100 food milestone (78 foods -> 100 foods)

[mrdbourke]

I've just created a new model based on EfficientNetB1-Lite (the TensorFlow JS version) that now powers Nutrify and published it to the Nutrify homepage.

In short, Nutrify can now identify 100 total foods (previous was 78).

The performance for each class varies quite a bit. But this will be upgraded in future models.

Next

Now we're up to 100 foods, the next step will be making the current model better (better data) and adding metadata to the website rather than just identifying the class of a food.

In essence, the next few updates will focus on making the current system better rather than adding more foods.

So the current list of foods will likely remain fixed for the next month or so.

App updates (to come)

• Add metadata (such as calories, protein, fat etc) for the first 100 foods (see list below) and make it display when a particular food is identified.
• Add a model cascade for identifying food/not food (Nutrify should recognize a photo of food or not).
• Display the top X (maybe 3?) predictions that can be clicked on and used.

Modelling updates (to come)

• Make training reproducible.
• Make data traceable, e.g. what data is used where?
• Have training and evaluation take place with results (e.g. acc, precision, recall per class)

Updates

The previous Nutrify model could identify the following 78 foods.

Note: If there's more than one food in an image, Nutrify still only identifies one class, this will change with future models.

# Previous 78 Foods Nutrify could classify
['Apple',
  'Artichoke',
  'BBQ sauce',
  'Bacon',
  'Bagel',
  'Banana',
  'Beef',
  'Beer',
  'Blueberries',
  'Bread',
  'Broccoli',
  'Butter',
  'Cabbage',
  'Candy',
  'Cantaloupe',
  'Carrot',
  'Cheese',
  'Chicken',
  'Chicken wings',
  'Cocktail',
  'Coconut',
  'Coffee',
  'Cookie',
  'Corn chips',
  'Cream',
  'Cucumber',
  'Doughnut',
  'Egg',
  'Fish',
  'Fries',
  'Grape',
  'Guacamole',
  'Hamburger',
  'Honey',
  'Ice cream',
  'Lemon',
  'Lime',
  'Lobster',
  'Mango',
  'Milk',
  'Muffin',
  'Mushroom',
  'Olive oil',
  'Olives',
  'Onion',
  'Orange',
  'Orange juice',
  'Pancake',
  'Pasta',
  'Pastry',
  'Pear',
  'Pepper',
  'Pineapple',
  'Pizza',
  'Pomegranate',
  'Popcorn',
  'Potato',
  'Prawns',
  'Pretzel',
  'Pumpkin',
  'Radish',
  'Rice',
  'Salad',
  'Salt',
  'Sandwich',
  'Sausages',
  'Soft drink',
  'Spinach',
  'Squid',
  'Strawberries',
  'Sushi',
  'Tea',
  'Tomato',
  'Tomato sauce',
  'Waffle',
  'Watermelon',
  'Wine',
  'Zucchini']

Note: Squid and Pastry classes were removed from the above list (the Squid data was poor and Pastry as a class is too vague for now), Nutrify focuses on whole foods.

And now with the addition of 24 more foods:

# 24 added foods to Nutrify
['Avocado',
  'Cherry',
  'Dumpling',
  'Egg tart',
  'Eggplant',
  'French fries',
  'Garlic',
  'Grapefruit',
  'Green beans',
  'Green onion',
  'Hamimelon',
  'Kiwi fruit',
  'Lettuce',
  'Meat ball',
  'Noodles',
  'Nuts',
  'Okra',
  'Papaya',
  'Peach',
  'Pie',
  'Plum',
  'Red cabbage',
  'Spring rolls',
  'Steak']

This brings the total to 100 identifiable foods (1000 is the goal for the end of the 2022).

