🍻 cheers

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 2.63,
-    "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_accuracy": 0.7145425627549962,
+    "eval_loss": 0.9821608662605286,
+    "eval_runtime": 154.2697,
+    "eval_samples_per_second": 187.477,
+    "eval_steps_per_second": 23.439,
+    "train_loss": 1.0496204544870478,
+    "train_runtime": 8957.3844,
+    "train_samples_per_second": 129.154,
+    "train_steps_per_second": 4.036
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.63,
-    "eval_accuracy": 0.7164442293064104,
-    "eval_loss": 0.99397212266922,
-    "eval_runtime": 156.2682,
-    "eval_samples_per_second": 185.079,
-    "eval_steps_per_second": 23.14
+    "eval_accuracy": 0.7145425627549962,
+    "eval_loss": 0.9821608662605286,
+    "eval_runtime": 154.2697,
+    "eval_samples_per_second": 187.477,
+    "eval_steps_per_second": 23.439
 }

runs/May30_08-16-10_adrian-development/events.out.tfevents.1685436542.adrian-development.8035.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b852ff8b5bb19b6752ebc94e3e857ce979b59379a6d0bbb2303077f4ff2c28d9
+size 369

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 2.63,
-    "train_loss": 1.0215468621504935,
-    "train_runtime": 9065.9357,
-    "train_samples_per_second": 127.608,
-    "train_steps_per_second": 3.988
+    "train_loss": 1.0496204544870478,
+    "train_runtime": 8957.3844,
+    "train_samples_per_second": 129.154,
+    "train_steps_per_second": 4.036
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.99397212266922,
+  "best_metric": 0.9821608662605286,
   "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": 1.9447379338957268e-05,
-      "loss": 2.1091,
+      "learning_rate": 1.7016456921587607e-05,
+      "loss": 2.1712,
       "step": 1000
     },
     {
       "epoch": 0.14,
-      "eval_accuracy": 0.6352603554387664,
-      "eval_loss": 1.5707892179489136,
-      "eval_runtime": 154.5668,
-      "eval_samples_per_second": 187.116,
-      "eval_steps_per_second": 23.394,
+      "eval_accuracy": 0.6206348108706176,
+      "eval_loss": 1.6429003477096558,
+      "eval_runtime": 157.1552,
+      "eval_samples_per_second": 184.035,
+      "eval_steps_per_second": 23.009,
       "step": 1000
     },
     {
       "epoch": 0.28,
-      "learning_rate": 1.8894758677914535e-05,
-      "loss": 1.4147,
+      "learning_rate": 1.65324298160697e-05,
+      "loss": 1.4661,
       "step": 2000
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.6584606873660189,
-      "eval_loss": 1.3137956857681274,
-      "eval_runtime": 154.1896,
-      "eval_samples_per_second": 187.574,
-      "eval_steps_per_second": 23.452,
+      "eval_accuracy": 0.6624714750017288,
+      "eval_loss": 1.3386942148208618,
+      "eval_runtime": 148.6316,
+      "eval_samples_per_second": 194.588,
+      "eval_steps_per_second": 24.329,
       "step": 2000
     },
     {
       "epoch": 0.41,
-      "learning_rate": 1.8341584843036927e-05,
-      "loss": 1.2355,
+      "learning_rate": 1.6048886737657307e-05,
+      "loss": 1.2803,
       "step": 3000
     },
     {
       "epoch": 0.41,
-      "eval_accuracy": 0.6820067768480741,
-      "eval_loss": 1.1873301267623901,
-      "eval_runtime": 155.4058,
-      "eval_samples_per_second": 186.106,
-      "eval_steps_per_second": 23.268,
+      "eval_accuracy": 0.6767166862595948,
+      "eval_loss": 1.2176014184951782,
+      "eval_runtime": 150.2003,
+      "eval_samples_per_second": 192.556,
+      "eval_steps_per_second": 24.075,
       "step": 3000
     },
     {
       "epoch": 0.55,
-      "learning_rate": 1.7788411008159317e-05,
