🍻 cheers

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 2.63,
-    "eval_accuracy": 0.7190028352119494,
-    "eval_loss": 0.9720757007598877,
-    "eval_runtime": 154.7452,
-    "eval_samples_per_second": 186.901,
-    "eval_steps_per_second": 23.367,
-    "train_loss": 0.9828558799342105,
-    "train_runtime": 8888.1365,
-    "train_samples_per_second": 130.161,
-    "train_steps_per_second": 4.068
+    "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_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.63,
-    "eval_accuracy": 0.7190028352119494,
-    "eval_loss": 0.9720757007598877,
-    "eval_runtime": 154.7452,
-    "eval_samples_per_second": 186.901,
-    "eval_steps_per_second": 23.367
+    "eval_accuracy": 0.7109466841850495,
+    "eval_loss": 0.9905902743339539,
+    "eval_runtime": 153.5849,
+    "eval_samples_per_second": 188.313,
+    "eval_steps_per_second": 23.544
 }

runs/May29_18-47-40_adrian-development/events.out.tfevents.1685388145.adrian-development.53969.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6b0e2ffbbe990159b9d0f2379bc1be31556ad7be8f44aa89a0c4c1a202e7b6d
+size 369

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 2.63,
-    "train_loss": 0.9828558799342105,
-    "train_runtime": 8888.1365,
-    "train_samples_per_second": 130.161,
-    "train_steps_per_second": 4.068
+    "train_loss": 0.8959425466437089,
+    "train_runtime": 9068.08,
+    "train_samples_per_second": 127.578,
+    "train_steps_per_second": 3.987
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9720757007598877,
+  "best_metric": 0.9905902743339539,
   "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": 2.4309224173696584e-05,
-      "loss": 2.0104,
+      "learning_rate": 7.292559811920896e-05,
+      "loss": 1.6049,
       "step": 1000
     },
     {
       "epoch": 0.14,
-      "eval_accuracy": 0.6474655971232971,
-      "eval_loss": 1.4711166620254517,
-      "eval_runtime": 155.8932,
-      "eval_samples_per_second": 185.524,
-      "eval_steps_per_second": 23.195,
+      "eval_accuracy": 0.6462900214369683,
+      "eval_loss": 1.2689716815948486,
+      "eval_runtime": 157.1043,
+      "eval_samples_per_second": 184.094,
+      "eval_steps_per_second": 23.017,
       "step": 1000
     },
     {
       "epoch": 0.28,
-      "learning_rate": 2.3617756880099574e-05,
-      "loss": 1.3423,
+      "learning_rate": 7.085327064029871e-05,
+      "loss": 1.1874,
       "step": 2000
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.6770624438143974,
-      "eval_loss": 1.237913966178894,
-      "eval_runtime": 149.8617,
-      "eval_samples_per_second": 192.991,
-      "eval_steps_per_second": 24.129,
+      "eval_accuracy": 0.6708042320724708,
+      "eval_loss": 1.1465643644332886,
+      "eval_runtime": 154.03,
+      "eval_samples_per_second": 187.769,
+      "eval_steps_per_second": 23.476,
       "step": 2000
     },
     {
       "epoch": 0.41,
-      "learning_rate": 2.2926289586502558e-05,
-      "loss": 1.1916,
+      "learning_rate": 6.878094316138846e-05,
+      "loss": 1.1217,
       "step": 3000
     },
     {
       "epoch": 0.41,
-      "eval_accuracy": 0.6885761703893231,
-      "eval_loss": 1.1475087404251099,
-      "eval_runtime": 151.1105,
-      "eval_samples_per_second": 191.396,
-      "eval_steps_per_second": 23.93,
+      "eval_accuracy": 0.6792061406541733,
+      "eval_loss": 1.1049177646636963,
+      "eval_runtime": 154.7974,
+      "eval_samples_per_second": 186.838,
+      "eval_steps_per_second": 23.36,
       "step": 3000
     },
     {
       "epoch": 0.55,
-      "learning_rate": 2.2235513760199144e-05,
