🍻 cheers

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 0.9924812030075187,
+    "eval_loss": 0.02442088909447193,
+    "eval_runtime": 2.5586,
+    "eval_samples_per_second": 51.982,
+    "eval_steps_per_second": 6.644,
+    "total_flos": 3.205097416476426e+17,
+    "train_loss": 0.11757667609132254,
+    "train_runtime": 133.442,
+    "train_samples_per_second": 30.995,
+    "train_steps_per_second": 1.948
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 0.9924812030075187,
+    "eval_loss": 0.02442088909447193,
+    "eval_runtime": 2.5586,
+    "eval_samples_per_second": 51.982,
+    "eval_steps_per_second": 6.644
+}

runs/Nov29_15-48-35_6c1c4745addc/events.out.tfevents.1669737769.6c1c4745addc.76.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d7fb5bfd0667f2275aff5795744bd0e093035f8a41618f32b3179fe6f321b3e
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "total_flos": 3.205097416476426e+17,
+    "train_loss": 0.11757667609132254,
+    "train_runtime": 133.442,
+    "train_samples_per_second": 30.995,
+    "train_steps_per_second": 1.948
+}

trainer_state.json

@@ -0,0 +1,199 @@
+{
+  "best_metric": 0.02442088909447193,
+  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
+  "epoch": 4.0,
+  "global_step": 260,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00019230769230769233,
+      "loss": 0.8847,
+      "step": 10
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00018461538461538463,
+      "loss": 0.4232,
+      "step": 20
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00017692307692307693,
+      "loss": 0.3126,
+      "step": 30
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00016923076923076923,
+      "loss": 0.2217,
+      "step": 40
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0001623076923076923,
+      "loss": 0.222,
+      "step": 50
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00015461538461538464,
+      "loss": 0.1696,
+      "step": 60
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014692307692307693,
+      "loss": 0.093,
+      "step": 70
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00013923076923076923,
+      "loss": 0.1037,
+      "step": 80
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00013153846153846156,
+      "loss": 0.057,
+      "step": 90
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00012384615384615385,
+      "loss": 0.069,
+      "step": 100
+    },
+    {
+      "epoch": 1.54,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.08042918890714645,
+      "eval_runtime": 1.7912,
+      "eval_samples_per_second": 74.252,
+      "eval_steps_per_second": 9.491,
+      "step": 100
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00011615384615384617,
+      "loss": 0.0754,
+      "step": 110
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00010846153846153846,
+      "loss": 0.0962,
+      "step": 120
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00010076923076923077,
+      "loss": 0.0324,
+      "step": 130
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 9.307692307692309e-05,
+      "loss": 0.0399,
+      "step": 140
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 8.538461538461538e-05,
+      "loss": 0.0242,
+      "step": 150
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 7.76923076923077e-05,
+      "loss": 0.0845,
+      "step": 160
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 7e-05,
+      "loss": 0.0169,
+      "step": 170
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 6.23076923076923e-05,
+      "loss": 0.0171,
+      "step": 180
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 5.461538461538461e-05,
+      "loss": 0.0125,
+      "step": 190
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 4.692307692307693e-05,
+      "loss": 0.0241,
+      "step": 200
+    },
+    {
+      "epoch": 3.08,
+      "eval_accuracy": 0.9924812030075187,
+      "eval_loss": 0.02442088909447193,
+      "eval_runtime": 2.2612,
+      "eval_samples_per_second": 58.818,
+      "eval_steps_per_second": 7.518,
+      "step": 200
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 3.923076923076923e-05,
+      "loss": 0.0192,
+      "step": 210
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.153846153846154e-05,
+      "loss": 0.011,
+      "step": 220
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 2.384615384615385e-05,
+      "loss": 0.0106,
+      "step": 230
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 1.6153846153846154e-05,
+      "loss": 0.0117,
+      "step": 240
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 8.461538461538462e-06,
+      "loss": 0.0106,
+      "step": 250
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 7.692307692307694e-07,
+      "loss": 0.0144,
+      "step": 260
+    },
+    {
+      "epoch": 4.0,
+      "step": 260,
+      "total_flos": 3.205097416476426e+17,
+      "train_loss": 0.11757667609132254,
+      "train_runtime": 133.442,
+      "train_samples_per_second": 30.995,
+      "train_steps_per_second": 1.948
+    }
+  ],
+  "max_steps": 260,
+  "num_train_epochs": 4,
+  "total_flos": 3.205097416476426e+17,
+  "trial_name": null,
+  "trial_params": null
+}