-      "loss": 1.1718,
+      "learning_rate": 1.55648596321394e-05,
+      "loss": 1.2063,
       "step": 4000
     },
     {
       "epoch": 0.55,
-      "eval_accuracy": 0.6836664131111265,
-      "eval_loss": 1.1535007953643799,
-      "eval_runtime": 155.1856,
-      "eval_samples_per_second": 186.37,
-      "eval_steps_per_second": 23.301,
+      "eval_accuracy": 0.682663716202199,
+      "eval_loss": 1.1588940620422363,
+      "eval_runtime": 148.1593,
+      "eval_samples_per_second": 195.209,
+      "eval_steps_per_second": 24.406,
       "step": 4000
     },
     {
       "epoch": 0.69,
-      "learning_rate": 1.7235237173281706e-05,
-      "loss": 1.1154,
+      "learning_rate": 1.5080832526621489e-05,
+      "loss": 1.1377,
       "step": 5000
     },
     {
       "epoch": 0.69,
-      "eval_accuracy": 0.6977041698361109,
-      "eval_loss": 1.092354655265808,
-      "eval_runtime": 153.5953,
-      "eval_samples_per_second": 188.3,
-      "eval_steps_per_second": 23.542,
+      "eval_accuracy": 0.6956642002627758,
+      "eval_loss": 1.0993205308914185,
+      "eval_runtime": 148.8741,
+      "eval_samples_per_second": 194.272,
+      "eval_steps_per_second": 24.289,
       "step": 5000
     },
     {
       "epoch": 0.83,
-      "learning_rate": 1.6682063338404095e-05,
-      "loss": 1.0914,
+      "learning_rate": 1.4597773475314616e-05,
+      "loss": 1.1039,
       "step": 6000
     },
     {
       "epoch": 0.83,
-      "eval_accuracy": 0.7002281999861697,
-      "eval_loss": 1.0666176080703735,
-      "eval_runtime": 153.4627,
-      "eval_samples_per_second": 188.463,
-      "eval_steps_per_second": 23.563,
+      "eval_accuracy": 0.7011963211396169,
+      "eval_loss": 1.064678430557251,
+      "eval_runtime": 148.5876,
+      "eval_samples_per_second": 194.646,
+      "eval_steps_per_second": 24.336,
       "step": 6000
     },
     {
       "epoch": 0.97,
-      "learning_rate": 1.6129995851196242e-05,
-      "loss": 1.052,
+      "learning_rate": 1.4113746369796708e-05,
+      "loss": 1.0691,
       "step": 7000
     },
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.7028559574026693,
-      "eval_loss": 1.0515823364257812,
-      "eval_runtime": 154.4506,
-      "eval_samples_per_second": 187.257,
-      "eval_steps_per_second": 23.412,
+      "eval_accuracy": 0.7036857755341954,
+      "eval_loss": 1.0476173162460327,
+      "eval_runtime": 150.2195,
+      "eval_samples_per_second": 192.532,
+      "eval_steps_per_second": 24.071,
       "step": 7000
     },
     {
       "epoch": 1.11,
-      "learning_rate": 1.5576822016318628e-05,
-      "loss": 0.9649,
+      "learning_rate": 1.3629719264278799e-05,
+      "loss": 0.9729,
       "step": 8000
     },
     {
       "epoch": 1.11,
-      "eval_accuracy": 0.7032708664684323,
-      "eval_loss": 1.0426146984100342,
-      "eval_runtime": 154.313,
-      "eval_samples_per_second": 187.424,
-      "eval_steps_per_second": 23.433,
+      "eval_accuracy": 0.7059677753958924,
+      "eval_loss": 1.0341469049453735,
+      "eval_runtime": 148.2309,
+      "eval_samples_per_second": 195.115,
+      "eval_steps_per_second": 24.394,
       "step": 8000
     },
     {
       "epoch": 1.24,
-      "learning_rate": 1.502364818144102e-05,
-      "loss": 0.9281,
+      "learning_rate": 1.3146176185866407e-05,
+      "loss": 0.9621,
       "step": 9000
     },
     {
       "epoch": 1.24,
-      "eval_accuracy": 0.7112578659843718,
-      "eval_loss": 1.027764081954956,
-      "eval_runtime": 154.0041,
-      "eval_samples_per_second": 187.8,
-      "eval_steps_per_second": 23.48,