-      "loss": 1.1322,
+      "learning_rate": 6.670654128059742e-05,
+      "loss": 1.0743,
       "step": 4000
     },
     {
       "epoch": 0.55,
-      "eval_accuracy": 0.6945923518428878,
-      "eval_loss": 1.0932817459106445,
-      "eval_runtime": 151.1375,
-      "eval_samples_per_second": 191.362,
-      "eval_steps_per_second": 23.925,
+      "eval_accuracy": 0.6763017771938317,
+      "eval_loss": 1.091710090637207,
+      "eval_runtime": 154.343,
+      "eval_samples_per_second": 187.388,
+      "eval_steps_per_second": 23.428,
       "step": 4000
     },
     {
       "epoch": 0.69,
-      "learning_rate": 2.1544737933895727e-05,
-      "loss": 1.1078,
+      "learning_rate": 6.463213939980638e-05,
+      "loss": 1.0546,
       "step": 5000
     },
     {
       "epoch": 0.69,
-      "eval_accuracy": 0.7007814120738538,
-      "eval_loss": 1.0583287477493286,
-      "eval_runtime": 152.0794,
-      "eval_samples_per_second": 190.177,
-      "eval_steps_per_second": 23.777,
+      "eval_accuracy": 0.6879883825461587,
+      "eval_loss": 1.0747572183609009,
+      "eval_runtime": 156.4913,
+      "eval_samples_per_second": 184.815,
+      "eval_steps_per_second": 23.107,
       "step": 5000
     },
     {
       "epoch": 0.83,
-      "learning_rate": 2.0853270640298714e-05,
-      "loss": 1.0718,
+      "learning_rate": 6.255981192089614e-05,
+      "loss": 1.0375,
       "step": 6000
     },
     {
       "epoch": 0.83,
-      "eval_accuracy": 0.7058640481294516,
-      "eval_loss": 1.0382111072540283,
-      "eval_runtime": 149.9231,
-      "eval_samples_per_second": 192.912,
-      "eval_steps_per_second": 24.119,
+      "eval_accuracy": 0.6867436553488694,
+      "eval_loss": 1.0563393831253052,
+      "eval_runtime": 154.5847,
+      "eval_samples_per_second": 187.095,
+      "eval_steps_per_second": 23.392,
       "step": 6000
     },
     {
       "epoch": 0.97,
-      "learning_rate": 2.01618033467017e-05,
-      "loss": 1.0419,
+      "learning_rate": 6.0485410040105095e-05,
+      "loss": 1.0244,
       "step": 7000
     },
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.7117419265610954,
-      "eval_loss": 1.0121501684188843,
-      "eval_runtime": 155.2505,
-      "eval_samples_per_second": 186.293,
-      "eval_steps_per_second": 23.291,
+      "eval_accuracy": 0.6980153516354333,
+      "eval_loss": 1.034531593322754,
+      "eval_runtime": 155.8798,
+      "eval_samples_per_second": 185.54,
+      "eval_steps_per_second": 23.197,
       "step": 7000
     },
     {
       "epoch": 1.11,
-      "learning_rate": 1.947033605310469e-05,
-      "loss": 0.9331,
+      "learning_rate": 5.8411008159314065e-05,
+      "loss": 0.8943,
       "step": 8000
     },
     {
       "epoch": 1.11,
-      "eval_accuracy": 0.7092178964110366,
-      "eval_loss": 1.014099359512329,
-      "eval_runtime": 154.5217,
-      "eval_samples_per_second": 187.171,
-      "eval_steps_per_second": 23.401,
+      "eval_accuracy": 0.7005739575409723,
+      "eval_loss": 1.0278058052062988,
+      "eval_runtime": 156.5423,
+      "eval_samples_per_second": 184.755,
+      "eval_steps_per_second": 23.099,
       "step": 8000
     },
     {
       "epoch": 1.24,
-      "learning_rate": 1.878025169409487e-05,
-      "loss": 0.8889,
+      "learning_rate": 5.633660627852302e-05,
+      "loss": 0.8324,
       "step": 9000
     },
     {
       "epoch": 1.24,
-      "eval_accuracy": 0.7111887144734113,
-      "eval_loss": 1.015520691871643,
-      "eval_runtime": 154.7731,
-      "eval_samples_per_second": 186.867,
-      "eval_steps_per_second": 23.363,
