+  "best_model_checkpoint": "vit-msn-small-beta-fia-manually-enhanced-HSV_test_3/checkpoint-17",
   "epoch": 28.571428571428573,
   "eval_steps": 500,
   "global_step": 50,
@@ -12,306 +12,306 @@
       "epoch": 0.5714285714285714,
       "eval_accuracy": 0.8873239436619719,
       "eval_loss": 0.5122641324996948,
-      "eval_runtime": 0.5861,
-      "eval_samples_per_second": 242.264,
-      "eval_steps_per_second": 5.118,
+      "eval_runtime": 0.6602,
+      "eval_samples_per_second": 215.093,
+      "eval_steps_per_second": 4.544,
       "step": 1
     },
     {
       "epoch": 1.7142857142857144,
-      "eval_accuracy": 0.8661971830985915,
-      "eval_loss": 0.5494502186775208,
-      "eval_runtime": 0.5821,
-      "eval_samples_per_second": 243.962,
-      "eval_steps_per_second": 5.154,
+      "eval_accuracy": 0.8873239436619719,
+      "eval_loss": 0.521920382976532,
+      "eval_runtime": 0.5873,
+      "eval_samples_per_second": 241.779,
+      "eval_steps_per_second": 5.108,
       "step": 3
     },
     {
       "epoch": 2.857142857142857,
-      "eval_accuracy": 0.8591549295774648,
-      "eval_loss": 0.6079620122909546,
-      "eval_runtime": 0.5831,
-      "eval_samples_per_second": 243.534,
+      "eval_accuracy": 0.8732394366197183,
+      "eval_loss": 0.5431370139122009,
+      "eval_runtime": 0.583,
+      "eval_samples_per_second": 243.549,
       "eval_steps_per_second": 5.145,
       "step": 5
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5590734481811523,
-      "eval_runtime": 0.6248,
-      "eval_samples_per_second": 227.269,
-      "eval_steps_per_second": 4.801,
+      "eval_loss": 0.5444017648696899,
+      "eval_runtime": 0.6551,
+      "eval_samples_per_second": 216.765,
+      "eval_steps_per_second": 4.58,
       "step": 7
     },
     {
       "epoch": 4.571428571428571,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5464029312133789,
-      "eval_runtime": 0.6237,
-      "eval_samples_per_second": 227.68,
-      "eval_steps_per_second": 4.81,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5335883498191833,
+      "eval_runtime": 0.6183,
+      "eval_samples_per_second": 229.646,
+      "eval_steps_per_second": 4.852,
       "step": 8
     },
     {
       "epoch": 5.714285714285714,
-      "grad_norm": 4.793102264404297,
+      "grad_norm": 4.845490455627441,
       "learning_rate": 9.523809523809525e-06,
-      "loss": 0.4241,
+      "loss": 0.4252,
       "step": 10
     },
     {
       "epoch": 5.714285714285714,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.5981650352478027,
-      "eval_runtime": 0.6008,
-      "eval_samples_per_second": 236.347,
-      "eval_steps_per_second": 4.993,
+      "eval_accuracy": 0.8873239436619719,
+      "eval_loss": 0.5234636068344116,
+      "eval_runtime": 0.6093,
+      "eval_samples_per_second": 233.044,
+      "eval_steps_per_second": 4.923,
       "step": 10
     },
     {
       "epoch": 6.857142857142857,
-      "eval_accuracy": 0.8169014084507042,
-      "eval_loss": 0.6497244238853455,
-      "eval_runtime": 0.6066,
-      "eval_samples_per_second": 234.109,
-      "eval_steps_per_second": 4.946,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5268605947494507,
+      "eval_runtime": 0.5951,
+      "eval_samples_per_second": 238.608,
+      "eval_steps_per_second": 5.041,
       "step": 12
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.852112676056338,
-      "eval_loss": 0.5927726030349731,
-      "eval_runtime": 0.627,
-      "eval_samples_per_second": 226.478,
-      "eval_steps_per_second": 4.785,
