Upload 6 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/example.jpg filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,58 @@
+# Imports
+import torch
+import gradio as gr
+from typing import Tuple, Dict
+from timeit import default_timer as timer
+import model
+import os
+
+# Class names
+with open("class_names.txt", "r") as f: 
+  class_names = [food.strip() for food in f.readlines()]
+
+# Create instance of model
+effnetb2, effnetb2_transforms = model.create_effnetb2_model(num_classes=len(class_names))
+# Load Weights
+effnetb2.load_state_dict(state_dict=torch.load("effnetb2_food101_20pct.pth",
+                                               map_location=torch.device("cpu") # hard-coded load to cpu
+                                               ))
+
+# Predict function
+def predict(img) -> Tuple[Dict, float]:
+  # Start timer
+  start = timer()
+  # Transform input image for use
+  img = effnetb2_transforms(img).unsqueeze(dim=0)
+  # Put model in eval mode
+  effnetb2.eval()
+  with torch.inference_mode():
+    logits = effnetb2(img)
+    pred_probs = torch.softmax(logits, dim=1).squeeze()
+    prediction = logits.argmax(dim=1).item()
+    prediction_label = class_names[prediction]
+  end = timer()
+  pred_dict = {class_names[i]: pred_probs[i].item() for i in range(len(class_names))}
+  delta_time = round(end-start, 4)
+  return pred_dict, delta_time
+
+# Gradio app
+
+# Create example list from within this file
+example_list = [ ["examples/" + example] for example in os.listdir("examples")]
+
+title = "FoodVision Big"
+description = "EfficientNetB2 feature extractor CV model to classify images of 101 types of food from the Food101 dataset."
+article = "Created for PyTorch ZTM course"
+
+demo = gr.Interface(fn=predict, # maps inputs to outputs
+                    inputs=gr.Image(type="pil"),
+                    outputs=[gr.Label(num_top_classes=5, label="Predictions"),  # for the prediction dictionary
+                             gr.Number(label="Prediction time (s)")
+                             ],
+                    examples=example_list,
+                    title=title,
+                    description=description,
+                    article=article,
+                    )
+
+demo.launch()

class_names.txt

@@ -0,0 +1,101 @@
+apple_pie
+baby_back_ribs
+baklava
+beef_carpaccio
+beef_tartare
+beet_salad
+beignets
+bibimbap
+bread_pudding
+breakfast_burrito
+bruschetta
+caesar_salad
+cannoli
+caprese_salad
+carrot_cake
+ceviche
+cheese_plate
+cheesecake
+chicken_curry
+chicken_quesadilla
+chicken_wings
+chocolate_cake
+chocolate_mousse
+churros
+clam_chowder
+club_sandwich
+crab_cakes
+creme_brulee
+croque_madame
+cup_cakes
+deviled_eggs
+donuts
+dumplings
+edamame
+eggs_benedict
+escargots
+falafel
+filet_mignon
+fish_and_chips
+foie_gras
+french_fries
+french_onion_soup
+french_toast
+fried_calamari
+fried_rice
+frozen_yogurt
+garlic_bread
+gnocchi
+greek_salad
+grilled_cheese_sandwich
+grilled_salmon
+guacamole
+gyoza
+hamburger
+hot_and_sour_soup
+hot_dog
+huevos_rancheros
+hummus
+ice_cream
+lasagna
+lobster_bisque
+lobster_roll_sandwich
+macaroni_and_cheese
+macarons
+miso_soup
+mussels
+nachos
+omelette
+onion_rings
+oysters
+pad_thai
+paella
+pancakes
+panna_cotta
+peking_duck
+pho
+pizza
+pork_chop
+poutine
+prime_rib
+pulled_pork_sandwich
+ramen
+ravioli
+red_velvet_cake
+risotto
+samosa
+sashimi
+scallops
+seaweed_salad
+shrimp_and_grits
+spaghetti_bolognese
+spaghetti_carbonara
+spring_rolls
+steak
+strawberry_shortcake
+sushi
+tacos
+takoyaki
+tiramisu
+tuna_tartare
+waffles

effnetb2_food101_20pct.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4481d999993e8ca8e47c0ddb485176713ddf132822d2c213a743fbf6695f4d92
+size 31833771

model.py

@@ -0,0 +1,21 @@
+
+import torchvision
+import torch
+from torch import nn
+
+"""Script to load and create and instance of effnetb2 """
+
+def create_effnetb2_model(num_classes=101):
+  weights=torchvision.models.EfficientNet_B2_Weights.DEFAULT
+  transforms=weights.transforms()
+  model=torchvision.models.efficientnet_b2(weights=weights)
+
+  for params in model.parameters():
+    params.requires_grad=False
+
+  model.classifier = nn.Sequential(
+      nn.Dropout(p=0.3, inplace=True),
+      nn.Linear(in_features=1408, out_features=num_classes)
+  )
+
+  return model, transforms

requirements.txt

@@ -0,0 +1,3 @@
+torch==2.0.1
+torchvision==0.15.2
+gradio==3.44.4