cheers again

README.md

@@ -2,6 +2,7 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # test-hasy-7
 
-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 None 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 HASY dataset.
 It achieves the following results on the evaluation set:
 - Loss: 4.0260
 - Accuracy: 0.2786

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.41995841995842,
-    "eval_loss": 3.498389959335327,
-    "eval_runtime": 1.6011,
-    "eval_samples_per_second": 300.417,
-    "eval_steps_per_second": 38.099,
-    "total_flos": 6.710386542096384e+17,
-    "train_loss": 3.9836128502808745,
-    "train_runtime": 79.2434,
-    "train_samples_per_second": 109.157,
-    "train_steps_per_second": 13.654
+    "epoch": 1.0,
+    "eval_accuracy": 0.2785862785862786,
+    "eval_loss": 4.025966167449951,
+    "eval_runtime": 2.0941,
+    "eval_samples_per_second": 229.689,
+    "eval_steps_per_second": 29.129,
+    "total_flos": 3.355193271048192e+17,
+    "train_loss": 4.313544084756961,
+    "train_runtime": 56.6601,
+    "train_samples_per_second": 76.332,
+    "train_steps_per_second": 9.548
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.41995841995842,
-    "eval_loss": 3.498389959335327,
-    "eval_runtime": 1.6011,
-    "eval_samples_per_second": 300.417,
-    "eval_steps_per_second": 38.099
+    "epoch": 1.0,
+    "eval_accuracy": 0.2785862785862786,
+    "eval_loss": 4.025966167449951,
+    "eval_runtime": 2.0941,
+    "eval_samples_per_second": 229.689,
+    "eval_steps_per_second": 29.129
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 6.710386542096384e+17,
-    "train_loss": 3.9836128502808745,
-    "train_runtime": 79.2434,
-    "train_samples_per_second": 109.157,
-    "train_steps_per_second": 13.654
+    "epoch": 1.0,
+    "total_flos": 3.355193271048192e+17,
+    "train_loss": 4.313544084756961,
+    "train_runtime": 56.6601,
+    "train_samples_per_second": 76.332,
+    "train_steps_per_second": 9.548
 }

trainer_state.json

@@ -1,61 +1,45 @@
 {
-  "best_metric": 0.41995841995842,
-  "best_model_checkpoint": "test-hasy-7/checkpoint-1082",
-  "epoch": 2.0,
+  "best_metric": 0.2785862785862786,
+  "best_model_checkpoint": "test-hasy-7/checkpoint-541",
+  "epoch": 1.0,
   "eval_steps": 500,
-  "global_step": 1082,
+  "global_step": 541,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.92,
-      "grad_norm": 2.9993338584899902,
-      "learning_rate": 1.0757855822550833e-05,
-      "loss": 4.2372,
+      "grad_norm": 3.252389669418335,
+      "learning_rate": 1.5157116451016638e-06,
+      "loss": 4.3305,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.367983367983368,
-      "eval_loss": 3.7456552982330322,
-      "eval_runtime": 1.5608,
-      "eval_samples_per_second": 308.179,
-      "eval_steps_per_second": 39.083,
+      "eval_accuracy": 0.2785862785862786,
+      "eval_loss": 4.025966167449951,
+      "eval_runtime": 2.0515,
+      "eval_samples_per_second": 234.458,
+      "eval_steps_per_second": 29.734,
       "step": 541
     },
     {
-      "epoch": 1.85,
-      "grad_norm": 4.245122909545898,
-      "learning_rate": 1.5157116451016638e-06,
-      "loss": 3.7808,
-      "step": 1000
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.41995841995842,
-      "eval_loss": 3.498389959335327,
-      "eval_runtime": 1.5566,
-      "eval_samples_per_second": 308.998,
-      "eval_steps_per_second": 39.187,
-      "step": 1082
-    },
-    {
-      "epoch": 2.0,
-      "step": 1082,
-      "total_flos": 6.710386542096384e+17,
-      "train_loss": 3.9836128502808745,
-      "train_runtime": 79.2434,
-      "train_samples_per_second": 109.157,
-      "train_steps_per_second": 13.654
+      "epoch": 1.0,
+      "step": 541,
+      "total_flos": 3.355193271048192e+17,
+      "train_loss": 4.313544084756961,
+      "train_runtime": 56.6601,
+      "train_samples_per_second": 76.332,
+      "train_steps_per_second": 9.548
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1082,
+  "max_steps": 541,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 1,
   "save_steps": 500,
-  "total_flos": 6.710386542096384e+17,
+  "total_flos": 3.355193271048192e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null