vit-base-kidney-stone-Michel_Daudon_-w256_1k_v1-_SUR

README.md

@@ -1,4 +1,5 @@
 ---
+library_name: transformers
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
@@ -25,30 +26,31 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6966475878986099
+      value: 0.7367130008176614
     - name: Precision
       type: precision
-      value: 0.7395056708514637
+      value: 0.7595070859879842
     - name: Recall
       type: recall
-      value: 0.6966475878986099
+      value: 0.7367130008176614
     - name: F1
       type: f1
-      value: 0.7121074355920559
+      value: 0.7402764231487609
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/cv-inside/vit-base-kidney-stone/runs/6e597kni)
 # vit-base-kidney-stone-Michel_Daudon_-w256_1k_v1-_SUR
 
 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.
 It achieves the following results on the evaluation set:
-- Loss: 0.9000
-- Accuracy: 0.6966
-- Precision: 0.7395
-- Recall: 0.6966
-- F1: 0.7121
+- Loss: 0.8522
+- Accuracy: 0.7367
+- Precision: 0.7595
+- Recall: 0.7367
+- F1: 0.7403
 
 ## Model description
 
@@ -71,21 +73,21 @@ The following hyperparameters were used during training:
 - train_batch_size: 32
 - eval_batch_size: 8
 - seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - num_epochs: 1
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
-| 0.0973        | 0.67  | 100  | 0.9000          | 0.6966   | 0.7395    | 0.6966 | 0.7121 |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
+| 0.117         | 0.6667 | 100  | 0.8522          | 0.7367   | 0.7595    | 0.7367 | 0.7403 |
 
 
 ### Framework versions
 
-- Transformers 4.37.2
-- Pytorch 2.1.1
-- Datasets 3.1.0
-- Tokenizers 0.15.2
+- Transformers 4.48.2
+- Pytorch 2.6.0+cu126
+- Datasets 3.2.0
+- Tokenizers 0.21.0

all_results.json

@@ -1,15 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.6966475878986099,
-    "eval_f1": 0.7121074355920559,
-    "eval_loss": 0.9000147581100464,
-    "eval_precision": 0.7395056708514637,
-    "eval_recall": 0.6966475878986099,
-    "eval_runtime": 21.0464,
-    "eval_samples_per_second": 58.11,
-    "eval_steps_per_second": 7.27,
-    "train_loss": 0.2848571586608887,
-    "train_runtime": 93.0462,
-    "train_samples_per_second": 51.587,
-    "train_steps_per_second": 1.612
+    "eval_accuracy": 0.7367130008176614,
+    "eval_f1": 0.7402764231487609,
+    "eval_loss": 0.8521540760993958,
+    "eval_precision": 0.7595070859879842,
+    "eval_recall": 0.7367130008176614,
+    "eval_runtime": 8.8089,
+    "eval_samples_per_second": 138.836,
+    "eval_steps_per_second": 17.369,
+    "total_flos": 3.71974885244928e+17,
+    "train_loss": 0.27923648834228515,
+    "train_runtime": 74.1686,
+    "train_samples_per_second": 64.717,
+    "train_steps_per_second": 2.022
 }

config.json

@@ -36,5 +36,5 @@
   "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.37.2"
+  "transformers_version": "4.48.2"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2402e84ac1d3376154cdb20eb7136f7a480c89215744196d27845b1c0dcc393c
+oid sha256:f4322113c6a295dbef1d54f4a14f71dc9b802a1413719b1812f4add0a69d2a3a
 size 343236280

preprocessor_config.json

@@ -1,4 +1,5 @@
 {
+  "do_convert_rgb": null,
   "do_normalize": true,
   "do_rescale": true,
   "do_resize": true,

test_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.6966475878986099,
-    "eval_f1": 0.7121074355920559,
-    "eval_loss": 0.9000147581100464,
-    "eval_precision": 0.7395056708514637,
-    "eval_recall": 0.6966475878986099,
-    "eval_runtime": 21.0464,
-    "eval_samples_per_second": 58.11,
-    "eval_steps_per_second": 7.27
+    "eval_accuracy": 0.7367130008176614,
+    "eval_f1": 0.7402764231487609,
+    "eval_loss": 0.8521540760993958,
+    "eval_precision": 0.7595070859879842,
+    "eval_recall": 0.7367130008176614,
+    "eval_runtime": 8.8089,
+    "eval_samples_per_second": 138.836,
+    "eval_steps_per_second": 17.369
 }

