🍻 cheers

README.md

@@ -3,6 +3,7 @@ library_name: transformers
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
 - generated_from_trainer
 datasets:
 - imagefolder
@@ -15,7 +16,7 @@ model-index:
       name: Image Classification
       type: image-classification
     dataset:
-      name: imagefolder
+      name: finetuned-websites
       type: imagefolder
       config: default
       split: train
@@ -31,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # finetuned-websites
 
-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 imagefolder 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 finetuned-websites dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.8349
 - Accuracy: 0.75

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.39705882352941174,
-    "eval_loss": 1.5659970045089722,
-    "eval_runtime": 3.0691,
-    "eval_samples_per_second": 22.156,
-    "eval_steps_per_second": 2.932,
+    "eval_accuracy": 0.75,
+    "eval_loss": 0.8348819613456726,
+    "eval_runtime": 3.7302,
+    "eval_samples_per_second": 18.23,
+    "eval_steps_per_second": 2.413,
     "total_flos": 2.9603266596900864e+17,
-    "train_loss": 1.4860339760780334,
-    "train_runtime": 184.3796,
-    "train_samples_per_second": 20.718,
-    "train_steps_per_second": 1.302
+    "train_loss": 0.6847666402657827,
+    "train_runtime": 183.3512,
+    "train_samples_per_second": 20.834,
+    "train_steps_per_second": 1.309
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.39705882352941174,
-    "eval_loss": 1.5659970045089722,
-    "eval_runtime": 3.0691,
-    "eval_samples_per_second": 22.156,
-    "eval_steps_per_second": 2.932
+    "eval_accuracy": 0.75,
+    "eval_loss": 0.8348819613456726,
+    "eval_runtime": 3.7302,
+    "eval_samples_per_second": 18.23,
+    "eval_steps_per_second": 2.413
 }

