nfn

app.py

@@ -31,10 +31,59 @@
 
 
 
+# import gradio as gr
+
+# def greet(name):
+#     return "Hello " + name + "!!"
+
+# iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+# iface.launch()
+
+
+
+"""Step 7: Deploying on Hugging Face Model Hub"""
 import gradio as gr
+import torch
+
+# Load the saved model from .pkl file
+model = torch.load("model.pth")
+
+# !pip install transformers
+
+# Define the preprocessing function
+preprocess = transforms.Compose([
+    transforms.Resize(224),
+    transforms.CenterCrop(224),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+])
+
+# Define the predict function
+def predict_food_image(image):
+    # Preprocess the input image
+    img = Image.fromarray(image.astype('uint8'), 'RGB')
+    img = preprocess(img)
+    img = img.unsqueeze(0)
+
+    # Run the model prediction
+    with torch.no_grad():
+        outputs = model(img)
+
+    # Post-process the outputs
+    predicted_class = torch.argmax(outputs).item()
+
+    # Return the predicted category
+    return predicted_class
 
-def greet(name):
-    return "Hello " + name + "!!"
+# Define the Gradio interface
+gr_interface = gr.Interface(
+    fn=predict_food_image,
+    inputs="image",
+    outputs="text",
+    title="Food Image Recognizer",
+    description="Upload an image of food and get the predicted category.",
+    allow_flagging=False
+)
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
-iface.launch()
+# Launch the interface
+gr_interface.launch(share=True)

food_model.pkl → model.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7b69eda976f7f38dfd369c21687943b70da604bb1c79e094983e3ebf2e341900
-size 87509175
+oid sha256:13ab178215e7cf49cd23173527cbaaeb10c7e7c160087b9a07c0e164cdfefb53
+size 262001181

recognizer_image.py

@@ -0,0 +1,152 @@
+# -*- coding: utf-8 -*-
+"""recognizer_image.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/1gRejGB4T2LDDXPQNTKBXJUahCXt2vA_G
+
+Step 1: Install the required libraries
+"""
+
+# !pip install torch torchvision
+# !pip install fastai==2.5.2
+# !pip install fastbook==0.0.16
+
+"""Step 2: Import the necessary libraries"""
+
+from fastai.vision.all import *
+from fastbook import *
+
+# """Step 3: Mount Google Drive"""
+
+# from google.colab import drive
+
+# # Mount Google Drive
+# drive.mount('/content/drive')
+
+"""Step 4: Set the data path"""
+
+data_path = '/content/drive/MyDrive/Master_Course_Data Science/Capstone_Project/foods_images'
+
+"""Step 5: Define data augmentation and data block"""
+
+data_augmentation = aug_transforms(mult=1.0, flip_vert=True, max_zoom=1.1, max_rotate=20.0, max_lighting=0.4)
+
+try:
+    food_dblock = DataBlock(
+        blocks=(ImageBlock, CategoryBlock),
+        get_items=get_image_files,
+        splitter=RandomSplitter(valid_pct=0.2, seed=42),
+        get_y=parent_label,
+        item_tfms=Resize(224),
+        batch_tfms=[*data_augmentation, Normalize.from_stats(*imagenet_stats)]
+    )
+except Exception as e:
+    print(f"Error in defining the data block: {e}")
+    sys.exit(1)
+
+"""Step 6: Create dataloaders"""
+
+try:
+    dls = food_dblock.dataloaders(data_path)
+except Exception as e:
+    print(f"Error in creating dataloaders: {e}")
+    sys.exit(1)
+
+"""Step 7: Show a batch of images"""
+
+try:
+    dls.show_batch()
+except Exception as e:
+    print(f"Error in showing batch of images: {e}")
+    sys.exit(1)
+
+"""Step 8: Train the model"""
+
+learn = cnn_learner(dls, resnet34, metrics=accuracy)
+try:
+    learn.fine_tune(4)
+except Exception as e:
+    print(f"Error in fine-tuning the model: {e}")
+    sys.exit(1)
+
+learn.save("/content/drive/MyDrive/Master_Course_Data Science/Capstone_Project/model")
+
+# !pip install gradio
+
+
+import gradio as gr
+
+from torchvision.transforms import transforms
+from PIL import Image
+
+# Define the prediction function
+def predict_food_image(img):
+    pred = learn.predict(img)[0]
+    return str(pred)
+
+# Create the Gradio interface
+gr_interface = gr.Interface(
+    fn=predict_food_image,
+    inputs="image",
+    outputs="text",
+    title="Food Image Recognizer",
+    description="Upload an image of food and get the predicted category.",
+    allow_flagging=False
+)
+
+# Launch the interface
+gr_interface.launch(share=True)
+
+"""Step 7: Deploying on Hugging Face Model Hub"""
+
+import torch
+
+# Load the saved model from .pkl file
+model = torch.load("/content/drive/MyDrive/Master_Course_Data Science/Capstone_Project/model.pth")
+
+# !pip install transformers
+
+# Define the preprocessing function
+preprocess = transforms.Compose([
+    transforms.Resize(224),
+    transforms.CenterCrop(224),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+])
+
+# Define the predict function
+def predict_food_image(image):
+    # Preprocess the input image
+    img = Image.fromarray(image.astype('uint8'), 'RGB')
+    img = preprocess(img)
+    img = img.unsqueeze(0)
+
+    # Run the model prediction
+    with torch.no_grad():
+        outputs = model(img)
+
+    # Post-process the outputs
+    predicted_class = torch.argmax(outputs).item()
+
+    # Return the predicted category
+    return predicted_class
+
+# Define the Gradio interface
+gr_interface = gr.Interface(
+    fn=predict_food_image,
+    inputs="image",
+    outputs="text",
+    title="Food Image Recognizer",
+    description="Upload an image of food and get the predicted category.",
+    allow_flagging=False
+)
+
+# Launch the interface
+gr_interface.launch(share=True)
+
+"""Notebook to Python Script Export"""
+
+!pip uninstall nbdev
+!pip install nbdev==1.1.13