🍻 cheers

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 2.63,
-    "eval_accuracy": 0.7109466841850495,
-    "eval_loss": 0.9905902743339539,
-    "eval_runtime": 153.5849,
-    "eval_samples_per_second": 188.313,
-    "eval_steps_per_second": 23.544,
-    "train_loss": 0.8959425466437089,
-    "train_runtime": 9068.08,
-    "train_samples_per_second": 127.578,
-    "train_steps_per_second": 3.987
+    "eval_accuracy": 0.7164442293064104,
+    "eval_loss": 0.99397212266922,
+    "eval_runtime": 156.2682,
+    "eval_samples_per_second": 185.079,
+    "eval_steps_per_second": 23.14,
+    "train_loss": 1.0215468621504935,
+    "train_runtime": 9065.9357,
+    "train_samples_per_second": 127.608,
+    "train_steps_per_second": 3.988
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.63,
-    "eval_accuracy": 0.7109466841850495,
-    "eval_loss": 0.9905902743339539,
-    "eval_runtime": 153.5849,
-    "eval_samples_per_second": 188.313,
-    "eval_steps_per_second": 23.544
+    "eval_accuracy": 0.7164442293064104,
+    "eval_loss": 0.99397212266922,
+    "eval_runtime": 156.2682,
+    "eval_samples_per_second": 185.079,
+    "eval_steps_per_second": 23.14
 }

runs/May29_21-28-35_adrian-development/events.out.tfevents.1685397805.adrian-development.93196.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d88de1a3bf83b0c00bd30e9cbe068672311b4155fba877d3101be05884fe0c39
+size 369

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 2.63,
-    "train_loss": 0.8959425466437089,
-    "train_runtime": 9068.08,
-    "train_samples_per_second": 127.578,
-    "train_steps_per_second": 3.987
+    "train_loss": 1.0215468621504935,
+    "train_runtime": 9065.9357,
+    "train_samples_per_second": 127.608,
+    "train_steps_per_second": 3.988
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9905902743339539,
+  "best_metric": 0.99397212266922,
   "best_model_checkpoint": "Transformers/vit-base-clothing-leafs-example-full-simple/checkpoint-14000",
   "epoch": 2.627575715668649,
   "global_step": 19000,
@@ -9,297 +9,297 @@
   "log_history": [
     {
       "epoch": 0.14,
-      "learning_rate": 7.292559811920896e-05,
-      "loss": 1.6049,
+      "learning_rate": 1.9447379338957268e-05,
+      "loss": 2.1091,
       "step": 1000
     },
     {
       "epoch": 0.14,
-      "eval_accuracy": 0.6462900214369683,
-      "eval_loss": 1.2689716815948486,
-      "eval_runtime": 157.1043,
-      "eval_samples_per_second": 184.094,
-      "eval_steps_per_second": 23.017,
+      "eval_accuracy": 0.6352603554387664,
+      "eval_loss": 1.5707892179489136,
+      "eval_runtime": 154.5668,
+      "eval_samples_per_second": 187.116,
+      "eval_steps_per_second": 23.394,
       "step": 1000
     },
     {
       "epoch": 0.28,
-      "learning_rate": 7.085327064029871e-05,
-      "loss": 1.1874,
+      "learning_rate": 1.8894758677914535e-05,
+      "loss": 1.4147,
       "step": 2000
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.6708042320724708,
-      "eval_loss": 1.1465643644332886,
-      "eval_runtime": 154.03,
-      "eval_samples_per_second": 187.769,
-      "eval_steps_per_second": 23.476,
+      "eval_accuracy": 0.6584606873660189,
+      "eval_loss": 1.3137956857681274,
+      "eval_runtime": 154.1896,
+      "eval_samples_per_second": 187.574,
+      "eval_steps_per_second": 23.452,
       "step": 2000
     },
     {
       "epoch": 0.41,
-      "learning_rate": 6.878094316138846e-05,
-      "loss": 1.1217,
+      "learning_rate": 1.8341584843036927e-05,
+      "loss": 1.2355,
       "step": 3000
     },
     {
       "epoch": 0.41,
-      "eval_accuracy": 0.6792061406541733,
-      "eval_loss": 1.1049177646636963,
-      "eval_runtime": 154.7974,
