Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
|
| 2 |
-
title: Satellite Classification Dashboardemoji: 🛰️colorFrom: bluecolorTo: purplesdk: gradiosdk_version: 5.0.
|
| 3 |
🛰️ Satellite Classification Dashboard
|
| 4 |
A Gradio-based application for classifying satellite images using pre-trained deep learning models. Upload a PNG, JPG, or JPEG image, select one or more models (Custom CNN, MobileNetV2, EfficientNetB0, DenseNet121), and view predictions with confidence scores and visualizations.
|
| 5 |
Quick Start
|
|
@@ -28,13 +28,30 @@ requests==2.32.3
|
|
| 28 |
protobuf==3.20.3
|
| 29 |
|
| 30 |
Troubleshooting
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
Error loading <model>: Unable to load model. Filepath is not an hdf5 file (or h5py is not available) or SavedModel
|
| 32 |
|
| 33 |
Cause: The model file is not a valid HDF5 or SavedModel, or h5py is missing.
|
| 34 |
Fix:
|
| 35 |
Ensure requirements.txt includes h5py==3.10.0.
|
| 36 |
-
Verify the model URL (e.g., https://huggingface.co/Bhavi23/Custom_CNN/resolve/main/best_multimodal_model.keras) is correct
|
| 37 |
-
|
|
|
|
|
|
|
| 38 |
python -c "import tensorflow as tf; model = tf.keras.models.load_model('best_multimodal_model.keras')"
|
| 39 |
|
| 40 |
|
|
@@ -73,15 +90,6 @@ Use the above Dockerfile if needed.
|
|
| 73 |
|
| 74 |
|
| 75 |
|
| 76 |
-
Missing configuration in README
|
| 77 |
-
|
| 78 |
-
Cause: The README.md lacked the YAML front matter.
|
| 79 |
-
Fix: This file includes the correct YAML header. Ensure it is saved as README.md:git add README.md
|
| 80 |
-
git commit -m "Update README.md"
|
| 81 |
-
git push
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
Support
|
| 86 |
|
| 87 |
Issues: Hugging Face Discussions
|
|
|
|
| 1 |
|
| 2 |
+
title: Satellite Classification Dashboardemoji: 🛰️colorFrom: bluecolorTo: purplesdk: gradiosdk_version: "5.0.2"app_file: app.pypinned: false
|
| 3 |
🛰️ Satellite Classification Dashboard
|
| 4 |
A Gradio-based application for classifying satellite images using pre-trained deep learning models. Upload a PNG, JPG, or JPEG image, select one or more models (Custom CNN, MobileNetV2, EfficientNetB0, DenseNet121), and view predictions with confidence scores and visualizations.
|
| 5 |
Quick Start
|
|
|
|
| 28 |
protobuf==3.20.3
|
| 29 |
|
| 30 |
Troubleshooting
|
| 31 |
+
Missing configuration in README
|
| 32 |
+
|
| 33 |
+
Cause: The README.md lacks the required YAML front matter or has incorrect syntax.
|
| 34 |
+
Fix:
|
| 35 |
+
Ensure this README.md is saved as README.md (case-sensitive) in the repository root.
|
| 36 |
+
Verify YAML syntax (2-space indentation, quoted sdk_version).
|
| 37 |
+
Push to repository:git add README.md
|
| 38 |
+
git commit -m "Add YAML front matter to README.md"
|
| 39 |
+
git push
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
Restart the Space in the Settings tab.
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
|
| 46 |
Error loading <model>: Unable to load model. Filepath is not an hdf5 file (or h5py is not available) or SavedModel
|
| 47 |
|
| 48 |
Cause: The model file is not a valid HDF5 or SavedModel, or h5py is missing.
|
| 49 |
Fix:
|
| 50 |
Ensure requirements.txt includes h5py==3.10.0.
|
| 51 |
+
Verify the model URL (e.g., https://huggingface.co/Bhavi23/Custom_CNN/resolve/main/best_multimodal_model.keras) is correct:curl -I https://huggingface.co/Bhavi23/Custom_CNN/resolve/main/best_multimodal_model.keras
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
Download and test the model locally:wget https://huggingface.co/Bhavi23/Custom_CNN/resolve/main/best_multimodal_model.keras
|
| 55 |
python -c "import tensorflow as tf; model = tf.keras.models.load_model('best_multimodal_model.keras')"
|
| 56 |
|
| 57 |
|
|
|
|
| 90 |
|
| 91 |
|
| 92 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
Support
|
| 94 |
|
| 95 |
Issues: Hugging Face Discussions
|