first commit

09_pretrained_effnetb2_feature_extractor_pizza_steak_sushi_20_percent.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:232b9150391e812a5c6ecba4348eb35c649f8c8baa1a390ecea7f8c6f5def965
+size 31307450

app.py

@@ -0,0 +1,65 @@
+import gradio as gr
+import os
+import torch
+
+from model import create_effnetb2_model
+from timeit import default_timer as timer
+from typing import Tuple, TypedDict
+
+# Setup class names
+class_names = ["pizza", "steak", "sushi"]
+
+# Create EffNetB2 model instance and transform
+effnetb2, effnetb2_transforms = create_effnetb2_model(num_classes=len(class_names))
+
+# Load model weights
+effnetb2.load_state_dict(
+    torch.load(
+        os.path.join("models", "09_pretrained_effnetb2_feature_extractor_pizza_steak_sushi_20_percent.pth"),
+        map_location=torch.device("cpu")
+        )
+    )
+
+# Predict function
+def predict(img) -> Tuple[Dict, float]:
+  # Start a timer
+  start_time = timer()
+
+  # Transform the input image for use with EffNetB2
+  img = effnetb2_transforms(img).unsqueeze(0)
+
+  # put model into eval mode, make prediction
+  effnetb2.eval()
+  with torch.inference_mode():
+    pred_probs = torch.softmax(effnetb2(img), dim=-1)
+
+  # Create a prediction label and predcition probability
+  pred_labels_and_probs = {class_names[i]: float(pred_probs[0][i]) for i in range(class_names)}
+
+  # Calculate pred time and pred dict
+  pred_time = round(timer() - start_time, 5)
+
+  return pred_labels_and_probs, pred_time
+
+# Gradio app
+# Create title, description and article strings
+title = "FoodVision Mini 🍕🥩🍣"
+description = "An EfficientNetB2 feature extractor computer vision model to classify images of food as pizza, steak or sushi."
+article = "Created at [09. PyTorch Model Deployment](https://www.learnpytorch.io/09_pytorch_model_deployment/)."
+
+# Create an example list
+example_list = [["examples/" + example] for example in os.listdir("examples")]
+
+# Create the Gradio demo
+demo = gr.Interface(fn=predict,
+                    inputs=gr.inputs.Image(type="pil"),
+                    outputs=[gr.outputs.Label(num_top_classes=3, label="Predictions"),
+                             gr.outputs.Number(label="Prediction time (s)")],
+                    examples=example_list,
+                    title=title,
+                    description=description,
+                    article=article)
+
+# Launch the demo!
+demo.launch(debug=False,
+            share=True)

model.py

@@ -0,0 +1,31 @@
+import torch, torchvision
+from torch import nn
+
+def create_effnetb2_model(num_classes: int = 3,
+                          seed:int=42):
+    """Creates an EfficientNetB2 feature extractor model and transforms.
+
+    Args:
+        num_classes (int, optional): Number of output neurons in the output layer. Defaults to 3
+        seed (int, optional): Random seed value. Defaults to 42.
+
+    Returns:
+        torchvision.models.efficientnet_b2: EffNetB2 feature extractor model
+
+    """
+    # 1. Setup pretrained EffNMetB2 weights
+    effnetb2_weights = torchvision.models.EfficientNet_B2_Weights.DEFAULT
+    effnetb2_transform = effnetb2_weights.transforms()
+    # 2. Setup pretrained model
+    effnetb2 = torchvision.models.efficientnet_b2(weights=effnetb2_weights)
+    # 3. Freeze the base layers
+    for param in effnetb2.parameters():
+        param.requires_grad = False
+
+    # 4. Change the classsifier to 3 classes
+    torch.manual_seed(seed)
+    effnetb2.classifier = nn.Sequential(
+        nn.Dropout(p=0.3, inplace=True),
+        nn.Linear(in_features=1408, out_features=num_classes))
+    
+    return effnetb2, effnetb2_transform

requirements.txt

@@ -0,0 +1,3 @@
+torch==1.12.0
+torchvision==0.13.0
+gradio==3.1.4