🍻 cheers

README.md

@@ -2,7 +2,11 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
+- vision
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: vit-base-beans
   results: []
@@ -13,7 +17,10 @@ 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: 0.0172
+- Accuracy: 1.0
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.024416429921984673,
-    "eval_runtime": 2.0478,
-    "eval_samples_per_second": 28.324,
-    "eval_steps_per_second": 3.907,
-    "total_flos": 1.7978141590585344e+16,
-    "train_loss": 0.3237314224243164,
-    "train_runtime": 22.8416,
-    "train_samples_per_second": 10.157,
-    "train_steps_per_second": 2.627
+    "eval_loss": 0.01724555343389511,
+    "eval_runtime": 2.9509,
+    "eval_samples_per_second": 29.822,
+    "eval_steps_per_second": 3.728,
+    "total_flos": 2.7277180344336384e+16,
+    "train_loss": 0.29457727345553314,
+    "train_runtime": 29.6451,
+    "train_samples_per_second": 11.874,
+    "train_steps_per_second": 2.968
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.024416429921984673,
-    "eval_runtime": 2.0478,
-    "eval_samples_per_second": 28.324,
-    "eval_steps_per_second": 3.907
+    "eval_loss": 0.01724555343389511,
+    "eval_runtime": 2.9509,
+    "eval_samples_per_second": 29.822,
+    "eval_steps_per_second": 3.728
 }

runs/Aug30_08-56-26_7e2de4304170/events.out.tfevents.1725008239.7e2de4304170

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:baa5827c224a251d2762b3ba503aef52f2838e578619c1a16d3f754a6a24d66c
+size 40

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "total_flos": 1.7978141590585344e+16,
-    "train_loss": 0.3237314224243164,
-    "train_runtime": 22.8416,
-    "train_samples_per_second": 10.157,
-    "train_steps_per_second": 2.627
+    "total_flos": 2.7277180344336384e+16,
+    "train_loss": 0.29457727345553314,
+    "train_runtime": 29.6451,
+    "train_samples_per_second": 11.874,
+    "train_steps_per_second": 2.968
 }

trainer_state.json

@@ -3,65 +3,79 @@
   "best_model_checkpoint": null,
   "epoch": 4.0,
   "eval_steps": 100,
-  "global_step": 60,
+  "global_step": 88,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.6666666666666666,
-      "grad_norm": 2.1028881072998047,
-      "learning_rate": 0.0001666666666666667,
-      "loss": 0.609,
+      "epoch": 0.45454545454545453,
+      "grad_norm": 2.2645957469940186,
+      "learning_rate": 0.00017727272727272728,
+      "loss": 0.5609,
       "step": 10
     },
     {
-      "epoch": 1.3333333333333333,
-      "grad_norm": 5.82706880569458,
-      "learning_rate": 0.00013333333333333334,
-      "loss": 0.6172,
+      "epoch": 0.9090909090909091,
+      "grad_norm": 3.18221116065979,
+      "learning_rate": 0.00015454545454545454,
+      "loss": 0.8267,
       "step": 20
     },
     {
-      "epoch": 2.0,
-      "grad_norm": 10.371822357177734,
-      "learning_rate": 0.0001,
-      "loss": 0.4026,
+      "epoch": 1.3636363636363638,
+      "grad_norm": 3.1673595905303955,
+      "learning_rate": 0.0001318181818181818,
+      "loss": 0.5182,
       "step": 30
     },
     {
-      "epoch": 2.6666666666666665,
-      "grad_norm": 0.941929817199707,
-      "learning_rate": 6.666666666666667e-05,
-      "loss": 0.171,
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.5910093784332275,
+      "learning_rate": 0.00010909090909090909,
+      "loss": 0.2435,
       "step": 40
     },
     {
-      "epoch": 3.3333333333333335,
-      "grad_norm": 0.1865934133529663,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.114,
+      "epoch": 2.2727272727272725,
+      "grad_norm": 3.6887409687042236,
+      "learning_rate": 8.636363636363637e-05,
+      "loss": 0.2545,
       "step": 50
     },
     {
-      "epoch": 4.0,
-      "grad_norm": 0.20858462154865265,
-      "learning_rate": 0.0,
-      "loss": 0.0287,
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.7244749665260315,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": 0.1302,
       "step": 60
     },
+    {
+      "epoch": 3.1818181818181817,
+      "grad_norm": 0.1487322300672531,
+      "learning_rate": 4.0909090909090915e-05,
+      "loss": 0.0252,
+      "step": 70
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.1414584219455719,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 0.0192,
+      "step": 80
+    },
     {
       "epoch": 4.0,
-      "step": 60,
-      "total_flos": 1.7978141590585344e+16,
-      "train_loss": 0.3237314224243164,
-      "train_runtime": 22.8416,
-      "train_samples_per_second": 10.157,
-      "train_steps_per_second": 2.627
+      "step": 88,
+      "total_flos": 2.7277180344336384e+16,
+      "train_loss": 0.29457727345553314,
+      "train_runtime": 29.6451,
+      "train_samples_per_second": 11.874,
+      "train_steps_per_second": 2.968
     }
   ],
   "logging_steps": 10,
-  "max_steps": 60,
+  "max_steps": 88,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 100,
@@ -77,7 +91,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.7978141590585344e+16,
+  "total_flos": 2.7277180344336384e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null