runs/Dec05_12-11-38_edab03f22c59/events.out.tfevents.1733400953.edab03f22c59.4058.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ff60e75deea3850000873a6821a2bfe39847ef82669f7f2152048dac3d26cd2
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 2.9603266596900864e+17,
-    "train_loss": 1.4860339760780334,
-    "train_runtime": 184.3796,
-    "train_samples_per_second": 20.718,
-    "train_steps_per_second": 1.302
+    "train_loss": 0.6847666402657827,
+    "train_runtime": 183.3512,
+    "train_samples_per_second": 20.834,
+    "train_steps_per_second": 1.309
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 1.5659970045089722,
+  "best_metric": 0.8348819613456726,
   "best_model_checkpoint": "finetuned-websites/checkpoint-200",
   "epoch": 10.0,
-  "eval_steps": 50,
+  "eval_steps": 100,
   "global_step": 240,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
@@ -10,223 +10,205 @@
   "log_history": [
     {
       "epoch": 0.4166666666666667,
-      "grad_norm": 4.125311851501465,
-      "learning_rate": 1.0000000000000002e-06,
-      "loss": 1.9272,
+      "grad_norm": 6.133713722229004,
+      "learning_rate": 1e-05,
+      "loss": 1.0596,
       "step": 10
     },
     {
       "epoch": 0.8333333333333334,
-      "grad_norm": 3.0228281021118164,
-      "learning_rate": 2.0000000000000003e-06,
-      "loss": 1.928,
+      "grad_norm": 5.39012336730957,
+      "learning_rate": 2e-05,
+      "loss": 1.0523,
       "step": 20
     },
     {
       "epoch": 1.25,
-      "grad_norm": 2.5070629119873047,
-      "learning_rate": 3e-06,
-      "loss": 1.921,
+      "grad_norm": 4.8409881591796875,
+      "learning_rate": 3e-05,
+      "loss": 1.02,
       "step": 30
     },
     {
       "epoch": 1.6666666666666665,
-      "grad_norm": 2.7787866592407227,
-      "learning_rate": 4.000000000000001e-06,
-      "loss": 1.8972,
+      "grad_norm": 5.128672122955322,
+      "learning_rate": 4e-05,
+      "loss": 0.9759,
       "step": 40
     },
     {
       "epoch": 2.0833333333333335,
-      "grad_norm": 2.7862040996551514,
-      "learning_rate": 5e-06,
-      "loss": 1.8678,
-      "step": 50
-    },
-    {
-      "epoch": 2.0833333333333335,
-      "eval_accuracy": 0.3088235294117647,
-      "eval_loss": 1.8965561389923096,
-      "eval_runtime": 2.6013,
-      "eval_samples_per_second": 26.141,
-      "eval_steps_per_second": 3.46,
+      "grad_norm": 6.329381465911865,
+      "learning_rate": 5e-05,
+      "loss": 0.9153,
       "step": 50
     },
     {
       "epoch": 2.5,
-      "grad_norm": 3.540709972381592,
-      "learning_rate": 6e-06,
-      "loss": 1.8526,
+      "grad_norm": 6.218069076538086,
+      "learning_rate": 6e-05,
+      "loss": 0.8539,
       "step": 60
     },
     {
       "epoch": 2.9166666666666665,
-      "grad_norm": 3.426781415939331,
-      "learning_rate": 7.000000000000001e-06,
-      "loss": 1.8198,
+      "grad_norm": 5.5271430015563965,
+      "learning_rate": 7.000000000000001e-05,
+      "loss": 0.8747,
       "step": 70
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 4.007580757141113,
-      "learning_rate": 8.000000000000001e-06,
-      "loss": 1.7369,
+      "grad_norm": 8.148289680480957,
+      "learning_rate": 8e-05,
+      "loss": 0.7716,
       "step": 80
     },
     {
       "epoch": 3.75,
-      "grad_norm": 4.467859745025635,
-      "learning_rate": 9e-06,
-      "loss": 1.7144,
+      "grad_norm": 5.050111293792725,
+      "learning_rate": 8.9e-05,
+      "loss": 0.8763,
       "step": 90
     },
     {
       "epoch": 4.166666666666667,
-      "grad_norm": 3.9431729316711426,
-      "learning_rate": 1e-05,
-      "loss": 1.6483,
+      "grad_norm": 5.70500373840332,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 0.6802,
       "step": 100
     },
     {
       "epoch": 4.166666666666667,
-      "eval_accuracy": 0.3382352941176471,
-      "eval_loss": 1.793816089630127,
-      "eval_runtime": 2.9764,
-      "eval_samples_per_second": 22.847,
-      "eval_steps_per_second": 3.024,
+      "eval_accuracy": 0.5735294117647058,
+      "eval_loss": 1.2561215162277222,
+      "eval_runtime": 2.1458,
+      "eval_samples_per_second": 31.69,
+      "eval_steps_per_second": 4.194,
       "step": 100
     },
     {
       "epoch": 4.583333333333333,
-      "grad_norm": 3.2608771324157715,
-      "learning_rate": 1.1000000000000001e-05,
-      "loss": 1.5856,
+      "grad_norm": 6.188072204589844,
+      "learning_rate": 0.000109,
+      "loss": 0.6251,
       "step": 110
     },
     {
       "epoch": 5.0,
-      "grad_norm": 3.638972520828247,
-      "learning_rate": 1.2e-05,
-      "loss": 1.5157,
+      "grad_norm": 9.144834518432617,