# All current foods Nutrify can identify
{0: 'Apple',
 1: 'Artichoke',
 2: 'Avocado',
 3: 'BBQ sauce',
 4: 'Bacon',
 5: 'Bagel',
 6: 'Banana',
 7: 'Beef',
 8: 'Beer',
 9: 'Blueberries',
 10: 'Bread',
 11: 'Broccoli',
 12: 'Butter',
 13: 'Cabbage',
 14: 'Candy',
 15: 'Cantaloupe',
 16: 'Carrot',
 17: 'Cheese',
 18: 'Cherry',
 19: 'Chicken wings',
 20: 'Cocktail',
 21: 'Coconut',
 22: 'Coffee',
 23: 'Cookie',
 24: 'Corn chips',
 25: 'Cream',
 26: 'Cucumber',
 27: 'Doughnut',
 28: 'Dumpling',
 29: 'Egg',
 30: 'Egg tart',
 31: 'Eggplant',
 32: 'Fish',
 33: 'French fries',
 34: 'Fries',
 35: 'Garlic',
 36: 'Grape',
 37: 'Grapefruit',
 38: 'Green beans',
 39: 'Green onion',
 40: 'Guacamole',
 41: 'Hamburger',
 42: 'Hamimelon',
 43: 'Honey',
 44: 'Ice cream',
 45: 'Kiwi fruit',
 46: 'Lemon',
 47: 'Lettuce',
 48: 'Lime',
 49: 'Lobster',
 50: 'Mango',
 51: 'Meat ball',
 52: 'Milk',
 53: 'Muffin',
 54: 'Mushroom',
 55: 'Noodles',
 56: 'Nuts',
 57: 'Okra',
 58: 'Olive oil',
 59: 'Olives',
 60: 'Onion',
 61: 'Orange',
 62: 'Orange juice',
 63: 'Pancake',
 64: 'Papaya',
 65: 'Pasta',
 66: 'Pastry',
 67: 'Peach',
 68: 'Pear',
 69: 'Pepper',
 70: 'Pie',
 71: 'Pineapple',
 72: 'Pizza',
 73: 'Plum',
 74: 'Pomegranate',
 75: 'Popcorn',
 76: 'Potato',
 77: 'Prawns',
 78: 'Pretzel',
 79: 'Pumpkin',
 80: 'Radish',
 81: 'Red cabbage',
 82: 'Rice',
 83: 'Salad',
 84: 'Salt',
 85: 'Sandwich',
 86: 'Sausages',
 87: 'Soft drink',
 88: 'Spinach',
 89: 'Spring rolls',
 90: 'Steak',
 91: 'Strawberries',
 92: 'Sushi',
 93: 'Tea',
 94: 'Tomato',
 95: 'Tomato sauce',
 96: 'Waffle',
 97: 'Watermelon',
 98: 'Wine',
 99: 'Zucchini'}

More class-based results to come soon.

---

This discussion was created from the release 100 food milestone (78 foods -> 100 foods).

[mrdbourke]

See the results:

Note: This is with Nutrify only predicting one class per image (some images may have more than one item or even many of the same items in the list below).

