vit-base-kidney-stone-4-Michel_Daudon_-w256_1k_v1-_SEC

README.md

@@ -0,0 +1,136 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: google/vit-base-patch16-224-in21k
+tags:
+- generated_from_trainer
+datasets:
+- imagefolder
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+model-index:
+- name: vit-base-kidney-stone-4-Michel_Daudon_-w256_1k_v1-_SEC
+  results:
+  - task:
+      name: Image Classification
+      type: image-classification
+    dataset:
+      name: imagefolder
+      type: imagefolder
+      config: default
+      split: test
+      args: default
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.9241666666666667
+    - name: Precision
+      type: precision
+      value: 0.9296490647145426
+    - name: Recall
+      type: recall
+      value: 0.9241666666666667
+    - name: F1
+      type: f1
+      value: 0.9247640186674816
+---
+
+<!-- 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. -->
+
+# vit-base-kidney-stone-4-Michel_Daudon_-w256_1k_v1-_SEC
+
+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: 0.2879
+- Accuracy: 0.9242
+- Precision: 0.9296
+- Recall: 0.9242
+- F1: 0.9248
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 16
+- eval_batch_size: 8
+- seed: 42
+- 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: 15
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch   | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
+|:-------------:|:-------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
+| 0.2837        | 0.3333  | 100  | 0.5470          | 0.8333   | 0.8693    | 0.8333 | 0.8325 |
+| 0.1498        | 0.6667  | 200  | 0.4199          | 0.8658   | 0.8833    | 0.8658 | 0.8647 |
+| 0.0979        | 1.0     | 300  | 0.4712          | 0.8783   | 0.9015    | 0.8783 | 0.8799 |
+| 0.009         | 1.3333  | 400  | 0.4957          | 0.885    | 0.8933    | 0.885  | 0.8819 |
+| 0.0226        | 1.6667  | 500  | 0.2879          | 0.9242   | 0.9296    | 0.9242 | 0.9248 |
+| 0.0722        | 2.0     | 600  | 0.4449          | 0.8875   | 0.8906    | 0.8875 | 0.8869 |
+| 0.0043        | 2.3333  | 700  | 0.3699          | 0.9125   | 0.9221    | 0.9125 | 0.9104 |
+| 0.0678        | 2.6667  | 800  | 0.6081          | 0.8792   | 0.8872    | 0.8792 | 0.8760 |
+| 0.1178        | 3.0     | 900  | 0.5728          | 0.8767   | 0.8748    | 0.8767 | 0.8744 |
+| 0.0297        | 3.3333  | 1000 | 0.3977          | 0.9258   | 0.9267    | 0.9258 | 0.9257 |
+| 0.0813        | 3.6667  | 1100 | 1.1116          | 0.8283   | 0.8462    | 0.8283 | 0.8153 |
+| 0.0336        | 4.0     | 1200 | 0.9246          | 0.82     | 0.8215    | 0.82   | 0.8155 |
+| 0.0291        | 4.3333  | 1300 | 0.6674          | 0.8808   | 0.8980    | 0.8808 | 0.8819 |
+| 0.1018        | 4.6667  | 1400 | 0.7256          | 0.8667   | 0.8760    | 0.8667 | 0.8641 |
+| 0.0739        | 5.0     | 1500 | 0.4149          | 0.8908   | 0.9082    | 0.8908 | 0.8913 |
+| 0.0017        | 5.3333  | 1600 | 0.3553          | 0.9208   | 0.9291    | 0.9208 | 0.9219 |
+| 0.0011        | 5.6667  | 1700 | 0.3934          | 0.915    | 0.9188    | 0.915  | 0.9157 |
+| 0.0056        | 6.0     | 1800 | 0.8180          | 0.8725   | 0.9139    | 0.8725 | 0.8733 |
+| 0.001         | 6.3333  | 1900 | 0.3790          | 0.9225   | 0.9216    | 0.9225 | 0.9217 |
+| 0.0055        | 6.6667  | 2000 | 0.6404          | 0.88     | 0.8910    | 0.88   | 0.8765 |
+| 0.0007        | 7.0     | 2100 | 0.5133          | 0.9017   | 0.9073    | 0.9017 | 0.9023 |
+| 0.0009        | 7.3333  | 2200 | 0.4628          | 0.92     | 0.9296    | 0.92   | 0.9189 |
+| 0.0007        | 7.6667  | 2300 | 0.8405          | 0.8617   | 0.8744    | 0.8617 | 0.8581 |
+| 0.1144        | 8.0     | 2400 | 1.0096          | 0.8592   | 0.8954    | 0.8592 | 0.8567 |
+| 0.0007        | 8.3333  | 2500 | 0.6318          | 0.8983   | 0.9113    | 0.8983 | 0.8977 |
+| 0.0005        | 8.6667  | 2600 | 0.4929          | 0.9075   | 0.9135    | 0.9075 | 0.9076 |
+| 0.0013        | 9.0     | 2700 | 0.6148          | 0.8883   | 0.8955    | 0.8883 | 0.8866 |
+| 0.001         | 9.3333  | 2800 | 1.0043          | 0.8392   | 0.8538    | 0.8392 | 0.8355 |
+| 0.0004        | 9.6667  | 2900 | 0.9713          | 0.8425   | 0.8556    | 0.8425 | 0.8390 |
+| 0.0004        | 10.0    | 3000 | 0.9737          | 0.865    | 0.8977    | 0.865  | 0.8634 |
+| 0.0004        | 10.3333 | 3100 | 0.8766          | 0.8683   | 0.8835    | 0.8683 | 0.8673 |
+| 0.0004        | 10.6667 | 3200 | 0.8620          | 0.8683   | 0.8808    | 0.8683 | 0.8672 |
+| 0.0003        | 11.0    | 3300 | 0.8669          | 0.8675   | 0.8803    | 0.8675 | 0.8665 |
+| 0.0003        | 11.3333 | 3400 | 0.8712          | 0.8667   | 0.8789    | 0.8667 | 0.8656 |
+| 0.0003        | 11.6667 | 3500 | 0.8732          | 0.8675   | 0.8797    | 0.8675 | 0.8665 |
+| 0.0003        | 12.0    | 3600 | 0.8754          | 0.8658   | 0.8782    | 0.8658 | 0.8648 |
+| 0.0003        | 12.3333 | 3700 | 0.8775          | 0.8658   | 0.8782    | 0.8658 | 0.8648 |
+| 0.0003        | 12.6667 | 3800 | 0.8797          | 0.865    | 0.8772    | 0.865  | 0.8640 |
+| 0.0003        | 13.0    | 3900 | 0.8816          | 0.865    | 0.8772    | 0.865  | 0.8640 |
+| 0.0003        | 13.3333 | 4000 | 0.8835          | 0.865    | 0.8772    | 0.865  | 0.8640 |
+| 0.0003        | 13.6667 | 4100 | 0.8844          | 0.865    | 0.8769    | 0.865  | 0.8639 |
+| 0.0003        | 14.0    | 4200 | 0.8852          | 0.8658   | 0.8775    | 0.8658 | 0.8648 |
+| 0.0002        | 14.3333 | 4300 | 0.8859          | 0.8667   | 0.8780    | 0.8667 | 0.8655 |
+| 0.0002        | 14.6667 | 4400 | 0.8865          | 0.8675   | 0.8786    | 0.8675 | 0.8664 |
+| 0.0002        | 15.0    | 4500 | 0.8868          | 0.8675   | 0.8786    | 0.8675 | 0.8664 |
+
+
+### Framework versions
+
+- Transformers 4.48.2
+- Pytorch 2.6.0+cu126
+- Datasets 3.1.0
+- Tokenizers 0.21.0

