first commit

.gitattributes

@@ -32,3 +32,4 @@ 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
+effnetb2_food101_dict.pth filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,60 @@
+
+import gradio as gr
+import os
+import torch
+
+from model import create_effnet_b2_instance
+from timeit import default_timer as timer
+from typing import Tuple, Dict
+
+# Setup class names
+with open("class_names.txt", "r") as f: 
+  class_names = [food_name.strip() for food_name in  f.readlines()]
+
+# Create Food101 compatible EffNetB2 instance
+effnet_transforms,effnetb2_food_101 = create_effnet_b2_instance(num_classes = len(class_names))
+
+# Load the saved model's state_dict()
+effnetb2_food_101.load_state_dict(torch.load("effnetb2_food101_dict.pth",map_location = torch.device("cpu")))
+
+
+def predict(img, model = effnetb2_food_101, transforms = effnet_transforms) -> Tuple[Dict,float]:
+  # start a timer
+  start_timer = timer()
+  # transform the image to be used by the model
+  prepreocpressed_image = transforms(img).unsqueeze(0)
+  # turn off regularization and parameters
+  model.eval()
+  with torch.inference_mode():
+    prediction = model(prepreocpressed_image)
+    probabilities = torch.softmax(prediction,dim = 1)
+  prob_dict = {class_names[i]: float(probabilities[0][i]) for i in range(len(class_names))}
+  # calculate the time
+  end_timer = timer()
+  total_time = end_timer - start_timer
+  return prob_dict,total_time
+
+
+# create the gradio app
+title = "FoodVision Big Classifier"
+description = "An EfficientNetB2 feature extractor trained on the Food101 Dataset to classify across 101 possible classes of food."
+article = "Model created using pytorch"
+
+example_list = [["examples/" + example] for example in os.listdir("examples")]
+
+# Create Gradio interface 
+demo = gr.Interface(
+    fn=predict,
+    inputs=gr.Image(type="pil"),
+    outputs=[
+        gr.Label(num_top_classes=5, label="Predictions"),
+        gr.Number(label="Prediction time (s)"),
+    ],
+    examples=example_list,
+    title=title,
+    description=description,
+    article=article,
+)
+
+# Launch the app!
+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_dict.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb4f75ce27688eb0ffb20ce848895ecb9794a2d2b776bf3453b8ba1e949a5e14
+size 31833069

model.py

@@ -0,0 +1,21 @@
+import torch
+import torchvision
+
+from torch import nn
+
+def create_effnet_b2_instance(num_classes = 3):
+  # fetch the model's pretrained weights
+  effnetb2_pretrained_weights = torchvision.models.EfficientNet_B2_Weights.DEFAULT
+  # fetch the preprocessing transforms
+  effnetb2_transforms = effnetb2_pretrained_weights.transforms()
+  # get the model and load the pretrained weighits
+  effnetb2 = torchvision.models.efficientnet_b2(weights=effnetb2_pretrained_weights)
+  # freeze the feature extractor
+  for param in effnetb2.parameters():
+    param.requires_grad = False
+  # fix the output 
+  effnetb2.classifier = nn.Sequential(
+      nn.Dropout(p = 0.3,inplace=True),
+      nn.Linear(in_features = 1408,out_features = num_classes)
+  )
+  return effnetb2_transforms,effnetb2

requirements.txt

@@ -0,0 +1,3 @@
+torch==2.0.0
+torchvision==0.15.1
+gradio==3.28.1