-      "eval_samples_per_second": 186.838,
-      "eval_steps_per_second": 23.36,
+      "eval_accuracy": 0.6820067768480741,
+      "eval_loss": 1.1873301267623901,
+      "eval_runtime": 155.4058,
+      "eval_samples_per_second": 186.106,
+      "eval_steps_per_second": 23.268,
       "step": 3000
     },
     {
       "epoch": 0.55,
-      "learning_rate": 6.670654128059742e-05,
-      "loss": 1.0743,
+      "learning_rate": 1.7788411008159317e-05,
+      "loss": 1.1718,
       "step": 4000
     },
     {
       "epoch": 0.55,
-      "eval_accuracy": 0.6763017771938317,
-      "eval_loss": 1.091710090637207,
-      "eval_runtime": 154.343,
-      "eval_samples_per_second": 187.388,
-      "eval_steps_per_second": 23.428,
+      "eval_accuracy": 0.6836664131111265,
+      "eval_loss": 1.1535007953643799,
+      "eval_runtime": 155.1856,
+      "eval_samples_per_second": 186.37,
+      "eval_steps_per_second": 23.301,
       "step": 4000
     },
     {
       "epoch": 0.69,
-      "learning_rate": 6.463213939980638e-05,
-      "loss": 1.0546,
+      "learning_rate": 1.7235237173281706e-05,
+      "loss": 1.1154,
       "step": 5000
     },
     {
       "epoch": 0.69,
-      "eval_accuracy": 0.6879883825461587,
-      "eval_loss": 1.0747572183609009,
-      "eval_runtime": 156.4913,
-      "eval_samples_per_second": 184.815,
-      "eval_steps_per_second": 23.107,
+      "eval_accuracy": 0.6977041698361109,
+      "eval_loss": 1.092354655265808,
+      "eval_runtime": 153.5953,
+      "eval_samples_per_second": 188.3,
+      "eval_steps_per_second": 23.542,
       "step": 5000
     },
     {
       "epoch": 0.83,
-      "learning_rate": 6.255981192089614e-05,
-      "loss": 1.0375,
+      "learning_rate": 1.6682063338404095e-05,
+      "loss": 1.0914,
       "step": 6000
     },
     {
       "epoch": 0.83,
-      "eval_accuracy": 0.6867436553488694,
-      "eval_loss": 1.0563393831253052,
-      "eval_runtime": 154.5847,
-      "eval_samples_per_second": 187.095,
-      "eval_steps_per_second": 23.392,
+      "eval_accuracy": 0.7002281999861697,
+      "eval_loss": 1.0666176080703735,
+      "eval_runtime": 153.4627,
+      "eval_samples_per_second": 188.463,
+      "eval_steps_per_second": 23.563,
       "step": 6000
     },
     {
       "epoch": 0.97,
-      "learning_rate": 6.0485410040105095e-05,
-      "loss": 1.0244,
+      "learning_rate": 1.6129995851196242e-05,
+      "loss": 1.052,
       "step": 7000
     },
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.6980153516354333,
-      "eval_loss": 1.034531593322754,
-      "eval_runtime": 155.8798,
-      "eval_samples_per_second": 185.54,
-      "eval_steps_per_second": 23.197,
+      "eval_accuracy": 0.7028559574026693,
+      "eval_loss": 1.0515823364257812,
+      "eval_runtime": 154.4506,
+      "eval_samples_per_second": 187.257,
+      "eval_steps_per_second": 23.412,
       "step": 7000
     },
     {
       "epoch": 1.11,
-      "learning_rate": 5.8411008159314065e-05,
-      "loss": 0.8943,
+      "learning_rate": 1.5576822016318628e-05,
+      "loss": 0.9649,
       "step": 8000
     },
     {
       "epoch": 1.11,
-      "eval_accuracy": 0.7005739575409723,
-      "eval_loss": 1.0278058052062988,
-      "eval_runtime": 156.5423,
-      "eval_samples_per_second": 184.755,
-      "eval_steps_per_second": 23.099,
+      "eval_accuracy": 0.7032708664684323,
+      "eval_loss": 1.0426146984100342,
+      "eval_runtime": 154.313,
+      "eval_samples_per_second": 187.424,
+      "eval_steps_per_second": 23.433,
       "step": 8000
     },
     {
       "epoch": 1.24,
-      "learning_rate": 5.633660627852302e-05,
-      "loss": 0.8324,
+      "learning_rate": 1.502364818144102e-05,
+      "loss": 0.9281,