+      "eval_accuracy": 0.7090450176336353,
+      "eval_loss": 1.0242159366607666,
+      "eval_runtime": 148.7996,
+      "eval_samples_per_second": 194.369,
+      "eval_steps_per_second": 24.301,
       "step": 9000
     },
     {
       "epoch": 1.38,
-      "learning_rate": 1.4471027520398286e-05,
-      "loss": 0.9131,
+      "learning_rate": 1.2662149080348498e-05,
+      "loss": 0.9344,
       "step": 10000
     },
     {
       "epoch": 1.38,
-      "eval_accuracy": 0.7106009266302469,
-      "eval_loss": 1.021926760673523,
-      "eval_runtime": 154.1315,
-      "eval_samples_per_second": 187.645,
-      "eval_steps_per_second": 23.46,
+      "eval_accuracy": 0.7103588963418851,
+      "eval_loss": 1.0159040689468384,
+      "eval_runtime": 149.5547,
+      "eval_samples_per_second": 193.387,
+      "eval_steps_per_second": 24.178,
       "step": 10000
     },
     {
       "epoch": 1.52,
-      "learning_rate": 1.3917853685520677e-05,
-      "loss": 0.9105,
+      "learning_rate": 1.217812197483059e-05,
+      "loss": 0.9364,
       "step": 11000
     },
     {
       "epoch": 1.52,
-      "eval_accuracy": 0.7136090173570292,
-      "eval_loss": 1.0093262195587158,
-      "eval_runtime": 155.6003,
-      "eval_samples_per_second": 185.874,
-      "eval_steps_per_second": 23.239,
+      "eval_accuracy": 0.7083189267685499,
+      "eval_loss": 1.016477346420288,
+      "eval_runtime": 149.6095,
+      "eval_samples_per_second": 193.317,
+      "eval_steps_per_second": 24.17,
       "step": 11000
     },
     {
       "epoch": 1.66,
-      "learning_rate": 1.3365233024477943e-05,
-      "loss": 0.9139,
+      "learning_rate": 1.169409486931268e-05,
+      "loss": 0.9235,
       "step": 12000
     },
     {
       "epoch": 1.66,
-      "eval_accuracy": 0.7157181384413249,
-      "eval_loss": 1.0021320581436157,
-      "eval_runtime": 154.9466,
-      "eval_samples_per_second": 186.658,
-      "eval_steps_per_second": 23.337,
+      "eval_accuracy": 0.708872138856234,
+      "eval_loss": 1.0138823986053467,
+      "eval_runtime": 155.2432,
+      "eval_samples_per_second": 186.301,
+      "eval_steps_per_second": 23.292,
       "step": 12000
     },
     {
       "epoch": 1.8,
-      "learning_rate": 1.2812059189600333e-05,
-      "loss": 0.901,
+      "learning_rate": 1.1210551790900289e-05,
+      "loss": 0.9264,
       "step": 13000
     },
     {
       "epoch": 1.8,
-      "eval_accuracy": 0.7147500172878778,
-      "eval_loss": 1.0019475221633911,
-      "eval_runtime": 156.866,
-      "eval_samples_per_second": 184.374,
-      "eval_steps_per_second": 23.052,
+      "eval_accuracy": 0.7109466841850495,
+      "eval_loss": 1.0029207468032837,
+      "eval_runtime": 154.1084,
+      "eval_samples_per_second": 187.673,
+      "eval_steps_per_second": 23.464,
       "step": 13000
     },
     {
       "epoch": 1.94,
-      "learning_rate": 1.22594385285576e-05,
-      "loss": 0.8916,
+      "learning_rate": 1.0727008712487899e-05,
+      "loss": 0.9161,
       "step": 14000
     },
     {
       "epoch": 1.94,
-      "eval_accuracy": 0.7164442293064104,
-      "eval_loss": 0.99397212266922,
-      "eval_runtime": 155.3636,
-      "eval_samples_per_second": 186.157,
-      "eval_steps_per_second": 23.274,
+      "eval_accuracy": 0.7145425627549962,
+      "eval_loss": 0.9821608662605286,
+      "eval_runtime": 153.7636,
+      "eval_samples_per_second": 188.094,
+      "eval_steps_per_second": 23.517,
       "step": 14000
     },
     {
       "epoch": 2.07,
-      "learning_rate": 1.1706817867514867e-05,
-      "loss": 0.8142,
