🍻 cheers

README.md

@@ -2,6 +2,7 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224
 tags:
+- image-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -18,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-augmentation
 
-This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5668
-- Accuracy: 0.8804
-- Precision: 0.8823
-- Recall: 0.8804
-- F1: 0.8789
+- Loss: 0.4287
+- Accuracy: 0.8592
+- Precision: 0.8580
+- Recall: 0.8592
+- F1: 0.8574
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 16.0,
+    "eval_accuracy": 0.8592233009708737,
+    "eval_f1": 0.857445188937737,
+    "eval_loss": 0.4286545217037201,
+    "eval_precision": 0.8579724186747698,
+    "eval_recall": 0.8592233009708737,
+    "eval_runtime": 24.0354,
+    "eval_samples_per_second": 119.989,
+    "eval_steps_per_second": 15.019,
+    "total_flos": 6.35834769042545e+18,
+    "train_loss": 0.3223209912160475,
+    "train_runtime": 1869.7645,
+    "train_samples_per_second": 274.259,
+    "train_steps_per_second": 17.168
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 16.0,
+    "eval_accuracy": 0.8592233009708737,
+    "eval_f1": 0.857445188937737,
+    "eval_loss": 0.4286545217037201,
+    "eval_precision": 0.8579724186747698,
+    "eval_recall": 0.8592233009708737,
+    "eval_runtime": 24.0354,
+    "eval_samples_per_second": 119.989,
+    "eval_steps_per_second": 15.019
+}

runs/Mar20_20-15-13_e02eeb96afb5/events.out.tfevents.1710967624.e02eeb96afb5.3312.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d1df6e9ca2710005fd95bd8e45a023489a0504264892bcb5e85f3802014590f
+size 560

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 16.0,
+    "total_flos": 6.35834769042545e+18,
+    "train_loss": 0.3223209912160475,
+    "train_runtime": 1869.7645,
+    "train_samples_per_second": 274.259,
+    "train_steps_per_second": 17.168
+}