number	class_name	f1_score	recall	precision	accuracy
1	Artichoke	0.976	1.0	0.952	0.709
79	Pumpkin	0.964	1.0	0.931	0.709
54	Mushroom	0.951	1.0	0.906	0.709
11	Broccoli	0.947	0.964	0.931	0.709
91	Strawberries	0.946	0.964	0.93	0.709
71	Pineapple	0.944	1.0	0.895	0.709
32	Fish	0.933	0.955	0.913	0.709
75	Popcorn	0.919	0.944	0.895	0.709
45	Kiwi fruit	0.918	0.875	0.966	0.709
9	Blueberries	0.917	0.88	0.957	0.709
56	Nuts	0.914	0.941	0.889	0.709
6	Banana	0.913	1.0	0.84	0.709
81	Red cabbage	0.912	1.0	0.839	0.709
72	Pizza	0.903	0.966	0.848	0.709
43	Honey	0.902	0.885	0.92	0.709
21	Coconut	0.898	0.88	0.917	0.709
49	Lobster	0.878	0.9	0.857	0.709
64	Papaya	0.877	0.862	0.893	0.709
38	Green beans	0.873	0.96	0.8	0.709
69	Pepper	0.868	0.943	0.805	0.709
47	Lettuce	0.851	0.909	0.8	0.709
89	Spring rolls	0.848	0.875	0.824	0.709
70	Pie	0.84	0.808	0.875	0.709
8	Beer	0.84	0.84	0.84	0.709
41	Hamburger	0.831	0.844	0.818	0.709
95	Tomato sauce	0.83	0.815	0.846	0.709
98	Wine	0.83	0.917	0.759	0.709
16	Carrot	0.825	0.816	0.833	0.709
35	Garlic	0.816	0.69	1.0	0.709
3	BBQ sauce	0.813	0.929	0.722	0.709
36	Grape	0.81	0.85	0.773	0.709
39	Green onion	0.8	0.9	0.72	0.709
30	Egg tart	0.793	0.767	0.821	0.709
60	Onion	0.791	0.739	0.85	0.709
94	Tomato	0.787	0.857	0.727	0.709
28	Dumpling	0.787	0.8	0.774	0.709
92	Sushi	0.78	0.885	0.697	0.709
97	Watermelon	0.772	0.759	0.786	0.709
19	Chicken wings	0.769	0.652	0.938	0.709
2	Avocado	0.769	0.69	0.87	0.709
74	Pomegranate	0.762	0.774	0.75	0.709
22	Coffee	0.759	0.688	0.846	0.709
13	Cabbage	0.757	0.737	0.778	0.709
29	Egg	0.744	0.696	0.8	0.709
57	Okra	0.735	0.692	0.783	0.709
31	Eggplant	0.722	0.684	0.765	0.709
86	Sausages	0.72	0.783	0.667	0.709
58	Olive oil	0.714	0.69	0.741	0.709
93	Tea	0.71	0.815	0.629	0.709
48	Lime	0.71	0.815	0.629	0.709
33	French fries	0.706	0.643	0.783	0.709
84	Salt	0.704	0.714	0.694	0.709
62	Orange juice	0.698	0.714	0.682	0.709
50	Mango	0.692	0.692	0.692	0.709
23	Cookie	0.69	0.8	0.606	0.709
82	Rice	0.683	0.609	0.778	0.709
14	Candy	0.68	0.708	0.654	0.709
85	Sandwich	0.679	0.72	0.643	0.709
10	Bread	0.678	0.588	0.8	0.709
65	Pasta	0.677	0.913	0.538	0.709
80	Radish	0.667	0.588	0.769	0.709
51	Meat ball	0.667	0.667	0.667	0.709
42	Hamimelon	0.653	0.571	0.762	0.709
67	Peach	0.653	0.593	0.727	0.709
68	Pear	0.651	0.583	0.737	0.709
87	Soft drink	0.65	0.684	0.619	0.709
37	Grapefruit	0.645	0.625	0.667	0.709
59	Olives	0.643	0.667	0.621	0.709
17	Cheese	0.642	0.654	0.63	0.709
26	Cucumber	0.625	0.625	0.625	0.709
61	Orange	0.622	0.609	0.636	0.709
0	Apple	0.615	0.571	0.667	0.709
83	Salad	0.603	0.815	0.478	0.709
78	Pretzel	0.6	0.562	0.643	0.709
77	Prawns	0.6	0.5	0.75	0.709
76	Potato	0.596	0.586	0.607	0.709
96	Waffle	0.596	0.609	0.583	0.709
52	Milk	0.595	0.611	0.579	0.709
18	Cherry	0.593	0.533	0.667	0.709
90	Steak	0.588	0.714	0.5	0.709
20	Cocktail	0.583	0.56	0.609	0.709
4	Bacon	0.576	0.548	0.607	0.709
63	Pancake	0.575	0.724	0.477	0.709
73	Plum	0.552	0.667	0.471	0.709
55	Noodles	0.55	0.44	0.733	0.709
53	Muffin	0.549	0.452	0.7	0.709
34	Fries	0.545	0.667	0.462	0.709
5	Bagel	0.521	0.528	0.514	0.709
44	Ice cream	0.493	0.621	0.409	0.709
88	Spinach	0.488	0.37	0.714	0.709
7	Beef	0.48	0.387	0.632	0.709
99	Zucchini	0.426	0.417	0.435	0.709
27	Doughnut	0.41	0.4	0.421	0.709
12	Butter	0.4	0.333	0.5	0.709
46	Lemon	0.381	0.348	0.421	0.709
40	Guacamole	0.356	0.364	0.348	0.709
24	Corn chips	0.348	0.32	0.381	0.709
66	Pastry	0.314	0.344	0.289	0.709
15	Cantaloupe	0.314	0.296	0.333	0.709
25	Cream	0.0	0.0	0.0	0.709