       "step": 9000
     },
     {
       "epoch": 1.24,
-      "eval_accuracy": 0.7050342299979254,
-      "eval_loss": 1.0215510129928589,
-      "eval_runtime": 154.4075,
-      "eval_samples_per_second": 187.31,
-      "eval_steps_per_second": 23.419,
+      "eval_accuracy": 0.7112578659843718,
+      "eval_loss": 1.027764081954956,
+      "eval_runtime": 154.0041,
+      "eval_samples_per_second": 187.8,
+      "eval_steps_per_second": 23.48,
       "step": 9000
     },
     {
       "epoch": 1.38,
-      "learning_rate": 5.426427879961277e-05,
-      "loss": 0.8411,
+      "learning_rate": 1.4471027520398286e-05,
+      "loss": 0.9131,
       "step": 10000
     },
     {
       "epoch": 1.38,
-      "eval_accuracy": 0.7016803817163405,
-      "eval_loss": 1.0213595628738403,
-      "eval_runtime": 154.6853,
-      "eval_samples_per_second": 186.973,
-      "eval_steps_per_second": 23.376,
+      "eval_accuracy": 0.7106009266302469,
+      "eval_loss": 1.021926760673523,
+      "eval_runtime": 154.1315,
+      "eval_samples_per_second": 187.645,
+      "eval_steps_per_second": 23.46,
       "step": 10000
     },
     {
       "epoch": 1.52,
-      "learning_rate": 5.2189876918821735e-05,
-      "loss": 0.8353,
+      "learning_rate": 1.3917853685520677e-05,
+      "loss": 0.9105,
       "step": 11000
     },
     {
       "epoch": 1.52,
-      "eval_accuracy": 0.704446442154761,
-      "eval_loss": 1.0081199407577515,
-      "eval_runtime": 156.0333,
-      "eval_samples_per_second": 185.358,
-      "eval_steps_per_second": 23.175,
+      "eval_accuracy": 0.7136090173570292,
+      "eval_loss": 1.0093262195587158,
+      "eval_runtime": 155.6003,
+      "eval_samples_per_second": 185.874,
+      "eval_steps_per_second": 23.239,
       "step": 11000
     },
     {
       "epoch": 1.66,
-      "learning_rate": 5.01154750380307e-05,
-      "loss": 0.8307,
+      "learning_rate": 1.3365233024477943e-05,
+      "loss": 0.9139,
       "step": 12000
     },
     {
       "epoch": 1.66,
-      "eval_accuracy": 0.7079040177027868,
-      "eval_loss": 1.0041202306747437,
-      "eval_runtime": 155.1738,
-      "eval_samples_per_second": 186.384,
-      "eval_steps_per_second": 23.303,
+      "eval_accuracy": 0.7157181384413249,
+      "eval_loss": 1.0021320581436157,
+      "eval_runtime": 154.9466,
+      "eval_samples_per_second": 186.658,
+      "eval_steps_per_second": 23.337,
       "step": 12000
     },
     {
       "epoch": 1.8,
-      "learning_rate": 4.8043147559120454e-05,
-      "loss": 0.8297,
+      "learning_rate": 1.2812059189600333e-05,
+      "loss": 0.901,
       "step": 13000
     },
     {
       "epoch": 1.8,
-      "eval_accuracy": 0.7103588963418851,
-      "eval_loss": 0.9978411793708801,
-      "eval_runtime": 150.0364,
-      "eval_samples_per_second": 192.767,
-      "eval_steps_per_second": 24.101,
+      "eval_accuracy": 0.7147500172878778,
+      "eval_loss": 1.0019475221633911,
+      "eval_runtime": 156.866,
+      "eval_samples_per_second": 184.374,
+      "eval_steps_per_second": 23.052,
       "step": 13000
     },
     {
       "epoch": 1.94,
-      "learning_rate": 4.596874567832941e-05,
-      "loss": 0.8095,
+      "learning_rate": 1.22594385285576e-05,
+      "loss": 0.8916,
       "step": 14000
     },
     {
       "epoch": 1.94,
-      "eval_accuracy": 0.7109466841850495,
-      "eval_loss": 0.9905902743339539,
-      "eval_runtime": 153.3635,
-      "eval_samples_per_second": 188.585,
-      "eval_steps_per_second": 23.578,
+      "eval_accuracy": 0.7164442293064104,
+      "eval_loss": 0.99397212266922,
+      "eval_runtime": 155.3636,
+      "eval_samples_per_second": 186.157,
+      "eval_steps_per_second": 23.274,
       "step": 14000