+      "learning_rate": 0.00011899999999999999,
+      "loss": 0.6689,
       "step": 120
     },
     {
       "epoch": 5.416666666666667,
-      "grad_norm": 3.3878164291381836,
-      "learning_rate": 1.3000000000000001e-05,
-      "loss": 1.4388,
+      "grad_norm": 5.840977668762207,
+      "learning_rate": 0.00012900000000000002,
+      "loss": 0.5262,
       "step": 130
     },
     {
       "epoch": 5.833333333333333,
-      "grad_norm": 4.086641311645508,
-      "learning_rate": 1.4000000000000001e-05,
-      "loss": 1.4522,
+      "grad_norm": 22.520524978637695,
+      "learning_rate": 0.00013900000000000002,
+      "loss": 0.6559,
       "step": 140
     },
     {
       "epoch": 6.25,
-      "grad_norm": 3.5550928115844727,
-      "learning_rate": 1.5e-05,
-      "loss": 1.3419,
-      "step": 150
-    },
-    {
-      "epoch": 6.25,
-      "eval_accuracy": 0.38235294117647056,
-      "eval_loss": 1.6614487171173096,
-      "eval_runtime": 2.1852,
-      "eval_samples_per_second": 31.118,
-      "eval_steps_per_second": 4.119,
+      "grad_norm": 7.3676676750183105,
+      "learning_rate": 0.000149,
+      "loss": 0.5186,
       "step": 150
     },
     {
       "epoch": 6.666666666666667,
-      "grad_norm": 4.774816036224365,
-      "learning_rate": 1.6000000000000003e-05,
-      "loss": 1.3147,
+      "grad_norm": 16.96535873413086,
+      "learning_rate": 0.00015900000000000002,
+      "loss": 0.4291,
       "step": 160
     },
     {
       "epoch": 7.083333333333333,
-      "grad_norm": 4.092309951782227,
-      "learning_rate": 1.7000000000000003e-05,
-      "loss": 1.2643,
+      "grad_norm": 4.878167629241943,
+      "learning_rate": 0.00016900000000000002,
+      "loss": 0.6311,
       "step": 170
     },
     {
       "epoch": 7.5,
-      "grad_norm": 4.992920398712158,
-      "learning_rate": 1.8e-05,
-      "loss": 1.1924,
+      "grad_norm": 5.731603145599365,
+      "learning_rate": 0.000179,
+      "loss": 0.4274,
       "step": 180
     },
     {
       "epoch": 7.916666666666667,
-      "grad_norm": 6.251681327819824,
-      "learning_rate": 1.9e-05,
-      "loss": 1.1461,
+      "grad_norm": 12.268874168395996,
+      "learning_rate": 0.000189,
+      "loss": 0.3916,
       "step": 190
     },
     {
       "epoch": 8.333333333333334,
-      "grad_norm": 5.500473499298096,
-      "learning_rate": 2e-05,
-      "loss": 1.1189,
+      "grad_norm": 19.5472469329834,
+      "learning_rate": 0.000199,
+      "loss": 0.3727,
       "step": 200
     },
     {
       "epoch": 8.333333333333334,
-      "eval_accuracy": 0.39705882352941174,
-      "eval_loss": 1.5659970045089722,
-      "eval_runtime": 2.0786,
-      "eval_samples_per_second": 32.715,
-      "eval_steps_per_second": 4.33,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.8348819613456726,
+      "eval_runtime": 2.0932,
+      "eval_samples_per_second": 32.486,
+      "eval_steps_per_second": 4.3,
       "step": 200
     },
     {
       "epoch": 8.75,
-      "grad_norm": 4.436168670654297,
-      "learning_rate": 2.1e-05,
-      "loss": 1.0734,
+      "grad_norm": 9.300850868225098,
+      "learning_rate": 0.00020899999999999998,
+      "loss": 0.5022,
       "step": 210
     },
     {
       "epoch": 9.166666666666666,
-      "grad_norm": 4.706101417541504,
-      "learning_rate": 2.2000000000000003e-05,
-      "loss": 0.9833,
+      "grad_norm": 2.52441668510437,
+      "learning_rate": 0.000219,
+      "loss": 0.3873,
       "step": 220
     },
     {
       "epoch": 9.583333333333334,
-      "grad_norm": 4.173801898956299,
-      "learning_rate": 2.3000000000000003e-05,
-      "loss": 0.9902,
+      "grad_norm": 7.8756794929504395,
+      "learning_rate": 0.000229,
+      "loss": 0.5663,
       "step": 230
     },
     {
       "epoch": 10.0,
-      "grad_norm": 5.312504291534424,
-      "learning_rate": 2.4e-05,
-      "loss": 0.9339,
+      "grad_norm": 7.985933780670166,
+      "learning_rate": 0.00023899999999999998,
+      "loss": 0.6521,
       "step": 240
     },
     {
       "epoch": 10.0,
       "step": 240,
       "total_flos": 2.9603266596900864e+17,
-      "train_loss": 1.4860339760780334,
-      "train_runtime": 184.3796,
-      "train_samples_per_second": 20.718,
-      "train_steps_per_second": 1.302
+      "train_loss": 0.6847666402657827,
+      "train_runtime": 183.3512,
+      "train_samples_per_second": 20.834,
+      "train_steps_per_second": 1.309
     }
   ],
   "logging_steps": 10,
   "max_steps": 240,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
-  "save_steps": 50,
+  "save_steps": 100,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {