Image Classification
Keras
LiteRT
TF-Keras
Safetensors
English
efficientnetv2-s
efficientnetv2
fgic
transfer-learning
gem-pooling
focal-loss
swa
grad-cam
calibration
temperature-scaling
computer-vision
tensorflow.js
Eval Results (legacy)
Instructions to use 0xgr3y/Arch-Building-Image-Classification with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use 0xgr3y/Arch-Building-Image-Classification with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://0xgr3y/Arch-Building-Image-Classification") - Notebooks
- Google Colab
- Kaggle
Upload V16 model_benchmark.json
Browse files- model_benchmark.json +32 -0
model_benchmark.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_file": "best_phase2_swa.keras",
|
| 3 |
+
"architecture": "DenseNet121+Conv2D(256)+BN+MaxPool2D+GeMPooling+Dense(256)+BN+Dropout(0.4)+Dense(8,softmax)",
|
| 4 |
+
"total_params": 9466953,
|
| 5 |
+
"trainable_params": 7884297,
|
| 6 |
+
"non_trainable_params": 1582656,
|
| 7 |
+
"input_shape": [
|
| 8 |
+
320,
|
| 9 |
+
320,
|
| 10 |
+
3
|
| 11 |
+
],
|
| 12 |
+
"num_classes": 8,
|
| 13 |
+
"model_size_mb": 97.78,
|
| 14 |
+
"saved_model_size_mb": 76.29,
|
| 15 |
+
"tflite_size_kb": 36726.35,
|
| 16 |
+
"tfjs_size_mb": 36.5,
|
| 17 |
+
"inference_ms_keras": 365.3,
|
| 18 |
+
"inference_ms_tflite": 157.9,
|
| 19 |
+
"metrics": {
|
| 20 |
+
"train_accuracy": 0.9988,
|
| 21 |
+
"val_accuracy": 0.9658,
|
| 22 |
+
"test_accuracy": 0.9688,
|
| 23 |
+
"test_loss": 0.4485,
|
| 24 |
+
"tta_accuracy": 0.968,
|
| 25 |
+
"top1_accuracy": 0.9688,
|
| 26 |
+
"top2_accuracy": 0.9859,
|
| 27 |
+
"top3_accuracy": 0.9933,
|
| 28 |
+
"macro_auc": 0.9986,
|
| 29 |
+
"ece": 0.1438,
|
| 30 |
+
"overfitting_gap": 0.03
|
| 31 |
+
}
|
| 32 |
+
}
|