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: 0.8841
 - Accuracy: 0.8647

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 5.0,
-    "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.3437380790710449,
-    "eval_runtime": 5.7697,
-    "eval_samples_per_second": 23.052,
-    "eval_steps_per_second": 0.52,
-    "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.0,
-    "train_runtime": 0.1392,
-    "train_samples_per_second": 37137.186,
-    "train_steps_per_second": 610.573
+    "epoch": 1.0,
+    "eval_accuracy": 0.8646616541353384,
+    "eval_loss": 0.8840949535369873,
+    "eval_runtime": 13.904,
+    "eval_samples_per_second": 9.566,
+    "eval_steps_per_second": 0.216,
+    "total_flos": 8.012743541191066e+16,
+    "train_loss": 0.9630403238184312,
+    "train_runtime": 131.2191,
+    "train_samples_per_second": 7.88,
+    "train_steps_per_second": 0.13
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 5.0,
-    "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.3437380790710449,
-    "eval_runtime": 5.7697,
-    "eval_samples_per_second": 23.052,
-    "eval_steps_per_second": 0.52
+    "epoch": 1.0,
+    "eval_accuracy": 0.8646616541353384,
+    "eval_loss": 0.8840949535369873,
+    "eval_runtime": 13.904,
+    "eval_samples_per_second": 9.566,
+    "eval_steps_per_second": 0.216
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 5.0,
-    "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.0,
-    "train_runtime": 0.1392,
-    "train_samples_per_second": 37137.186,
-    "train_steps_per_second": 610.573
+    "epoch": 1.0,
+    "total_flos": 8.012743541191066e+16,
+    "train_loss": 0.9630403238184312,
+    "train_runtime": 131.2191,
+    "train_samples_per_second": 7.88,
+    "train_steps_per_second": 0.13
 }

trainer_state.json

@@ -1,128 +1,43 @@
 {
-  "best_metric": 0.3437380790710449,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-85",
-  "epoch": 5.0,
+  "best_metric": 0.8840949535369873,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-17",
+  "epoch": 1.0,
   "eval_steps": 500,
-  "global_step": 85,
+  "global_step": 17,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.3447251319885254,
-      "learning_rate": 1.7647058823529414e-05,
-      "loss": 0.9867,
+      "grad_norm": 1.3551151752471924,
+      "learning_rate": 8.23529411764706e-06,
+      "loss": 0.9983,
       "step": 10
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.924812030075188,
-      "eval_loss": 0.7572891116142273,
-      "eval_runtime": 2.5634,
-      "eval_samples_per_second": 51.884,
-      "eval_steps_per_second": 1.17,
+      "eval_accuracy": 0.8646616541353384,
+      "eval_loss": 0.8840949535369873,
+      "eval_runtime": 13.6157,
+      "eval_samples_per_second": 9.768,
+      "eval_steps_per_second": 0.22,
       "step": 17
     },
     {
-      "epoch": 1.1764705882352942,
-      "grad_norm": 1.191501259803772,
-      "learning_rate": 1.5294117647058822e-05,
-      "loss": 0.8137,
-      "step": 20
-    },
-    {
-      "epoch": 1.7647058823529411,
-      "grad_norm": 1.1548258066177368,
-      "learning_rate": 1.2941176470588238e-05,
-      "loss": 0.6901,
-      "step": 30
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.5546020269393921,
-      "eval_runtime": 1.7725,
-      "eval_samples_per_second": 75.037,
-      "eval_steps_per_second": 1.693,
-      "step": 34
-    },
-    {
-      "epoch": 2.3529411764705883,
-      "grad_norm": 1.2297497987747192,
-      "learning_rate": 1.0588235294117648e-05,
-      "loss": 0.5758,
-      "step": 40
-    },
-    {
-      "epoch": 2.9411764705882355,
-      "grad_norm": 1.0743614435195923,
-      "learning_rate": 8.23529411764706e-06,
-      "loss": 0.4983,
-      "step": 50
-    },
-    {
-      "epoch": 3.0,
-      "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.4202447533607483,
-      "eval_runtime": 1.5463,
-      "eval_samples_per_second": 86.01,
-      "eval_steps_per_second": 1.94,
-      "step": 51
-    },
-    {
-      "epoch": 3.5294117647058822,
-      "grad_norm": 1.227348804473877,
-      "learning_rate": 5.882352941176471e-06,
-      "loss": 0.454,
-      "step": 60
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.36091482639312744,
-      "eval_runtime": 1.5527,
-      "eval_samples_per_second": 85.658,
-      "eval_steps_per_second": 1.932,
-      "step": 68
-    },
-    {
-      "epoch": 4.117647058823529,
-      "grad_norm": 1.0482817888259888,
-      "learning_rate": 3.529411764705883e-06,
-      "loss": 0.391,
-      "step": 70
-    },
-    {
-      "epoch": 4.705882352941177,
-      "grad_norm": 1.1002055406570435,
-      "learning_rate": 1.1764705882352942e-06,
-      "loss": 0.3921,
-      "step": 80
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.3437380790710449,
-      "eval_runtime": 1.5995,
-      "eval_samples_per_second": 83.153,
-      "eval_steps_per_second": 1.876,
-      "step": 85
-    },
-    {
-      "epoch": 5.0,
-      "step": 85,
-      "total_flos": 4.006371770595533e+17,
-      "train_loss": 0.0,
-      "train_runtime": 0.1392,
-      "train_samples_per_second": 37137.186,
-      "train_steps_per_second": 610.573
+      "epoch": 1.0,
+      "step": 17,
+      "total_flos": 8.012743541191066e+16,
+      "train_loss": 0.9630403238184312,
+      "train_runtime": 131.2191,
+      "train_samples_per_second": 7.88,
+      "train_steps_per_second": 0.13
     }
   ],
   "logging_steps": 10,
-  "max_steps": 85,
+  "max_steps": 17,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 1,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -136,7 +51,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.006371770595533e+17,
+  "total_flos": 8.012743541191066e+16,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null