train_results.json

@@ -1,7 +1,8 @@
 {
     "epoch": 1.0,
-    "train_loss": 0.2848571586608887,
-    "train_runtime": 93.0462,
-    "train_samples_per_second": 51.587,
-    "train_steps_per_second": 1.612
+    "total_flos": 3.71974885244928e+17,
+    "train_loss": 0.27923648834228515,
+    "train_runtime": 74.1686,
+    "train_samples_per_second": 64.717,
+    "train_steps_per_second": 2.022
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9000147581100464,
+  "best_metric": 0.8521540760993958,
   "best_model_checkpoint": "vit-base-kidney-stone-Michel_Daudon_-w256_1k_v1-_SUR\\checkpoint-100",
   "epoch": 1.0,
   "eval_steps": 100,
@@ -9,205 +9,235 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.03,
+      "epoch": 0.03333333333333333,
+      "grad_norm": 1.9283610582351685,
       "learning_rate": 0.00019333333333333333,
-      "loss": 1.6122,
+      "loss": 1.6434,
       "step": 5
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.06666666666666667,
+      "grad_norm": 2.5478172302246094,
       "learning_rate": 0.0001866666666666667,
-      "loss": 1.1761,
+      "loss": 1.251,
       "step": 10
     },
     {
       "epoch": 0.1,
+      "grad_norm": 1.469027042388916,
       "learning_rate": 0.00018,
-      "loss": 0.8586,
+      "loss": 0.9211,
       "step": 15
     },
     {
-      "epoch": 0.13,
+      "epoch": 0.13333333333333333,
+      "grad_norm": 1.6149502992630005,
       "learning_rate": 0.00017333333333333334,
-      "loss": 0.6213,
+      "loss": 0.6539,
       "step": 20
     },
     {
-      "epoch": 0.17,
+      "epoch": 0.16666666666666666,
+      "grad_norm": 2.1767466068267822,
       "learning_rate": 0.0001666666666666667,
-      "loss": 0.4837,
+      "loss": 0.5089,
       "step": 25
     },
     {
       "epoch": 0.2,
+      "grad_norm": 2.2109217643737793,
       "learning_rate": 0.00016,
-      "loss": 0.4179,
+      "loss": 0.3865,
       "step": 30
     },
     {
-      "epoch": 0.23,
+      "epoch": 0.23333333333333334,
+      "grad_norm": 1.071829080581665,
       "learning_rate": 0.00015333333333333334,
-      "loss": 0.3188,
+      "loss": 0.3182,
       "step": 35
     },
     {
-      "epoch": 0.27,
+      "epoch": 0.26666666666666666,
+      "grad_norm": 4.0049943923950195,
       "learning_rate": 0.00014666666666666666,
-      "loss": 0.2558,
+      "loss": 0.3118,
       "step": 40
     },
     {
       "epoch": 0.3,
+      "grad_norm": 1.0740928649902344,
       "learning_rate": 0.00014,
-      "loss": 0.2618,
+      "loss": 0.2968,
       "step": 45
     },
     {
-      "epoch": 0.33,
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5208770036697388,
       "learning_rate": 0.00013333333333333334,
-      "loss": 0.2922,
+      "loss": 0.1959,
       "step": 50
     },
     {
-      "epoch": 0.37,
+      "epoch": 0.36666666666666664,
+      "grad_norm": 1.7292317152023315,
       "learning_rate": 0.00012666666666666666,
-      "loss": 0.2404,
+      "loss": 0.1556,
       "step": 55
     },
     {
       "epoch": 0.4,
+      "grad_norm": 1.2045090198516846,
       "learning_rate": 0.00012,
-      "loss": 0.1868,
+      "loss": 0.1415,
       "step": 60
     },
     {
-      "epoch": 0.43,
+      "epoch": 0.43333333333333335,
+      "grad_norm": 0.33121228218078613,
       "learning_rate": 0.00011333333333333334,
-      "loss": 0.119,
+      "loss": 0.1212,
       "step": 65
     },
     {
-      "epoch": 0.47,
+      "epoch": 0.4666666666666667,
+      "grad_norm": 2.1612660884857178,
       "learning_rate": 0.00010666666666666667,
-      "loss": 0.1586,
+      "loss": 0.1637,
       "step": 70
     },
     {
       "epoch": 0.5,
+      "grad_norm": 0.30413818359375,
       "learning_rate": 0.0001,
-      "loss": 0.1398,
+      "loss": 0.116,
       "step": 75
     },
     {
-      "epoch": 0.53,
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.23730212450027466,
       "learning_rate": 9.333333333333334e-05,
-      "loss": 0.1169,
+      "loss": 0.0844,
       "step": 80
     },
     {
-      "epoch": 0.57,
+      "epoch": 0.5666666666666667,
+      "grad_norm": 3.734177827835083,
       "learning_rate": 8.666666666666667e-05,
-      "loss": 0.1334,
+      "loss": 0.093,
