🍻 cheers

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 2.63,
-    "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
+    "epoch": 0.97,
+    "eval_accuracy": 0.24808104557084573,
+    "eval_loss": 2.6692957878112793,
+    "eval_runtime": 153.5066,
+    "eval_samples_per_second": 188.409,
+    "eval_steps_per_second": 23.556,
+    "train_loss": 1.2184505571637836,
+    "train_runtime": 3232.5785,
+    "train_samples_per_second": 357.883,
+    "train_steps_per_second": 11.185
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.63,
-    "eval_accuracy": 0.7145425627549962,
-    "eval_loss": 0.9821608662605286,
-    "eval_runtime": 154.2697,
-    "eval_samples_per_second": 187.477,
-    "eval_steps_per_second": 23.439
+    "epoch": 0.97,
+    "eval_accuracy": 0.24808104557084573,
+    "eval_loss": 2.6692957878112793,
+    "eval_runtime": 153.5066,
+    "eval_samples_per_second": 188.409,
+    "eval_steps_per_second": 23.556
 }

runs/May30_10-58-11_adrian-development/events.out.tfevents.1685440502.adrian-development.46083.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24860b6daae6c9703fc3eff95767ab0b504bc688fbb3c91b73d9141c835d3f51
+size 363

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 2.63,
-    "train_loss": 1.0496204544870478,
-    "train_runtime": 8957.3844,
-    "train_samples_per_second": 129.154,
-    "train_steps_per_second": 4.036
+    "epoch": 0.97,
+    "train_loss": 1.2184505571637836,
+    "train_runtime": 3232.5785,
+    "train_samples_per_second": 357.883,
+    "train_steps_per_second": 11.185
 }