And here's the F1-scores visually:

[mrdbourke]

Just went through a bunch of data filtering to:

• Remove duplicates.
• Cleanup bottom 10 classes with images that were poor quality or not food (e.g. images in the "Peach" class contained photos of women/the Peach character from Mario Kart) - I removed these. Future datasets of Nutrify will go through a "contains_person" model to see if the image contains a person or not and if so, delete it.
• Doing this meant going from 13,402 total images -> 11,942 images.

I noticed a few issues as well of some images containing two or more of the same class, for example, images of carrots also containing cucumbers. In that case, which do you give the label to? (since as of now, Nutrify is only one class per image)

Future models will be able to detect more than one food class per image.

Anyway, here are the new results (ordered by the difference in F1), the "difference" columns are the difference between the above results and the new model added in 2249533:

|    | class         | f1_score | recall | precision | accuracy | f1_difference | precision_difference | recall_difference |
| -- | ------------- | -------- | ------ | --------- | -------- | ------------- | -------------------- | ----------------- |
| 40 | Guacamole     | 0.733    | 0.611  | 0.917     | 0.73     | 0.377         | 0.569                | 0.247             |
| 24 | Corn chips    | 0.682    | 0.6    | 0.789     | 0.73     | 0.334         | 0.408                | 0.28              |
| 25 | Cream         | 0.286    | 0.25   | 0.333     | 0.73     | 0.286         | 0.333                | 0.25              |
| 27 | Doughnut      | 0.684    | 0.65   | 0.722     | 0.73     | 0.274         | 0.301                | 0.25              |
| 90 | Steak         | 0.847    | 1.0    | 0.735     | 0.73     | 0.259         | 0.235                | 0.286             |
| 44 | Ice cream     | 0.73     | 0.697  | 0.767     | 0.73     | 0.237         | 0.358                | 0.076             |
| 37 | Grapefruit    | 0.828    | 0.857  | 0.8       | 0.73     | 0.183         | 0.133                | 0.232             |
| 82 | Rice          | 0.863    | 0.957  | 0.786     | 0.73     | 0.18          | 0.008                | 0.348             |
| 53 | Muffin        | 0.724    | 0.677  | 0.778     | 0.73     | 0.175         | 0.078                | 0.225             |
| 99 | Zucchini      | 0.595    | 0.524  | 0.688     | 0.73     | 0.169         | 0.253                | 0.107             |
| 74 | Pomegranate   | 0.906    | 1.0    | 0.828     | 0.73     | 0.144         | 0.078                | 0.226             |
| 46 | Lemon         | 0.513    | 0.556  | 0.476     | 0.73     | 0.132         | 0.055                | 0.208             |
| 14 | Candy         | 0.812    | 0.848  | 0.778     | 0.73     | 0.132         | 0.124                | 0.14              |
| 63 | Pancake       | 0.706    | 0.72   | 0.692     | 0.73     | 0.131         | 0.215                | -0.004            |
| 31 | Eggplant      | 0.847    | 0.862  | 0.833     | 0.73     | 0.125         | 0.068                | 0.178             |
| 20 | Cocktail      | 0.7      | 0.778  | 0.636     | 0.73     | 0.117         | 0.027                | 0.218             |
| 7  | Beef          | 0.595    | 0.524  | 0.688     | 0.73     | 0.115         | 0.056                | 0.137             |
| 3  | BBQ sauce     | 0.912    | 0.939  | 0.886     | 0.73     | 0.099         | 0.164                | 0.01              |