trainer_state.json

@@ -0,0 +1,334 @@
+{
+  "best_metric": 0.4286545217037201,
+  "best_model_checkpoint": "./vit-augmentation/checkpoint-1926",
+  "epoch": 16.0,
+  "eval_steps": 500,
+  "global_step": 5136,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 6.7570695877075195,
+      "learning_rate": 4.1428571428571437e-05,
+      "loss": 0.9124,
+      "step": 321
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7805131761442441,
+      "eval_f1": 0.768284202243805,
+      "eval_loss": 0.6024736166000366,
+      "eval_precision": 0.778781843539936,
+      "eval_recall": 0.7805131761442441,
+      "eval_runtime": 24.6229,
+      "eval_samples_per_second": 117.127,
+      "eval_steps_per_second": 14.661,
+      "step": 321
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 7.467956066131592,
+      "learning_rate": 8.311688311688312e-05,
+      "loss": 0.5876,
+      "step": 642
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7864077669902912,
+      "eval_f1": 0.7820269177484748,
+      "eval_loss": 0.5819488167762756,
+      "eval_precision": 0.7990366623686466,
+      "eval_recall": 0.7864077669902912,
+      "eval_runtime": 24.6924,
+      "eval_samples_per_second": 116.797,
+      "eval_steps_per_second": 14.62,
+      "step": 642
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.2375881671905518,
+      "learning_rate": 9.985334621908699e-05,
+      "loss": 0.5415,
+      "step": 963
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8040915395284327,
+      "eval_f1": 0.7864667726550387,
+      "eval_loss": 0.6148691773414612,
+      "eval_precision": 0.7943404456778191,
+      "eval_recall": 0.8040915395284327,
+      "eval_runtime": 23.4113,
+      "eval_samples_per_second": 123.188,
+      "eval_steps_per_second": 15.42,
+      "step": 963
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.80470085144043,
+      "learning_rate": 9.894936461151184e-05,
+      "loss": 0.4815,
+      "step": 1284
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8294036061026352,
+      "eval_f1": 0.8114752843394228,
+      "eval_loss": 0.4653749465942383,
+      "eval_precision": 0.8259066316761086,
+      "eval_recall": 0.8294036061026352,
+      "eval_runtime": 22.9294,
+      "eval_samples_per_second": 125.778,
+      "eval_steps_per_second": 15.744,
+      "step": 1284
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 8.0819091796875,
+      "learning_rate": 9.723506398349735e-05,
+      "loss": 0.4263,
+      "step": 1605
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8259361997226075,
+      "eval_f1": 0.8022540151903795,
+      "eval_loss": 0.5480949282646179,
+      "eval_precision": 0.8315275661759924,
+      "eval_recall": 0.8259361997226075,
+      "eval_runtime": 22.776,
+      "eval_samples_per_second": 126.625,
+      "eval_steps_per_second": 15.85,
+      "step": 1605
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.5868643522262573,
+      "learning_rate": 9.473882326123909e-05,
+      "loss": 0.3515,
+      "step": 1926
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8592233009708737,
+      "eval_f1": 0.857445188937737,
+      "eval_loss": 0.4286545217037201,
+      "eval_precision": 0.8579724186747698,
+      "eval_recall": 0.8592233009708737,
+      "eval_runtime": 22.7677,
+      "eval_samples_per_second": 126.671,
+      "eval_steps_per_second": 15.856,
+      "step": 1926
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 7.5292840003967285,
+      "learning_rate": 9.15019657867844e-05,
+      "loss": 0.3144,
+      "step": 2247
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8363384188626907,
+      "eval_f1": 0.8269755362382184,
+      "eval_loss": 0.5004583597183228,
+      "eval_precision": 0.8320109172887173,
+      "eval_recall": 0.8363384188626907,
+      "eval_runtime": 22.8371,
+      "eval_samples_per_second": 126.286,
+      "eval_steps_per_second": 15.808,
+      "step": 2247
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 9.491310119628906,
+      "learning_rate": 8.757807524194037e-05,
+      "loss": 0.2736,
+      "step": 2568
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8294036061026352,
+      "eval_f1": 0.8301623012381248,
+      "eval_loss": 0.5306037664413452,
+      "eval_precision": 0.8447810859364984,
+      "eval_recall": 0.8294036061026352,
+      "eval_runtime": 22.6737,
+      "eval_samples_per_second": 127.196,
+      "eval_steps_per_second": 15.922,
+      "step": 2568
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 5.960662841796875,
+      "learning_rate": 8.303210861170248e-05,
+      "loss": 0.2519,
+      "step": 2889
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.8578363384188626,
+      "eval_f1": 0.8534043679644312,
+      "eval_loss": 0.4732687175273895,
+      "eval_precision": 0.8533865418942713,
+      "eval_recall": 0.8578363384188626,
+      "eval_runtime": 22.5796,
+      "eval_samples_per_second": 127.726,
+      "eval_steps_per_second": 15.988,
+      "step": 2889
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 8.973663330078125,
+      "learning_rate": 7.793932087141109e-05,
+      "loss": 0.2227,
+      "step": 3210
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.8585298196948682,
+      "eval_f1": 0.851216238380984,
+      "eval_loss": 0.4905049800872803,
+      "eval_precision": 0.8519657875781843,
+      "eval_recall": 0.8585298196948682,
+      "eval_runtime": 24.7137,
+      "eval_samples_per_second": 116.696,
+      "eval_steps_per_second": 14.607,
+      "step": 3210
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 3.6425747871398926,
+      "learning_rate": 7.23840191986112e-05,
+      "loss": 0.1724,
+      "step": 3531
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.8654646324549237,
+      "eval_f1": 0.862832998223053,
+      "eval_loss": 0.5050108432769775,
+      "eval_precision": 0.8670507633275476,
+      "eval_recall": 0.8654646324549237,
+      "eval_runtime": 24.5636,
+      "eval_samples_per_second": 117.409,
+      "eval_steps_per_second": 14.697,
+      "step": 3531
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 14.814494132995605,
+      "learning_rate": 6.64581673326787e-05,
+      "loss": 0.1596,
+      "step": 3852
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.8685852981969486,
+      "eval_f1": 0.8631462417126987,
+      "eval_loss": 0.526269257068634,
+      "eval_precision": 0.8656616085283582,
+      "eval_recall": 0.8685852981969486,
+      "eval_runtime": 24.7987,
+      "eval_samples_per_second": 116.296,
+      "eval_steps_per_second": 14.557,
+      "step": 3852
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.3064632415771484,
+      "learning_rate": 6.027949045818934e-05,
+      "loss": 0.1397,
+      "step": 4173
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.8533287101248266,
+      "eval_f1": 0.848769203208603,
+      "eval_loss": 0.7043444514274597,
+      "eval_precision": 0.8702783792779959,
+      "eval_recall": 0.8533287101248266,
+      "eval_runtime": 24.4467,
+      "eval_samples_per_second": 117.971,
+      "eval_steps_per_second": 14.767,
+      "step": 4173
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 0.027323821559548378,
+      "learning_rate": 5.391170860718704e-05,
+      "loss": 0.1298,
+      "step": 4494
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.8678918169209431,
+      "eval_f1": 0.8631936421240504,
+      "eval_loss": 0.6274930834770203,
+      "eval_precision": 0.8734415776541141,
+      "eval_recall": 0.8678918169209431,
+      "eval_runtime": 24.4385,
+      "eval_samples_per_second": 118.01,
+      "eval_steps_per_second": 14.772,
+      "step": 4494
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 18.411346435546875,
+      "learning_rate": 4.7479171433782145e-05,
+      "loss": 0.1029,
+      "step": 4815
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.8807212205270458,
+      "eval_f1": 0.8772158509937853,
+      "eval_loss": 0.5564317107200623,
+      "eval_precision": 0.8775761761640355,
+      "eval_recall": 0.8807212205270458,
+      "eval_runtime": 24.4883,
+      "eval_samples_per_second": 117.771,
+      "eval_steps_per_second": 14.742,
+      "step": 4815
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 17.630950927734375,
+      "learning_rate": 4.1088364635238816e-05,
+      "loss": 0.0893,
+      "step": 5136
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.880374479889043,
+      "eval_f1": 0.8789241435580383,
+      "eval_loss": 0.5668464303016663,
+      "eval_precision": 0.8823340407710801,
+      "eval_recall": 0.880374479889043,
+      "eval_runtime": 24.4769,
+      "eval_samples_per_second": 117.825,
+      "eval_steps_per_second": 14.749,
+      "step": 5136
+    },
+    {
+      "epoch": 16.0,
+      "step": 5136,
+      "total_flos": 6.35834769042545e+18,
+      "train_loss": 0.3223209912160475,
+      "train_runtime": 1869.7645,
+      "train_samples_per_second": 274.259,
+      "train_steps_per_second": 17.168
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 32100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 100,
+  "save_steps": 500,
+  "total_flos": 6.35834769042545e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}