trainer_state.json

@@ -1,310 +1,130 @@
 {
-  "best_metric": 0.9821608662605286,
-  "best_model_checkpoint": "Transformers/vit-base-clothing-leafs-example-full-simple/checkpoint-14000",
-  "epoch": 2.627575715668649,
-  "global_step": 19000,
+  "best_metric": 2.6692957878112793,
+  "best_model_checkpoint": "Transformers/vit-base-clothing-leafs-example-full-simple/checkpoint-2000",
+  "epoch": 0.968054211035818,
+  "global_step": 7000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.14,
-      "learning_rate": 1.7016456921587607e-05,
-      "loss": 2.1712,
+      "learning_rate": 0.0019446826165122391,
+      "loss": 2.7222,
       "step": 1000
     },
     {
       "epoch": 0.14,
-      "eval_accuracy": 0.6206348108706176,
-      "eval_loss": 1.6429003477096558,
-      "eval_runtime": 157.1552,
-      "eval_samples_per_second": 184.035,
-      "eval_steps_per_second": 23.009,
+      "eval_accuracy": 0.25281792407164094,
+      "eval_loss": 2.676572322845459,
+      "eval_runtime": 156.0521,
+      "eval_samples_per_second": 185.336,
+      "eval_steps_per_second": 23.172,
       "step": 1000
     },
     {
       "epoch": 0.28,
-      "learning_rate": 1.65324298160697e-05,
-      "loss": 1.4661,
+      "learning_rate": 0.001889586502558429,
+      "loss": 2.6768,
       "step": 2000
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.6624714750017288,
-      "eval_loss": 1.3386942148208618,
-      "eval_runtime": 148.6316,
-      "eval_samples_per_second": 194.588,
-      "eval_steps_per_second": 24.329,
+      "eval_accuracy": 0.24808104557084573,
+      "eval_loss": 2.6692957878112793,
+      "eval_runtime": 154.1979,
+      "eval_samples_per_second": 187.564,
+      "eval_steps_per_second": 23.45,
       "step": 2000
     },
     {
       "epoch": 0.41,
-      "learning_rate": 1.6048886737657307e-05,
-      "loss": 1.2803,
+      "learning_rate": 0.0018360392753422764,
+      "loss": 2.6689,
       "step": 3000
     },
     {
       "epoch": 0.41,
-      "eval_accuracy": 0.6767166862595948,
-      "eval_loss": 1.2176014184951782,
-      "eval_runtime": 150.2003,
-      "eval_samples_per_second": 192.556,
-      "eval_steps_per_second": 24.075,
+      "eval_accuracy": 0.22304819860313949,
+      "eval_loss": NaN,
+      "eval_runtime": 151.9095,
+      "eval_samples_per_second": 190.39,
+      "eval_steps_per_second": 23.804,
       "step": 3000
     },
     {
       "epoch": 0.55,
-      "learning_rate": 1.55648596321394e-05,
-      "loss": 1.2063,
+      "learning_rate": 0.0017879131517079244,
+      "loss": 0.4613,
       "step": 4000
     },
     {
       "epoch": 0.55,
-      "eval_accuracy": 0.682663716202199,
-      "eval_loss": 1.1588940620422363,
-      "eval_runtime": 148.1593,
-      "eval_samples_per_second": 195.209,
-      "eval_steps_per_second": 24.406,
+      "eval_accuracy": 0.051241269621741234,
+      "eval_loss": NaN,
+      "eval_runtime": 152.778,
+      "eval_samples_per_second": 189.307,
+      "eval_steps_per_second": 23.668,
       "step": 4000
     },
     {
       "epoch": 0.69,
-      "learning_rate": 1.5080832526621489e-05,
-      "loss": 1.1377,
+      "learning_rate": 0.0017325957682201632,
+      "loss": 0.0,
       "step": 5000
     },
     {
       "epoch": 0.69,
-      "eval_accuracy": 0.6956642002627758,
-      "eval_loss": 1.0993205308914185,
-      "eval_runtime": 148.8741,
-      "eval_samples_per_second": 194.272,
-      "eval_steps_per_second": 24.289,
+      "eval_accuracy": 0.051241269621741234,
+      "eval_loss": NaN,
+      "eval_runtime": 152.584,
+      "eval_samples_per_second": 189.548,
+      "eval_steps_per_second": 23.698,
       "step": 5000
     },
     {
       "epoch": 0.83,
-      "learning_rate": 1.4597773475314616e-05,
-      "loss": 1.1039,
+      "learning_rate": 0.0016772783847324023,
+      "loss": 0.0,
       "step": 6000
     },
     {
       "epoch": 0.83,
-      "eval_accuracy": 0.7011963211396169,
-      "eval_loss": 1.064678430557251,
-      "eval_runtime": 148.5876,
-      "eval_samples_per_second": 194.646,
-      "eval_steps_per_second": 24.336,
+      "eval_accuracy": 0.051241269621741234,
+      "eval_loss": NaN,
+      "eval_runtime": 152.8491,
+      "eval_samples_per_second": 189.219,
+      "eval_steps_per_second": 23.657,
       "step": 6000
     },
     {
       "epoch": 0.97,
-      "learning_rate": 1.4113746369796708e-05,
-      "loss": 1.0691,
+      "learning_rate": 0.001621961001244641,
+      "loss": 0.0,
       "step": 7000
     },
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.7036857755341954,
-      "eval_loss": 1.0476173162460327,
-      "eval_runtime": 150.2195,
-      "eval_samples_per_second": 192.532,
-      "eval_steps_per_second": 24.071,
+      "eval_accuracy": 0.051241269621741234,
+      "eval_loss": NaN,
+      "eval_runtime": 151.1266,
+      "eval_samples_per_second": 191.376,
+      "eval_steps_per_second": 23.927,
       "step": 7000
     },
     {
-      "epoch": 1.11,
-      "learning_rate": 1.3629719264278799e-05,
-      "loss": 0.9729,
-      "step": 8000
-    },
-    {
-      "epoch": 1.11,
-      "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": 1.3146176185866407e-05,
-      "loss": 0.9621,
-      "step": 9000
-    },
-    {
-      "epoch": 1.24,
-      "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": 1.2662149080348498e-05,
-      "loss": 0.9344,
-      "step": 10000
-    },
-    {
-      "epoch": 1.38,
-      "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": 1.217812197483059e-05,
-      "loss": 0.9364,
-      "step": 11000
-    },
-    {
-      "epoch": 1.52,
-      "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": 1.169409486931268e-05,
-      "loss": 0.9235,
-      "step": 12000
-    },
-    {
-      "epoch": 1.66,
-      "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": 1.1210551790900289e-05,
-      "loss": 0.9264,
-      "step": 13000
-    },
-    {
-      "epoch": 1.8,
-      "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": 1.0727008712487899e-05,
-      "loss": 0.9161,
-      "step": 14000
-    },
-    {
-      "epoch": 1.94,
-      "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": 1.024298160696999e-05,
-      "loss": 0.8473,
-      "step": 15000
-    },
-    {
-      "epoch": 2.07,
-      "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": 9.75895450145208e-06,
-      "loss": 0.7759,
-      "step": 16000
-    },
-    {
-      "epoch": 2.21,
-      "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": 9.274927395934171e-06,
-      "loss": 0.7743,
-      "step": 17000
-    },
-    {
-      "epoch": 2.35,
-      "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": 8.790900290416263e-06,
-      "loss": 0.7699,
-      "step": 18000
-    },
-    {
-      "epoch": 2.49,
-      "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": 8.307357212003871e-06,
-      "loss": 0.7691,
-      "step": 19000
-    },
-    {
-      "epoch": 2.63,
-      "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": 1.0496204544870478,
-      "train_runtime": 8957.3844,
-      "train_samples_per_second": 129.154,
-      "train_steps_per_second": 4.036
+      "epoch": 0.97,
+      "step": 7000,
+      "total_flos": 1.736505102237696e+19,
+      "train_loss": 1.2184505571637836,
+      "train_runtime": 3232.5785,
+      "train_samples_per_second": 357.883,
+      "train_steps_per_second": 11.185
     }
   ],
   "max_steps": 36155,
   "num_train_epochs": 5,
-  "total_flos": 4.713138424140411e+19,
+  "total_flos": 1.736505102237696e+19,
   "trial_name": null,
   "trial_params": null
 }