+      "eval_accuracy": 0.7050342299979254,
+      "eval_loss": 1.0215510129928589,
+      "eval_runtime": 154.4075,
+      "eval_samples_per_second": 187.31,
+      "eval_steps_per_second": 23.419,
       "step": 9000
     },
     {
       "epoch": 1.38,
-      "learning_rate": 1.8088784400497857e-05,
-      "loss": 0.8884,
+      "learning_rate": 5.426427879961277e-05,
+      "loss": 0.8411,
       "step": 10000
     },
     {
       "epoch": 1.38,
-      "eval_accuracy": 0.7148191687988382,
-      "eval_loss": 0.9903624653816223,
-      "eval_runtime": 154.1179,
-      "eval_samples_per_second": 187.662,
-      "eval_steps_per_second": 23.463,
+      "eval_accuracy": 0.7016803817163405,
+      "eval_loss": 1.0213595628738403,
+      "eval_runtime": 154.6853,
+      "eval_samples_per_second": 186.973,
+      "eval_steps_per_second": 23.376,
       "step": 10000
     },
     {
       "epoch": 1.52,
-      "learning_rate": 1.739800857419444e-05,
-      "loss": 0.8761,
+      "learning_rate": 5.2189876918821735e-05,
+      "loss": 0.8353,
       "step": 11000
     },
     {
       "epoch": 1.52,
-      "eval_accuracy": 0.7140239264227923,
-      "eval_loss": 0.9896508455276489,
-      "eval_runtime": 153.2726,
-      "eval_samples_per_second": 188.697,
-      "eval_steps_per_second": 23.592,
+      "eval_accuracy": 0.704446442154761,
+      "eval_loss": 1.0081199407577515,
+      "eval_runtime": 156.0333,
+      "eval_samples_per_second": 185.358,
+      "eval_steps_per_second": 23.175,
       "step": 11000
     },
     {
       "epoch": 1.66,
-      "learning_rate": 1.670654128059743e-05,
-      "loss": 0.8735,
+      "learning_rate": 5.01154750380307e-05,
+      "loss": 0.8307,
       "step": 12000
     },
     {
       "epoch": 1.66,
-      "eval_accuracy": 0.7168591383721734,
-      "eval_loss": 0.9818286895751953,
-      "eval_runtime": 152.8255,
-      "eval_samples_per_second": 189.249,
-      "eval_steps_per_second": 23.661,
+      "eval_accuracy": 0.7079040177027868,
+      "eval_loss": 1.0041202306747437,
+      "eval_runtime": 155.1738,
+      "eval_samples_per_second": 186.384,
+      "eval_steps_per_second": 23.303,
       "step": 12000
     },
     {
       "epoch": 1.8,
-      "learning_rate": 1.6015073987000413e-05,
-      "loss": 0.8815,
+      "learning_rate": 4.8043147559120454e-05,
+      "loss": 0.8297,
       "step": 13000
     },
     {
       "epoch": 1.8,
-      "eval_accuracy": 0.7158910172187263,
-      "eval_loss": 0.9801143407821655,
-      "eval_runtime": 142.9942,
-      "eval_samples_per_second": 202.26,
-      "eval_steps_per_second": 25.288,
+      "eval_accuracy": 0.7103588963418851,
+      "eval_loss": 0.9978411793708801,
+      "eval_runtime": 150.0364,
+      "eval_samples_per_second": 192.767,
+      "eval_steps_per_second": 24.101,
       "step": 13000
     },
     {
       "epoch": 1.94,
-      "learning_rate": 1.5323606693403403e-05,
-      "loss": 0.8698,
+      "learning_rate": 4.596874567832941e-05,
+      "loss": 0.8095,
       "step": 14000
     },
     {
       "epoch": 1.94,
-      "eval_accuracy": 0.7190028352119494,
-      "eval_loss": 0.9720757007598877,
-      "eval_runtime": 143.2081,
-      "eval_samples_per_second": 201.958,
-      "eval_steps_per_second": 25.25,
+      "eval_accuracy": 0.7109466841850495,
+      "eval_loss": 0.9905902743339539,
+      "eval_runtime": 153.3635,
+      "eval_samples_per_second": 188.585,
+      "eval_steps_per_second": 23.578,
       "step": 14000
     },
     {
       "epoch": 2.07,
-      "learning_rate": 1.4632139399806392e-05,
-      "loss": 0.786,