+      "eval_accuracy": 0.8873239436619719,
+      "eval_loss": 0.5106273889541626,
+      "eval_runtime": 0.576,
+      "eval_samples_per_second": 246.532,
+      "eval_steps_per_second": 5.208,
       "step": 14
     },
     {
       "epoch": 8.571428571428571,
-      "eval_accuracy": 0.852112676056338,
-      "eval_loss": 0.5711137056350708,
-      "eval_runtime": 0.6061,
-      "eval_samples_per_second": 234.301,
-      "eval_steps_per_second": 4.95,
+      "eval_accuracy": 0.8873239436619719,
+      "eval_loss": 0.5047743916511536,
+      "eval_runtime": 0.6164,
+      "eval_samples_per_second": 230.361,
+      "eval_steps_per_second": 4.867,
       "step": 15
     },
     {
       "epoch": 9.714285714285714,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5468315482139587,
-      "eval_runtime": 0.5742,
-      "eval_samples_per_second": 247.304,
-      "eval_steps_per_second": 5.225,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5013329386711121,
+      "eval_runtime": 0.5967,
+      "eval_samples_per_second": 237.985,
+      "eval_steps_per_second": 5.028,
       "step": 17
     },
     {
       "epoch": 10.857142857142858,
-      "eval_accuracy": 0.852112676056338,
-      "eval_loss": 0.5482771992683411,
-      "eval_runtime": 0.5841,
-      "eval_samples_per_second": 243.112,
-      "eval_steps_per_second": 5.136,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5104676485061646,
+      "eval_runtime": 0.6082,
+      "eval_samples_per_second": 233.487,
+      "eval_steps_per_second": 4.933,
       "step": 19
     },
     {
       "epoch": 11.428571428571429,
-      "grad_norm": 5.585962772369385,
+      "grad_norm": 4.298947811126709,
       "learning_rate": 7.1428571428571436e-06,
-      "loss": 0.4152,
+      "loss": 0.4413,
       "step": 20
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.5783097147941589,
-      "eval_runtime": 0.5721,
-      "eval_samples_per_second": 248.198,
-      "eval_steps_per_second": 5.244,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.525558352470398,
+      "eval_runtime": 0.6369,
+      "eval_samples_per_second": 222.939,
+      "eval_steps_per_second": 4.71,
       "step": 21
     },
     {
       "epoch": 12.571428571428571,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.5835375785827637,
-      "eval_runtime": 0.6296,
-      "eval_samples_per_second": 225.543,
-      "eval_steps_per_second": 4.765,
+      "eval_accuracy": 0.8732394366197183,
+      "eval_loss": 0.5303365588188171,
+      "eval_runtime": 0.5782,
+      "eval_samples_per_second": 245.582,
+      "eval_steps_per_second": 5.188,
       "step": 22
     },
     {
       "epoch": 13.714285714285714,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.5668258666992188,
-      "eval_runtime": 0.5869,
-      "eval_samples_per_second": 241.934,
-      "eval_steps_per_second": 5.111,
+      "eval_accuracy": 0.8661971830985915,
+      "eval_loss": 0.5218045115470886,
+      "eval_runtime": 0.5767,
+      "eval_samples_per_second": 246.238,
+      "eval_steps_per_second": 5.202,
       "step": 24
     },
     {
       "epoch": 14.857142857142858,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.555620014667511,
-      "eval_runtime": 0.5858,
-      "eval_samples_per_second": 242.413,
-      "eval_steps_per_second": 5.121,
+      "eval_accuracy": 0.8591549295774648,
+      "eval_loss": 0.5187973380088806,
+      "eval_runtime": 0.5779,
+      "eval_samples_per_second": 245.713,
+      "eval_steps_per_second": 5.191,
       "step": 26