+      "learning_rate": 1.024298160696999e-05,
+      "loss": 0.8473,
       "step": 15000
     },
     {
       "epoch": 2.07,
-      "eval_accuracy": 0.7175506534817786,
-      "eval_loss": 1.0117239952087402,
-      "eval_runtime": 155.3204,
-      "eval_samples_per_second": 186.209,
-      "eval_steps_per_second": 23.281,
+      "eval_accuracy": 0.7135398658460688,
+      "eval_loss": 1.004128098487854,
+      "eval_runtime": 154.0463,
+      "eval_samples_per_second": 187.749,
+      "eval_steps_per_second": 23.473,
       "step": 15000
     },
     {
       "epoch": 2.21,
-      "learning_rate": 1.1153644032637257e-05,
-      "loss": 0.7494,
+      "learning_rate": 9.75895450145208e-06,
+      "loss": 0.7759,
       "step": 16000
     },
     {
       "epoch": 2.21,
-      "eval_accuracy": 0.7156144111748842,
-      "eval_loss": 1.014885663986206,
-      "eval_runtime": 155.0656,
-      "eval_samples_per_second": 186.515,
-      "eval_steps_per_second": 23.319,
+      "eval_accuracy": 0.7136781688679897,
+      "eval_loss": 1.0011224746704102,
+      "eval_runtime": 153.3797,
+      "eval_samples_per_second": 188.565,
+      "eval_steps_per_second": 23.575,
       "step": 16000
     },
     {
       "epoch": 2.35,
-      "learning_rate": 1.0600470197759648e-05,
-      "loss": 0.7489,
+      "learning_rate": 9.274927395934171e-06,
+      "loss": 0.7743,
       "step": 17000
     },
     {
       "epoch": 2.35,
-      "eval_accuracy": 0.7148537445543185,
-      "eval_loss": 1.014554500579834,
-      "eval_runtime": 154.5182,
-      "eval_samples_per_second": 187.175,
-      "eval_steps_per_second": 23.402,
+      "eval_accuracy": 0.7119148053384966,
+      "eval_loss": 1.002486228942871,
+      "eval_runtime": 153.5134,
+      "eval_samples_per_second": 188.4,
+      "eval_steps_per_second": 23.555,
       "step": 17000
     },
     {
       "epoch": 2.49,
-      "learning_rate": 1.0047296362882037e-05,
-      "loss": 0.7392,
+      "learning_rate": 8.790900290416263e-06,
+      "loss": 0.7699,
       "step": 18000
     },
     {
       "epoch": 2.49,
-      "eval_accuracy": 0.7146117142659567,
-      "eval_loss": 1.0146572589874268,
-      "eval_runtime": 155.0602,
-      "eval_samples_per_second": 186.521,
-      "eval_steps_per_second": 23.32,
+      "eval_accuracy": 0.7147500172878778,
+      "eval_loss": 0.9976186156272888,
+      "eval_runtime": 155.4031,
+      "eval_samples_per_second": 186.11,
+      "eval_steps_per_second": 23.269,
       "step": 18000
     },
     {
       "epoch": 2.63,
-      "learning_rate": 9.494675701839304e-06,
-      "loss": 0.7447,
+      "learning_rate": 8.307357212003871e-06,
+      "loss": 0.7691,
       "step": 19000
     },
     {
       "epoch": 2.63,
-      "eval_accuracy": 0.715856441463246,
-      "eval_loss": 1.011325716972351,
-      "eval_runtime": 155.8908,
-      "eval_samples_per_second": 185.527,
-      "eval_steps_per_second": 23.196,
+      "eval_accuracy": 0.7145771385104764,
+      "eval_loss": 1.0039151906967163,
+      "eval_runtime": 154.2535,
+      "eval_samples_per_second": 187.497,
+      "eval_steps_per_second": 23.442,
       "step": 19000
     },
     {
       "epoch": 2.63,
       "step": 19000,
       "total_flos": 4.713138424140411e+19,
-      "train_loss": 1.0215468621504935,
-      "train_runtime": 9065.9357,
-      "train_samples_per_second": 127.608,
-      "train_steps_per_second": 3.988
+      "train_loss": 1.0496204544870478,
+      "train_runtime": 8957.3844,
+      "train_samples_per_second": 129.154,
+      "train_steps_per_second": 4.036
     }
   ],
   "max_steps": 36155,