     },
     {
       "epoch": 2.07,
-      "learning_rate": 4.3894343797538375e-05,
-      "loss": 0.7007,
+      "learning_rate": 1.1706817867514867e-05,
+      "loss": 0.8142,
       "step": 15000
     },
     {
       "epoch": 2.07,
-      "eval_accuracy": 0.7077657146808658,
-      "eval_loss": 1.0342744588851929,
-      "eval_runtime": 155.1599,
-      "eval_samples_per_second": 186.401,
-      "eval_steps_per_second": 23.305,
+      "eval_accuracy": 0.7175506534817786,
+      "eval_loss": 1.0117239952087402,
+      "eval_runtime": 155.3204,
+      "eval_samples_per_second": 186.209,
+      "eval_steps_per_second": 23.281,
       "step": 15000
     },
     {
       "epoch": 2.21,
-      "learning_rate": 4.1822016318628124e-05,
-      "loss": 0.5791,
+      "learning_rate": 1.1153644032637257e-05,
+      "loss": 0.7494,
       "step": 16000
     },
     {
       "epoch": 2.21,
-      "eval_accuracy": 0.701472927183459,
-      "eval_loss": 1.0633156299591064,
-      "eval_runtime": 153.3675,
-      "eval_samples_per_second": 188.58,
-      "eval_steps_per_second": 23.577,
+      "eval_accuracy": 0.7156144111748842,
+      "eval_loss": 1.014885663986206,
+      "eval_runtime": 155.0656,
+      "eval_samples_per_second": 186.515,
+      "eval_steps_per_second": 23.319,
       "step": 16000
     },
     {
       "epoch": 2.35,
-      "learning_rate": 3.974761443783709e-05,
-      "loss": 0.5879,
+      "learning_rate": 1.0600470197759648e-05,
+      "loss": 0.7489,
       "step": 17000
     },
     {
       "epoch": 2.35,
-      "eval_accuracy": 0.7047230481986031,
-      "eval_loss": 1.0511373281478882,
-      "eval_runtime": 155.8456,
-      "eval_samples_per_second": 185.581,
-      "eval_steps_per_second": 23.202,
+      "eval_accuracy": 0.7148537445543185,
+      "eval_loss": 1.014554500579834,
+      "eval_runtime": 154.5182,
+      "eval_samples_per_second": 187.175,
+      "eval_steps_per_second": 23.402,
       "step": 17000
     },
     {
       "epoch": 2.49,
-      "learning_rate": 3.7675286958926836e-05,
-      "loss": 0.599,
+      "learning_rate": 1.0047296362882037e-05,
+      "loss": 0.7392,
       "step": 18000
     },
     {
       "epoch": 2.49,
-      "eval_accuracy": 0.7046884724431229,
-      "eval_loss": 1.0590227842330933,
-      "eval_runtime": 155.3559,
-      "eval_samples_per_second": 186.166,
-      "eval_steps_per_second": 23.276,
+      "eval_accuracy": 0.7146117142659567,
+      "eval_loss": 1.0146572589874268,
+      "eval_runtime": 155.0602,
+      "eval_samples_per_second": 186.521,
+      "eval_steps_per_second": 23.32,
       "step": 18000
     },
     {
       "epoch": 2.63,
-      "learning_rate": 3.56008850781358e-05,
-      "loss": 0.5784,
+      "learning_rate": 9.494675701839304e-06,
+      "loss": 0.7447,
       "step": 19000
     },
     {
       "epoch": 2.63,
-      "eval_accuracy": 0.7059331996404121,
-      "eval_loss": 1.0469636917114258,
-      "eval_runtime": 153.9954,
-      "eval_samples_per_second": 187.811,
-      "eval_steps_per_second": 23.481,
+      "eval_accuracy": 0.715856441463246,
+      "eval_loss": 1.011325716972351,
+      "eval_runtime": 155.8908,
+      "eval_samples_per_second": 185.527,
+      "eval_steps_per_second": 23.196,
       "step": 19000
     },
     {
       "epoch": 2.63,
       "step": 19000,
       "total_flos": 4.713138424140411e+19,
-      "train_loss": 0.8959425466437089,
-      "train_runtime": 9068.08,
-      "train_samples_per_second": 127.578,
-      "train_steps_per_second": 3.987
+      "train_loss": 1.0215468621504935,
+      "train_runtime": 9065.9357,
+      "train_samples_per_second": 127.608,
+      "train_steps_per_second": 3.988
     }
   ],
   "max_steps": 36155,