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.5564189553260803,
-      "eval_runtime": 0.5987,
-      "eval_samples_per_second": 237.192,
-      "eval_steps_per_second": 5.011,
+      "eval_accuracy": 0.8591549295774648,
+      "eval_loss": 0.5202071666717529,
+      "eval_runtime": 0.5943,
+      "eval_samples_per_second": 238.943,
+      "eval_steps_per_second": 5.048,
       "step": 28
     },
     {
       "epoch": 16.571428571428573,
-      "eval_accuracy": 0.8450704225352113,
-      "eval_loss": 0.5590547919273376,
-      "eval_runtime": 0.6483,
-      "eval_samples_per_second": 219.036,
-      "eval_steps_per_second": 4.628,
+      "eval_accuracy": 0.8591549295774648,
+      "eval_loss": 0.5252342820167542,
+      "eval_runtime": 0.5791,
+      "eval_samples_per_second": 245.212,
+      "eval_steps_per_second": 5.181,
       "step": 29
     },
     {
       "epoch": 17.142857142857142,
-      "grad_norm": 5.258753299713135,
+      "grad_norm": 4.189113140106201,
       "learning_rate": 4.761904761904762e-06,
-      "loss": 0.4367,
+      "loss": 0.437,
       "step": 30
     },
     {
       "epoch": 17.714285714285715,
       "eval_accuracy": 0.8591549295774648,
-      "eval_loss": 0.5619198679924011,
-      "eval_runtime": 0.6281,
-      "eval_samples_per_second": 226.062,
-      "eval_steps_per_second": 4.776,
+      "eval_loss": 0.5384504199028015,
+      "eval_runtime": 0.6761,
+      "eval_samples_per_second": 210.04,
+      "eval_steps_per_second": 4.437,
       "step": 31
     },
     {
       "epoch": 18.857142857142858,
       "eval_accuracy": 0.8591549295774648,
-      "eval_loss": 0.5809253454208374,
-      "eval_runtime": 0.5953,
-      "eval_samples_per_second": 238.516,
-      "eval_steps_per_second": 5.039,
+      "eval_loss": 0.5456344485282898,
+      "eval_runtime": 0.5817,
+      "eval_samples_per_second": 244.116,
+      "eval_steps_per_second": 5.157,
       "step": 33
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.8661971830985915,
-      "eval_loss": 0.5810067057609558,
-      "eval_runtime": 0.626,
-      "eval_samples_per_second": 226.835,
-      "eval_steps_per_second": 4.792,
+      "eval_accuracy": 0.8732394366197183,
+      "eval_loss": 0.5408889055252075,
+      "eval_runtime": 0.5889,
+      "eval_samples_per_second": 241.115,
+      "eval_steps_per_second": 5.094,
       "step": 35
     },
     {
       "epoch": 20.571428571428573,
       "eval_accuracy": 0.8661971830985915,
-      "eval_loss": 0.5768489837646484,
-      "eval_runtime": 0.682,
-      "eval_samples_per_second": 208.199,
-      "eval_steps_per_second": 4.399,
+      "eval_loss": 0.5374722480773926,
+      "eval_runtime": 0.6624,
+      "eval_samples_per_second": 214.362,
+      "eval_steps_per_second": 4.529,
       "step": 36
     },
     {
       "epoch": 21.714285714285715,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5590782761573792,
-      "eval_runtime": 0.665,
-      "eval_samples_per_second": 213.54,
-      "eval_steps_per_second": 4.511,
+      "eval_accuracy": 0.8661971830985915,
+      "eval_loss": 0.5355854034423828,
+      "eval_runtime": 0.6285,
+      "eval_samples_per_second": 225.953,
+      "eval_steps_per_second": 4.774,
       "step": 38
     },
     {
       "epoch": 22.857142857142858,
-      "grad_norm": 4.620666027069092,
+      "grad_norm": 5.2482404708862305,
       "learning_rate": 2.380952380952381e-06,
-      "loss": 0.4241,
+      "loss": 0.4343,
       "step": 40
     },
     {
       "epoch": 22.857142857142858,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5452097654342651,