       "step": 85
     },
     {
       "epoch": 0.6,
+      "grad_norm": 1.837829828262329,
       "learning_rate": 8e-05,
-      "loss": 0.115,
+      "loss": 0.1055,
       "step": 90
     },
     {
-      "epoch": 0.63,
+      "epoch": 0.6333333333333333,
+      "grad_norm": 0.41846963763237,
       "learning_rate": 7.333333333333333e-05,
-      "loss": 0.1426,
+      "loss": 0.0708,
       "step": 95
     },
     {
-      "epoch": 0.67,
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.37256985902786255,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 0.0973,
+      "loss": 0.117,
       "step": 100
     },
     {
-      "epoch": 0.67,
-      "eval_accuracy": 0.6966475878986099,
-      "eval_f1": 0.7121074355920559,
-      "eval_loss": 0.9000147581100464,
-      "eval_precision": 0.7395056708514637,
-      "eval_recall": 0.6966475878986099,
-      "eval_runtime": 20.3384,
-      "eval_samples_per_second": 60.132,
-      "eval_steps_per_second": 7.523,
+      "epoch": 0.6666666666666666,
+      "eval_accuracy": 0.7367130008176614,
+      "eval_f1": 0.7402764231487609,
+      "eval_loss": 0.8521540760993958,
+      "eval_precision": 0.7595070859879842,
+      "eval_recall": 0.7367130008176614,
+      "eval_runtime": 10.0946,
+      "eval_samples_per_second": 121.154,
+      "eval_steps_per_second": 15.157,
       "step": 100
     },
     {
       "epoch": 0.7,
+      "grad_norm": 1.851919412612915,
       "learning_rate": 6e-05,
-      "loss": 0.0894,
+      "loss": 0.1259,
       "step": 105
     },
     {
-      "epoch": 0.73,
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.20213304460048676,
       "learning_rate": 5.333333333333333e-05,
-      "loss": 0.0841,
+      "loss": 0.0944,
       "step": 110
     },
     {
-      "epoch": 0.77,
+      "epoch": 0.7666666666666667,
+      "grad_norm": 2.9062812328338623,
       "learning_rate": 4.666666666666667e-05,
-      "loss": 0.1361,
+      "loss": 0.0958,
       "step": 115
     },
     {
       "epoch": 0.8,
+      "grad_norm": 0.41974860429763794,
       "learning_rate": 4e-05,
-      "loss": 0.0673,
+      "loss": 0.0599,
       "step": 120
     },
     {
-      "epoch": 0.83,
+      "epoch": 0.8333333333333334,
+      "grad_norm": 0.17554304003715515,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0614,
+      "loss": 0.057,
       "step": 125
     },
     {
-      "epoch": 0.87,
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.9086737036705017,
       "learning_rate": 2.6666666666666667e-05,
-      "loss": 0.062,
+      "loss": 0.0569,
       "step": 130
     },
     {
       "epoch": 0.9,
+      "grad_norm": 4.543064594268799,
       "learning_rate": 2e-05,
-      "loss": 0.0775,
+      "loss": 0.0609,
       "step": 135
     },
     {
-      "epoch": 0.93,
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.3038415312767029,
       "learning_rate": 1.3333333333333333e-05,
-      "loss": 0.1043,
+      "loss": 0.0565,
       "step": 140
     },
     {
-      "epoch": 0.97,
+      "epoch": 0.9666666666666667,
+      "grad_norm": 0.17050418257713318,
       "learning_rate": 6.666666666666667e-06,
-      "loss": 0.0585,
+      "loss": 0.0524,
       "step": 145
     },
     {
       "epoch": 1.0,
+      "grad_norm": 0.18213202059268951,
       "learning_rate": 0.0,
-      "loss": 0.0568,
+      "loss": 0.0612,
       "step": 150
     },
     {
       "epoch": 1.0,
       "step": 150,
       "total_flos": 3.71974885244928e+17,
-      "train_loss": 0.2848571586608887,
-      "train_runtime": 93.0462,
-      "train_samples_per_second": 51.587,
-      "train_steps_per_second": 1.612
+      "train_loss": 0.27923648834228515,
+      "train_runtime": 74.1686,
+      "train_samples_per_second": 64.717,
+      "train_steps_per_second": 2.022
     }
   ],
   "logging_steps": 5,
@@ -215,6 +245,18 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
   "total_flos": 3.71974885244928e+17,
   "train_batch_size": 32,
   "trial_name": null,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e65756886f5d398824a6282f0add6486de2cefc2f866666dd6288860afed6eb
-size 4728
+oid sha256:7c23cc4ac9f96efe3810e18c145117c73b8e78af31fb6d7c567d0849333c8a95
+size 5432