| import gradio as gr |
| import datasets |
| import torch |
| from transformers import AutoFeatureExtractor, AutoModelForImageClassification |
| from transformers import ViTImageProcessor, ViTForImageClassification |
|
|
| dataset = datasets.load_dataset("beans") |
| image_processor = ViTImageProcessor.from_pretrained("google/vit-base-patch16-224") |
|
|
| extractor = AutoFeatureExtractor.from_pretrained("saved_model_files", from_pt=True) |
| model = AutoModelForImageClassification.from_pretrained("saved_model_files") |
| model.config.to_json_file("./saved_model_files/config.json") |
|
|
|
|
|
|
| labels = dataset['train'].features['labels'].names |
|
|
| def classify(im): |
| features = image_processor(im, return_tensors='pt') |
| logits = model(features["pixel_values"])[-1] |
| probability = torch.nn.functional.softmax(logits, dim=-1) |
| probs = probability[0].detach().numpy() |
| confidences = {label: float(probs[i]) for i, label in enumerate(labels)} |
| return confidences |
|
|
| interface = gr.Interface(fn=classify, inputs="image", outputs="label") |
|
|
| interface.launch(debug=True) |
|
|