+      "learning_rate": 4.3894343797538375e-05,
+      "loss": 0.7007,
       "step": 15000
     },
     {
       "epoch": 2.07,
-      "eval_accuracy": 0.7172740474379365,
-      "eval_loss": 0.9991583824157715,
-      "eval_runtime": 142.9764,
-      "eval_samples_per_second": 202.285,
-      "eval_steps_per_second": 25.291,
+      "eval_accuracy": 0.7077657146808658,
+      "eval_loss": 1.0342744588851929,
+      "eval_runtime": 155.1599,
+      "eval_samples_per_second": 186.401,
+      "eval_steps_per_second": 23.305,
       "step": 15000
     },
     {
       "epoch": 2.21,
-      "learning_rate": 1.3940672106209377e-05,
-      "loss": 0.693,
+      "learning_rate": 4.1822016318628124e-05,
+      "loss": 0.5791,
       "step": 16000
     },
     {
       "epoch": 2.21,
-      "eval_accuracy": 0.7165133808173708,
-      "eval_loss": 0.9988262057304382,
-      "eval_runtime": 144.3789,
-      "eval_samples_per_second": 200.32,
-      "eval_steps_per_second": 25.045,
+      "eval_accuracy": 0.701472927183459,
+      "eval_loss": 1.0633156299591064,
+      "eval_runtime": 153.3675,
+      "eval_samples_per_second": 188.58,
+      "eval_steps_per_second": 23.577,
       "step": 16000
     },
     {
       "epoch": 2.35,
-      "learning_rate": 1.3249204812612365e-05,
-      "loss": 0.7043,
+      "learning_rate": 3.974761443783709e-05,
+      "loss": 0.5879,
       "step": 17000
     },
     {
       "epoch": 2.35,
-      "eval_accuracy": 0.7158910172187263,
-      "eval_loss": 0.9988804459571838,
-      "eval_runtime": 144.0582,
-      "eval_samples_per_second": 200.766,
-      "eval_steps_per_second": 25.101,
+      "eval_accuracy": 0.7047230481986031,
+      "eval_loss": 1.0511373281478882,
+      "eval_runtime": 155.8456,
+      "eval_samples_per_second": 185.581,
+      "eval_steps_per_second": 23.202,
       "step": 17000
     },
     {
       "epoch": 2.49,
-      "learning_rate": 1.255773751901535e-05,
-      "loss": 0.6995,
+      "learning_rate": 3.7675286958926836e-05,
+      "loss": 0.599,
       "step": 18000
     },
     {
       "epoch": 2.49,
-      "eval_accuracy": 0.7175852292372589,
-      "eval_loss": 0.9933487176895142,
-      "eval_runtime": 154.478,
-      "eval_samples_per_second": 187.224,
-      "eval_steps_per_second": 23.408,
+      "eval_accuracy": 0.7046884724431229,
+      "eval_loss": 1.0590227842330933,
+      "eval_runtime": 155.3559,
+      "eval_samples_per_second": 186.166,
+      "eval_steps_per_second": 23.276,
       "step": 18000
     },
     {
       "epoch": 2.63,
-      "learning_rate": 1.1866961692711935e-05,
-      "loss": 0.6821,
+      "learning_rate": 3.56008850781358e-05,
+      "loss": 0.5784,
       "step": 19000
     },
     {
       "epoch": 2.63,
-      "eval_accuracy": 0.718415047368785,
-      "eval_loss": 1.001592993736267,
-      "eval_runtime": 155.8353,
-      "eval_samples_per_second": 185.593,
-      "eval_steps_per_second": 23.204,
+      "eval_accuracy": 0.7059331996404121,
+      "eval_loss": 1.0469636917114258,
+      "eval_runtime": 153.9954,
+      "eval_samples_per_second": 187.811,
+      "eval_steps_per_second": 23.481,
       "step": 19000
     },
     {
       "epoch": 2.63,
       "step": 19000,
       "total_flos": 4.713138424140411e+19,
-      "train_loss": 0.9828558799342105,
-      "train_runtime": 8888.1365,
-      "train_samples_per_second": 130.161,
-      "train_steps_per_second": 4.068
+      "train_loss": 0.8959425466437089,
+      "train_runtime": 9068.08,
+      "train_samples_per_second": 127.578,
+      "train_steps_per_second": 3.987
     }
   ],
   "max_steps": 36155,