End of training

README.md

@@ -2,6 +2,8 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 metrics:
 - accuracy
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-beans
 
-This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2273
 - Accuracy: 0.9699

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 1.0,
-    "eval_accuracy": 0.8646616541353384,
-    "eval_loss": 0.8840346336364746,
-    "eval_runtime": 4.7663,
-    "eval_samples_per_second": 27.904,
-    "eval_steps_per_second": 0.629,
-    "total_flos": 8.012743541191066e+16,
-    "train_loss": 0.9630123587215648,
-    "train_runtime": 54.0236,
-    "train_samples_per_second": 19.14,
-    "train_steps_per_second": 0.315
+    "epoch": 7.0,
+    "eval_accuracy": 0.9699248120300752,
+    "eval_loss": 0.22725073993206024,
+    "eval_runtime": 4.2156,
+    "eval_samples_per_second": 31.55,
+    "eval_steps_per_second": 0.712,
+    "total_flos": 5.608920478833746e+17,
+    "train_loss": 0.4668655235226415,
+    "train_runtime": 325.4253,
+    "train_samples_per_second": 22.242,
+    "train_steps_per_second": 0.366
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "eval_accuracy": 0.8646616541353384,
-    "eval_loss": 0.8840346336364746,
-    "eval_runtime": 4.7663,
-    "eval_samples_per_second": 27.904,
-    "eval_steps_per_second": 0.629
+    "epoch": 7.0,
+    "eval_accuracy": 0.9699248120300752,
+    "eval_loss": 0.22725073993206024,
+    "eval_runtime": 4.2156,
+    "eval_samples_per_second": 31.55,
+    "eval_steps_per_second": 0.712
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "total_flos": 8.012743541191066e+16,
-    "train_loss": 0.9630123587215648,
-    "train_runtime": 54.0236,
-    "train_samples_per_second": 19.14,
-    "train_steps_per_second": 0.315
+    "epoch": 7.0,
+    "total_flos": 5.608920478833746e+17,
+    "train_loss": 0.4668655235226415,
+    "train_runtime": 325.4253,
+    "train_samples_per_second": 22.242,
+    "train_steps_per_second": 0.366
 }

trainer_state.json

@@ -1,43 +1,167 @@
 {
-  "best_metric": 0.8840346336364746,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-17",
-  "epoch": 1.0,
+  "best_metric": 0.22725073993206024,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-119",
+  "epoch": 7.0,
   "eval_steps": 500,
-  "global_step": 17,
+  "global_step": 119,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.354477882385254,
-      "learning_rate": 8.23529411764706e-06,
-      "loss": 0.9983,
+      "grad_norm": 1.3721389770507812,
+      "learning_rate": 1.831932773109244e-05,
+      "loss": 0.9859,
       "step": 10
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8646616541353384,
-      "eval_loss": 0.8840346336364746,
-      "eval_runtime": 4.3311,
-      "eval_samples_per_second": 30.708,
-      "eval_steps_per_second": 0.693,
+      "eval_accuracy": 0.9323308270676691,
+      "eval_loss": 0.7494019269943237,
+      "eval_runtime": 4.6478,
+      "eval_samples_per_second": 28.616,
+      "eval_steps_per_second": 0.645,
       "step": 17
     },
     {
-      "epoch": 1.0,
-      "step": 17,
-      "total_flos": 8.012743541191066e+16,
-      "train_loss": 0.9630123587215648,
-      "train_runtime": 54.0236,
-      "train_samples_per_second": 19.14,
-      "train_steps_per_second": 0.315
+      "epoch": 1.1764705882352942,
+      "grad_norm": 1.188028335571289,
+      "learning_rate": 1.6638655462184877e-05,
+      "loss": 0.8079,
+      "step": 20
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 1.1527937650680542,
+      "learning_rate": 1.4957983193277313e-05,
+      "loss": 0.6764,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9624060150375939,
+      "eval_loss": 0.5287379026412964,
+      "eval_runtime": 4.6427,
+      "eval_samples_per_second": 28.647,
+      "eval_steps_per_second": 0.646,
+      "step": 34
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 1.2383160591125488,
+      "learning_rate": 1.3277310924369749e-05,
+      "loss": 0.5503,
+      "step": 40
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 1.011650800704956,
+      "learning_rate": 1.1596638655462186e-05,
+      "loss": 0.4614,
+      "step": 50
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9624060150375939,
+      "eval_loss": 0.37249577045440674,
+      "eval_runtime": 4.5756,
+      "eval_samples_per_second": 29.067,
+      "eval_steps_per_second": 0.656,
+      "step": 51
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 1.1885125637054443,
+      "learning_rate": 9.915966386554622e-06,
+      "loss": 0.4053,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.2979891300201416,
+      "eval_runtime": 4.9698,
+      "eval_samples_per_second": 26.762,
+      "eval_steps_per_second": 0.604,
+      "step": 68
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 1.0263020992279053,
+      "learning_rate": 8.23529411764706e-06,
+      "loss": 0.3269,
+      "step": 70
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 1.331243634223938,
+      "learning_rate": 6.5546218487394966e-06,
+      "loss": 0.3172,
+      "step": 80
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.2541128396987915,
+      "eval_runtime": 4.376,
+      "eval_samples_per_second": 30.393,
+      "eval_steps_per_second": 0.686,
+      "step": 85
+    },
+    {
+      "epoch": 5.294117647058823,
+      "grad_norm": 1.0003427267074585,
+      "learning_rate": 4.873949579831933e-06,
+      "loss": 0.2773,
+      "step": 90
+    },
+    {
+      "epoch": 5.882352941176471,
+      "grad_norm": 0.8368270397186279,
+      "learning_rate": 3.1932773109243696e-06,
+      "loss": 0.2542,
+      "step": 100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.22862975299358368,
+      "eval_runtime": 4.1954,
+      "eval_samples_per_second": 31.701,
+      "eval_steps_per_second": 0.715,
+      "step": 102
+    },
+    {
+      "epoch": 6.470588235294118,
+      "grad_norm": 1.1537479162216187,
+      "learning_rate": 1.5126050420168068e-06,
+      "loss": 0.2629,
+      "step": 110
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.22725073993206024,
+      "eval_runtime": 4.8721,
+      "eval_samples_per_second": 27.298,
+      "eval_steps_per_second": 0.616,
+      "step": 119
+    },
+    {
+      "epoch": 7.0,
+      "step": 119,
+      "total_flos": 5.608920478833746e+17,
+      "train_loss": 0.4668655235226415,
+      "train_runtime": 325.4253,
+      "train_samples_per_second": 22.242,
+      "train_steps_per_second": 0.366
     }
   ],
   "logging_steps": 10,
-  "max_steps": 17,
+  "max_steps": 119,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 7,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -51,7 +175,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.012743541191066e+16,
+  "total_flos": 5.608920478833746e+17,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null