| 87 | Soft drink    | 0.743    | 0.619  | 0.929     | 0.73     | 0.093         | 0.31                 | -0.065            |
| 66 | Pastry        | 0.4      | 0.333  | 0.5       | 0.73     | 0.086         | 0.211                | -0.011            |
| 96 | Waffle        | 0.679    | 0.621  | 0.75      | 0.73     | 0.083         | 0.167                | 0.012             |
| 10 | Bread         | 0.76     | 0.76   | 0.76      | 0.73     | 0.082         | -0.04                | 0.172             |
| 85 | Sandwich      | 0.75     | 0.667  | 0.857     | 0.73     | 0.071         | 0.214                | -0.053            |
| 88 | Spinach       | 0.558    | 0.462  | 0.706     | 0.73     | 0.07          | -0.008               | 0.092             |
| 92 | Sushi         | 0.84     | 0.875  | 0.808     | 0.73     | 0.06          | 0.111                | -0.01             |
| 77 | Prawns        | 0.655    | 0.562  | 0.783     | 0.73     | 0.055         | 0.033                | 0.062             |
| 94 | Tomato        | 0.84     | 0.778  | 0.913     | 0.73     | 0.053         | 0.186                | -0.079            |
| 83 | Salad         | 0.654    | 0.708  | 0.607     | 0.73     | 0.051         | 0.129                | -0.107            |
| 4  | Bacon         | 0.625    | 0.526  | 0.769     | 0.73     | 0.049         | 0.162                | -0.022            |
| 42 | Hamimelon     | 0.702    | 0.69   | 0.714     | 0.73     | 0.049         | -0.048               | 0.119             |
| 55 | Noodles       | 0.594    | 0.731  | 0.5       | 0.73     | 0.044         | -0.233               | 0.291             |
| 49 | Lobster       | 0.914    | 0.889  | 0.941     | 0.73     | 0.036         | 0.084                | -0.011            |
| 52 | Milk          | 0.622    | 0.56   | 0.7       | 0.73     | 0.027         | 0.121                | -0.051            |
| 98 | Wine          | 0.857    | 0.889  | 0.828     | 0.73     | 0.027         | 0.069                | -0.028            |
| 72 | Pizza         | 0.929    | 0.929  | 0.929     | 0.73     | 0.026         | 0.081                | -0.037            |
| 35 | Garlic        | 0.84     | 0.778  | 0.913     | 0.73     | 0.024         | -0.087               | 0.088             |
| 41 | Hamburger     | 0.852    | 0.958  | 0.767     | 0.73     | 0.021         | -0.051               | 0.114             |
| 91 | Strawberries  | 0.966    | 1.0    | 0.933     | 0.73     | 0.02          | 0.003                | 0.036             |
| 11 | Broccoli      | 0.964    | 0.931  | 1.0       | 0.73     | 0.017         | 0.069                | -0.033            |
| 22 | Coffee        | 0.774    | 0.857  | 0.706     | 0.73     | 0.015         | -0.14                | 0.169             |
| 23 | Cookie        | 0.704    | 0.594  | 0.864     | 0.73     | 0.014         | 0.258                | -0.206            |
| 6  | Banana        | 0.923    | 0.96   | 0.889     | 0.73     | 0.01          | 0.049                | -0.04             |
| 45 | Kiwi fruit    | 0.926    | 0.862  | 1.0       | 0.73     | 0.008         | 0.034                | -0.013            |
| 89 | Spring rolls  | 0.853    | 1.0    | 0.744     | 0.73     | 0.005         | -0.08                | 0.125             |
| 54 | Mushroom      | 0.954    | 0.969  | 0.939     | 0.73     | 0.003         | 0.033                | -0.031            |
| 32 | Fish          | 0.93     | 0.952  | 0.909     | 0.73     | -0.003        | -0.004               | -0.003            |
| 26 | Cucumber      | 0.622    | 0.667  | 0.583     | 0.73     | -0.003        | -0.042               | 0.042             |
