🍻 cheers

README.md

@@ -0,0 +1,85 @@
+---
+license: apache-2.0
+base_model: google/vit-base-patch16-224-in21k
+tags:
+- image-classification
+- generated_from_trainer
+datasets:
+- generator
+metrics:
+- accuracy
+- f1
+model-index:
+- name: stool-condition-classification
+  results:
+  - task:
+      name: Image Classification
+      type: image-classification
+    dataset:
+      name: stool-image
+      type: generator
+      config: default
+      split: train
+      args: default
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8580527752502275
+    - name: F1
+      type: f1
+      value: 0.8173302107728336
+---
+
+<!-- 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. -->
+
+# stool-condition-classification
+
+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 stool-image dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3669
+- Auroc: 0.9121
+- Accuracy: 0.8581
+- Sensitivity: 0.7756
+- Specificty: 0.9153
+- F1: 0.8173
+
+## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 1
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Auroc  | Accuracy | Sensitivity | Specificty | F1     |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|:-----------:|:----------:|:------:|
+| 0.4071        | 0.98  | 100  | 0.4415          | 0.8876 | 0.8179   | 0.6629      | 0.9552     | 0.7738 |
+
+
+### Framework versions
+
+- Transformers 4.36.1
+- Pytorch 2.0.1
+- Datasets 2.15.0
+- Tokenizers 0.15.0

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.8580527752502275,
+    "eval_auroc": 0.912140044512926,
+    "eval_f1": 0.8173302107728336,
+    "eval_loss": 0.366916298866272,
+    "eval_runtime": 68.1665,
+    "eval_samples_per_second": 16.122,
+    "eval_sensitivity": 0.7755555555555556,
+    "eval_specificty": 0.9152542372881356,
+    "eval_steps_per_second": 2.024,
+    "train_loss": 0.4936492688515607,
+    "train_runtime": 82.0698,
+    "train_samples_per_second": 19.8,
+    "train_steps_per_second": 1.243
+}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "./vit-base-UCEIS-strong-aug/checkpoint-2000",
+  "_name_or_path": "google/vit-base-patch16-224-in21k",
   "architectures": [
     "ViTForImageClassification"
   ],
@@ -9,8 +9,8 @@
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
-    "0": "Normal",
-    "1": "Abnormal"
+    "0": "0",
+    "1": "1"
   },
   "image_size": 224,
   "initializer_range": 0.02,

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.8580527752502275,
+    "eval_auroc": 0.912140044512926,
+    "eval_f1": 0.8173302107728336,
+    "eval_loss": 0.366916298866272,
+    "eval_runtime": 68.1665,
+    "eval_samples_per_second": 16.122,
+    "eval_sensitivity": 0.7755555555555556,
+    "eval_specificty": 0.9152542372881356,
+    "eval_steps_per_second": 2.024
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7eff3b0a5fe7e6eb0b633bd5ee7de7b8970ed276a00429d75c84e7e149f0b9cc
+oid sha256:67e325baa94230e86d326557616c9574e14bc969ba434562d3a52bc97b6bbe6a
 size 343223968

runs/Jan05_17-12-10_DESKTOP-BDBS5RV/events.out.tfevents.1704442331.DESKTOP-BDBS5RV

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2c397fcf5e38bf12ed1ead4460427d233c83b4f80fc3400763b76d9140ff17a
+size 6732

runs/Jan05_17-12-10_DESKTOP-BDBS5RV/events.out.tfevents.1704442481.DESKTOP-BDBS5RV

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a81b132f789c2ad0ed1bbcf06af19bfedcb7cab2578529e47b56f2e50bd6d3ad
+size 338

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 1.0,
+    "train_loss": 0.4936492688515607,
+    "train_runtime": 82.0698,
+    "train_samples_per_second": 19.8,
+    "train_steps_per_second": 1.243
+}

trainer_state.json

@@ -0,0 +1,103 @@
+{
+  "best_metric": 0.8875705740958131,
+  "best_model_checkpoint": "./stool-condition-classification\\checkpoint-100",
+  "epoch": 1.0,
+  "eval_steps": 100,
+  "global_step": 102,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0001803921568627451,
+      "loss": 0.5897,
+      "step": 10
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00016078431372549022,
+      "loss": 0.479,
+      "step": 20
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0001411764705882353,
+      "loss": 0.5314,
+      "step": 30
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00012156862745098039,
+      "loss": 0.5547,
+      "step": 40
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00010196078431372549,
+      "loss": 0.4872,
+      "step": 50
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 8.23529411764706e-05,
+      "loss": 0.505,
+      "step": 60
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 6.274509803921569e-05,
+      "loss": 0.5257,
+      "step": 70
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.313725490196079e-05,
+      "loss": 0.4404,
+      "step": 80
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.3529411764705884e-05,
+      "loss": 0.4018,
+      "step": 90
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 3.92156862745098e-06,
+      "loss": 0.4071,
+      "step": 100
+    },
+    {
+      "epoch": 0.98,
+      "eval_accuracy": 0.8179419525065963,
+      "eval_auroc": 0.8875705740958131,
+      "eval_f1": 0.7737704918032786,
+      "eval_loss": 0.44151929020881653,
+      "eval_runtime": 19.7299,
+      "eval_samples_per_second": 19.209,
+      "eval_sensitivity": 0.6629213483146067,
+      "eval_specificty": 0.9552238805970149,
+      "eval_steps_per_second": 2.433,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "step": 102,
+      "total_flos": 1.25924483123712e+17,
+      "train_loss": 0.4936492688515607,
+      "train_runtime": 82.0698,
+      "train_samples_per_second": 19.8,
+      "train_steps_per_second": 1.243
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 102,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 1.25924483123712e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd4157da5d8d9f13bebf7070514c0e1f61d4b4e845588f68818e3014effdf48d
+size 4283