End of training

README.md

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5346
-- Accuracy: 0.85
+- Loss: 0.5017
+- Accuracy: 0.8667
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 9.03,
+    "eval_accuracy": 0.8666666666666667,
+    "eval_loss": 0.5017222166061401,
+    "eval_runtime": 1.2494,
+    "eval_samples_per_second": 48.024,
+    "eval_steps_per_second": 1.601,
+    "train_loss": 0.20223496982029507,
+    "train_runtime": 419.929,
+    "train_samples_per_second": 23.194,
+    "train_steps_per_second": 0.167
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.03,
+    "eval_accuracy": 0.8666666666666667,
+    "eval_loss": 0.5017222166061401,
+    "eval_runtime": 1.2494,
+    "eval_samples_per_second": 48.024,
+    "eval_steps_per_second": 1.601
+}

runs/Apr25_10-42-39_DESKTOP-SKBE9FB/events.out.tfevents.1714063845.DESKTOP-SKBE9FB.12604.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce5a97ebbcdf1e05bab2f22d8f1cdb00bc8687918a8a84b0d43d3ae7d08d3a6c
+size 405

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 9.03,
+    "train_loss": 0.20223496982029507,
+    "train_runtime": 419.929,
+    "train_samples_per_second": 23.194,
+    "train_steps_per_second": 0.167
+}

trainer_state.json

@@ -0,0 +1,162 @@
+{
+  "best_metric": 0.8666666666666667,
+  "best_model_checkpoint": "vit-base-patch16-224-dmae-va-U5-100bc\\checkpoint-7",
+  "epoch": 9.03225806451613,
+  "eval_steps": 500,
+  "global_step": 70,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.9,
+      "eval_accuracy": 0.8666666666666667,
+      "eval_loss": 0.5017222166061401,
+      "eval_runtime": 0.94,
+      "eval_samples_per_second": 63.829,
+      "eval_steps_per_second": 2.128,
+      "step": 7
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 0.3168,
+      "step": 10
+    },
+    {
+      "epoch": 1.94,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.5970409512519836,
+      "eval_runtime": 0.955,
+      "eval_samples_per_second": 62.829,
+      "eval_steps_per_second": 2.094,
+      "step": 15
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 3.787878787878788e-05,
+      "loss": 0.2613,
+      "step": 20
+    },
+    {
+      "epoch": 2.97,
+      "eval_accuracy": 0.8166666666666667,
+      "eval_loss": 0.5442367792129517,
+      "eval_runtime": 1.0051,
+      "eval_samples_per_second": 59.698,
+      "eval_steps_per_second": 1.99,
+      "step": 23
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 3.0303030303030306e-05,
+      "loss": 0.222,
+      "step": 30
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7666666666666667,
+      "eval_loss": 0.7156000733375549,
+      "eval_runtime": 1.4807,
+      "eval_samples_per_second": 40.521,
+      "eval_steps_per_second": 1.351,
+      "step": 31
+    },
+    {
+      "epoch": 4.9,
+      "eval_accuracy": 0.85,
+      "eval_loss": 0.5175150036811829,
+      "eval_runtime": 0.952,
+      "eval_samples_per_second": 63.028,
+      "eval_steps_per_second": 2.101,
+      "step": 38
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 0.1783,
+      "step": 40
+    },
+    {
+      "epoch": 5.94,
+      "eval_accuracy": 0.8166666666666667,
+      "eval_loss": 0.6034969687461853,
+      "eval_runtime": 0.9995,
+      "eval_samples_per_second": 60.029,
+      "eval_steps_per_second": 2.001,
+      "step": 46
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 1.5151515151515153e-05,
+      "loss": 0.168,
+      "step": 50
+    },
+    {
+      "epoch": 6.97,
+      "eval_accuracy": 0.85,
+      "eval_loss": 0.5045068860054016,
+      "eval_runtime": 0.979,
+      "eval_samples_per_second": 61.287,
+      "eval_steps_per_second": 2.043,
+      "step": 54
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 7.5757575757575764e-06,
+      "loss": 0.1456,
+      "step": 60
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.85,
+      "eval_loss": 0.49230921268463135,
+      "eval_runtime": 0.9805,
+      "eval_samples_per_second": 61.19,
+      "eval_steps_per_second": 2.04,
+      "step": 62
+    },
+    {
+      "epoch": 8.9,
+      "eval_accuracy": 0.85,
+      "eval_loss": 0.5346192121505737,
+      "eval_runtime": 0.9662,
+      "eval_samples_per_second": 62.096,
+      "eval_steps_per_second": 2.07,
+      "step": 69
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.0,
+      "loss": 0.1236,
+      "step": 70
+    },
+    {
+      "epoch": 9.03,
+      "eval_accuracy": 0.85,
+      "eval_loss": 0.5345727801322937,
+      "eval_runtime": 1.0025,
+      "eval_samples_per_second": 59.851,
+      "eval_steps_per_second": 1.995,
+      "step": 70
+    },
+    {
+      "epoch": 9.03,
+      "step": 70,
+      "total_flos": 6.817867456880148e+17,
+      "train_loss": 0.20223496982029507,
+      "train_runtime": 419.929,
+      "train_samples_per_second": 23.194,
+      "train_steps_per_second": 0.167
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 70,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 6.817867456880148e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}