import gradio as gr import os from PIL import Image import numpy as np import pickle import io import sys import torch import subprocess import h5py from sklearn.metrics import confusion_matrix import matplotlib.pyplot as plt # Paths to the predefined images folder RAW_PATH = os.path.join("images", "raw") EMBEDDINGS_PATH = os.path.join("images", "embeddings") # Specific values for percentage of data for training percentage_values = [10, 30, 50, 70, 100] # Custom class to capture print output class PrintCapture(io.StringIO): def __init__(self): super().__init__() self.output = [] def write(self, txt): self.output.append(txt) super().write(txt) def get_output(self): return ''.join(self.output) # Function to load and display predefined images based on user selection def display_predefined_images(percentage_idx): percentage = percentage_values[percentage_idx] raw_image_path = os.path.join(RAW_PATH, f"percentage_{percentage}_complexity_16.png") # Assume complexity 16 for simplicity embeddings_image_path = os.path.join(EMBEDDINGS_PATH, f"percentage_{percentage}_complexity_16.png") raw_image = Image.open(raw_image_path) embeddings_image = Image.open(embeddings_image_path) return raw_image, embeddings_image # Function to create random images for LoS/NLoS classification results def create_random_image(size=(300, 300)): random_image = np.random.rand(*size, 3) * 255 return Image.fromarray(random_image.astype('uint8')) # Function to dynamically load a Python module from a given file path def load_module_from_path(module_name, file_path): spec = importlib.util.spec_from_file_location(module_name, file_path) module = importlib.util.module_from_spec(spec) spec.loader.exec_module(module) return module # Function to split dataset into training and test sets based on user selection def split_dataset(channels, labels, percentage_idx): percentage = percentage_values[percentage_idx] / 100 num_samples = channels.shape[0] train_size = int(num_samples * percentage) indices = np.arange(num_samples) np.random.shuffle(indices) train_idx, test_idx = indices[:train_size], indices[train_size:] train_data, test_data = channels[train_idx], channels[test_idx] train_labels, test_labels = labels[train_idx], labels[test_idx] return train_data, test_data, train_labels, test_labels # Function to generate confusion matrix plot def plot_confusion_matrix(y_true, y_pred, title): cm = confusion_matrix(y_true, y_pred) plt.figure(figsize=(4, 4)) plt.imshow(cm, cmap='Blues') plt.title(title) plt.xlabel('Predicted') plt.ylabel('Actual') plt.colorbar() plt.xticks([0, 1], labels=[0, 1]) plt.yticks([0, 1], labels=[0, 1]) plt.tight_layout() plt.savefig(f"{title}.png") return Image.open(f"{title}.png") def identical_train_test_split(output_emb, output_raw, labels, percentage): N = output_emb.shape[0] indices = torch.randperm(N) split_index = int(N * percentage) train_indices = indices[:split_index] test_indices = indices[split_index:] train_emb = output_emb[train_indices] test_emb = output_emb[test_indices] train_raw = output_raw[train_indices] test_raw = output_raw[test_indices] train_labels = labels[train_indices] test_labels = labels[test_indices] return train_emb, test_emb, train_raw, test_raw, train_labels, test_labels # Function to classify test data based on distance to class centroids def classify_based_on_distance(train_data, train_labels, test_data): centroid_0 = train_data[train_labels == 0].mean(dim=0) centroid_1 = train_data[train_labels == 1].mean(dim=0) predictions = [] for test_point in test_data: dist_0 = torch.norm(test_point - centroid_0) dist_1 = torch.norm(test_point - centroid_1) predictions.append(0 if dist_0 < dist_1 else 1) return torch.tensor(predictions) # Store the original working directory when the app starts original_dir = os.getcwd() def process_hdf5_file(uploaded_file, percentage_idx): capture = PrintCapture() sys.stdout = capture try: model_repo_url = "https://huggingface.co/sadjadalikhani/LWM" model_repo_dir = "./LWM" if not os.path.exists(model_repo_dir): subprocess.run(["git", "clone", model_repo_url, model_repo_dir], check=True) repo_work_dir = os.path.join(original_dir, model_repo_dir) if os.path.exists(repo_work_dir): os.chdir(repo_work_dir) lwm_model_path = os.path.join(os.getcwd(), 'lwm_model.py') input_preprocess_path = os.path.join(os.getcwd(), 'input_preprocess.py') inference_path = os.path.join(os.getcwd(), 'inference.py') lwm_model = load_module_from_path("lwm_model", lwm_model_path) input_preprocess = load_module_from_path("input_preprocess", input_preprocess_path) inference = load_module_from_path("inference", inference_path) device = 'cpu' model = lwm_model.LWM.from_pretrained(device=device) with h5py.File(uploaded_file.name, 'r') as f: channels = np.array(f['channels']) labels = np.array(f['labels']) preprocessed_chs = input_preprocess.tokenizer(manual_data=channels) output_emb = inference.lwm_inference(preprocessed_chs, 'channel_emb', model) output_raw = inference.create_raw_dataset(preprocessed_chs, device) train_data_emb, test_data_emb, train_data_raw, test_data_raw, train_labels, test_labels = identical_train_test_split( output_emb.view(len(output_emb),-1), output_raw.view(len(output_raw),-1), labels, percentage_idx) pred_raw = classify_based_on_distance(train_data_raw, train_labels, test_data_raw) pred_emb = classify_based_on_distance(train_data_emb, train_labels, test_data_emb) raw_cm_image = plot_confusion_matrix(test_labels, pred_raw, title="Confusion Matrix (Raw Channels)") emb_cm_image = plot_confusion_matrix(test_labels, pred_emb, title="Confusion Matrix (Embeddings)") return raw_cm_image, emb_cm_image, capture.get_output() except Exception as e: return str(e), str(e), capture.get_output() finally: os.chdir(original_dir) sys.stdout = sys.__stdout__ def los_nlos_classification(file, percentage_idx): if file is not None: return process_hdf5_file(file, percentage_idx) else: return display_predefined_images(percentage_idx), None # Define the Gradio interface with a compact, minimal layout with gr.Blocks(css=""" .slider-container { text-align: center; margin-bottom: 20px; } .image-row { justify-content: center; margin-top: 10px; } .output-box { max-width: 600px; margin: 0 auto; } """) as demo: # Contact Section gr.Markdown("""
Wireless Model Email
""") # Tabs for Beam Prediction and LoS/NLoS Classification with gr.Tab("Beam Prediction Task"): gr.Markdown("### Beam Prediction Task") with gr.Row(): with gr.Column(elem_id="slider-container"): percentage_slider_bp = gr.Slider(minimum=0, maximum=4, step=1, value=0, label="Training Data (%)") with gr.Row(elem_id="image-row"): raw_img_bp = gr.Image(label="Raw Channels", type="pil", width=300, height=300) embeddings_img_bp = gr.Image(label="Embeddings", type="pil", width=300, height=300) percentage_slider_bp.change(fn=display_predefined_images, inputs=[percentage_slider_bp], outputs=[raw_img_bp, embeddings_img_bp]) with gr.Tab("LoS/NLoS Classification Task"): gr.Markdown("### LoS/NLoS Classification Task") file_input = gr.File(label="Upload HDF5 Dataset", file_types=[".h5"]) with gr.Row(): with gr.Column(elem_id="slider-container"): percentage_slider_los = gr.Slider(minimum=0, maximum=4, step=1, value=0, label="Training Data (%)") with gr.Row(elem_id="image-row"): raw_img_los = gr.Image(label="Raw Channels", type="pil", width=300, height=300) embeddings_img_los = gr.Image(label="Embeddings", type="pil", width=300, height=300) output_textbox = gr.Textbox(label="Console Output", lines=8, elem_classes="output-box") file_input.change(fn=los_nlos_classification, inputs=[file_input, percentage_slider_los], outputs=[raw_img_los, embeddings_img_los, output_textbox]) percentage_slider_los.change(fn=los_nlos_classification, inputs=[file_input, percentage_slider_los], outputs=[raw_img_los, embeddings_img_los, output_textbox]) # Launch the app if __name__ == "__main__": demo.launch()