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 0.9241666666666667,
+    "eval_f1": 0.9247640186674816,
+    "eval_loss": 0.287933349609375,
+    "eval_precision": 0.9296490647145426,
+    "eval_recall": 0.9241666666666667,
+    "eval_runtime": 9.6386,
+    "eval_samples_per_second": 124.499,
+    "eval_steps_per_second": 15.562,
+    "total_flos": 5.57962327867392e+18,
+    "train_loss": 0.042441848201884166,
+    "train_runtime": 1173.35,
+    "train_samples_per_second": 61.363,
+    "train_steps_per_second": 3.835
+}

config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "google/vit-base-patch16-224-in21k",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "SEC-Subtype_IVa",
+    "1": "SEC-Subtype_IVa2",
+    "2": "SEC-Subtype_IVc",
+    "3": "SEC-Subtype_IVd",
+    "4": "SEC-Subtype_Ia",
+    "5": "SEC-Subtype_Va"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "SEC-Subtype_IVa": "0",
+    "SEC-Subtype_IVa2": "1",
+    "SEC-Subtype_IVc": "2",
+    "SEC-Subtype_IVd": "3",
+    "SEC-Subtype_Ia": "4",
+    "SEC-Subtype_Va": "5"
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.2"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f42e15369e9eb805ebe840e2ce1c5c3888f2755cda6711929fec33575c0e3f0
+size 343236280

preprocessor_config.json

@@ -0,0 +1,23 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTFeatureExtractor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

test_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 0.9241666666666667,
+    "eval_f1": 0.9247640186674816,
+    "eval_loss": 0.287933349609375,
+    "eval_precision": 0.9296490647145426,
+    "eval_recall": 0.9241666666666667,
+    "eval_runtime": 9.6386,
+    "eval_samples_per_second": 124.499,
+    "eval_steps_per_second": 15.562
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 15.0,
+    "total_flos": 5.57962327867392e+18,
+    "train_loss": 0.042441848201884166,
+    "train_runtime": 1173.35,
+    "train_samples_per_second": 61.363,
+    "train_steps_per_second": 3.835
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50c7a14f25c6bd8435b76ecbfb50b62c02e13fc401f0b92a77d4504b9572968f
+size 5432