Ammar2k's picture
first commit
93267bd
import gradio as gr
import os
import torch
from model import create_vit
from timeit import default_timer as timer
from typing import Tuple, Dict
class_names = ["buildings", "forest", "glacier", "mountain", "sea", "street"]
vit_model, vit_transforms = create_vit(num_classes=len(class_names),
seed=42)
vit_model.load_state_dict(
torch.load(
f="pretrained_vit_feature_extractor_scene_recognition.pth",
map_location=torch.device("cpu")
)
)
def predict(img):
start_timer = timer()
img = vit_transforms(img).unsqueeze(0)
vit_model.eval()
with torch.inference_mode():
pred_prob = torch.softmax(vit_model(img), dim=1)
pred_labels_and_probs = {class_names[i]: float(pred_prob[0][i]) for i in range(len(class_names))}
pred_time = round(timer() - start_timer, 5)
return pred_labels_and_probs, pred_time
title = "Scene Recognition: Intel Image Classification"
description = "A ViT feature extractor Computer Vision model to classify images of scenes from 1 out of 6 classes."
article = "Access project repository at [GitHub](https://github.com/Ammar2k/intel_image_classification)"
example_list = [["examples/" + example] for example in os.listdir("examples")]
demo = gr.Interface(fn=predict,
inputs=gr.Image(type="pil"),
outputs=[gr.Label(num_top_classes=6, label="Predictions"),
gr.Number(label="Prediction time(s)")],
examples=example_list,
title=title,
description=description,
article=article
)
demo.launch()