-      "eval_runtime": 0.5858,
-      "eval_samples_per_second": 242.391,
-      "eval_steps_per_second": 5.121,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5328315496444702,
+      "eval_runtime": 0.5751,
+      "eval_samples_per_second": 246.919,
+      "eval_steps_per_second": 5.217,
       "step": 40
     },
     {
       "epoch": 24.0,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5387392640113831,
-      "eval_runtime": 0.5833,
-      "eval_samples_per_second": 243.447,
-      "eval_steps_per_second": 5.143,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.531771719455719,
+      "eval_runtime": 0.5972,
+      "eval_samples_per_second": 237.757,
+      "eval_steps_per_second": 5.023,
       "step": 42
     },
     {
       "epoch": 24.571428571428573,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5397770404815674,
-      "eval_runtime": 0.6666,
-      "eval_samples_per_second": 213.023,
-      "eval_steps_per_second": 4.5,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5329975485801697,
+      "eval_runtime": 0.6133,
+      "eval_samples_per_second": 231.529,
+      "eval_steps_per_second": 4.891,
       "step": 43
     },
     {
       "epoch": 25.714285714285715,
-      "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5457538962364197,
-      "eval_runtime": 0.5797,
-      "eval_samples_per_second": 244.962,
-      "eval_steps_per_second": 5.175,
+      "eval_accuracy": 0.8802816901408451,
+      "eval_loss": 0.5333565473556519,
+      "eval_runtime": 0.5803,
+      "eval_samples_per_second": 244.697,
+      "eval_steps_per_second": 5.17,
       "step": 45
     },
     {
       "epoch": 26.857142857142858,
       "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5509300827980042,
-      "eval_runtime": 0.6395,
-      "eval_samples_per_second": 222.035,
-      "eval_steps_per_second": 4.691,
+      "eval_loss": 0.5332355499267578,
+      "eval_runtime": 0.6475,
+      "eval_samples_per_second": 219.301,
+      "eval_steps_per_second": 4.633,
       "step": 47
     },
     {
       "epoch": 28.0,
       "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5549753904342651,
-      "eval_runtime": 0.5878,
-      "eval_samples_per_second": 241.566,
-      "eval_steps_per_second": 5.103,
+      "eval_loss": 0.5341118574142456,
+      "eval_runtime": 0.5963,
+      "eval_samples_per_second": 238.119,
+      "eval_steps_per_second": 5.031,
       "step": 49
     },
     {
       "epoch": 28.571428571428573,
-      "grad_norm": 4.701329708099365,
+      "grad_norm": 4.288495063781738,
       "learning_rate": 0.0,
-      "loss": 0.4171,
+      "loss": 0.4271,
       "step": 50
     },
     {
       "epoch": 28.571428571428573,
       "eval_accuracy": 0.8732394366197183,
-      "eval_loss": 0.5557973980903625,
-      "eval_runtime": 0.5942,
-      "eval_samples_per_second": 238.984,
-      "eval_steps_per_second": 5.049,
+      "eval_loss": 0.5343306064605713,
+      "eval_runtime": 0.6337,
+      "eval_samples_per_second": 224.087,
+      "eval_steps_per_second": 4.734,
       "step": 50
     },
     {
       "epoch": 28.571428571428573,
       "step": 50,
       "total_flos": 2.3842598606630093e+17,
-      "train_loss": 0.4234132957458496,
-      "train_runtime": 127.7102,
-      "train_samples_per_second": 166.784,
-      "train_steps_per_second": 0.392
+      "train_loss": 0.4329641246795654,
+      "train_runtime": 127.8111,
+      "train_samples_per_second": 166.652,
+      "train_steps_per_second": 0.391
     }
   ],
   "logging_steps": 10,