| 81 | Red cabbage   | 0.906    | 0.906  | 0.906     | 0.73     | -0.006        | 0.067                | -0.094            |
| 8  | Beer          | 0.829    | 0.85   | 0.81      | 0.73     | -0.011        | -0.03                | 0.01              |
| 86 | Sausages      | 0.708    | 0.739  | 0.68      | 0.73     | -0.012        | 0.013                | -0.044            |
| 33 | French fries  | 0.693    | 0.812  | 0.605     | 0.73     | -0.013        | -0.178               | 0.169             |
| 17 | Cheese        | 0.629    | 0.579  | 0.688     | 0.73     | -0.013        | 0.058                | -0.075            |
| 84 | Salt          | 0.69     | 0.625  | 0.769     | 0.73     | -0.014        | 0.075                | -0.089            |
| 15 | Cantaloupe    | 0.3      | 0.261  | 0.353     | 0.73     | -0.014        | 0.02                 | -0.035            |
| 38 | Green beans   | 0.857    | 0.882  | 0.833     | 0.73     | -0.016        | 0.033                | -0.078            |
| 78 | Pretzel       | 0.583    | 0.452  | 0.824     | 0.73     | -0.017        | 0.181                | -0.11             |
| 47 | Lettuce       | 0.833    | 0.962  | 0.735     | 0.73     | -0.018        | -0.065               | 0.053             |
| 70 | Pie           | 0.821    | 0.941  | 0.727     | 0.73     | -0.019        | -0.148               | 0.133             |
| 5  | Bagel         | 0.5      | 0.444  | 0.571     | 0.73     | -0.021        | 0.057                | -0.084            |
| 30 | Egg tart      | 0.772    | 0.917  | 0.667     | 0.73     | -0.021        | -0.154               | 0.15              |
| 50 | Mango         | 0.667    | 0.647  | 0.688     | 0.73     | -0.025        | -0.004               | -0.045            |
| 16 | Carrot        | 0.8      | 0.718  | 0.903     | 0.73     | -0.025        | 0.07                 | -0.098            |
| 60 | Onion         | 0.765    | 0.619  | 1.0       | 0.73     | -0.026        | 0.15                 | -0.12             |
| 19 | Chicken wings | 0.739    | 0.68   | 0.81      | 0.73     | -0.03         | -0.128               | 0.028             |
| 76 | Potato        | 0.564    | 0.478  | 0.688     | 0.73     | -0.032        | 0.081                | -0.108            |
| 73 | Plum          | 0.519    | 0.737  | 0.4       | 0.73     | -0.033        | -0.071               | 0.07              |
| 29 | Egg           | 0.71     | 1.0    | 0.55      | 0.73     | -0.034        | -0.25                | 0.304             |
| 69 | Pepper        | 0.833    | 0.741  | 0.952     | 0.73     | -0.035        | 0.147                | -0.202            |
| 12 | Butter        | 0.364    | 0.235  | 0.8       | 0.73     | -0.036        | 0.3                  | -0.098            |
| 13 | Cabbage       | 0.711    | 0.593  | 0.889     | 0.73     | -0.046        | 0.111                | -0.144            |
| 58 | Olive oil     | 0.667    | 0.652  | 0.682     | 0.73     | -0.047        | -0.059               | -0.038            |
| 61 | Orange        | 0.571    | 0.455  | 0.769     | 0.73     | -0.051        | 0.133                | -0.154            |
| 79 | Pumpkin       | 0.909    | 0.893  | 0.926     | 0.73     | -0.055        | -0.005               | -0.107            |
| 71 | Pineapple     | 0.889    | 0.952  | 0.833     | 0.73     | -0.055        | -0.062               | -0.048            |
| 57 | Okra          | 0.68     | 0.654  | 0.708     | 0.73     | -0.055        | -0.075               | -0.038            |
| 93 | Tea           | 0.654    | 0.739  | 0.586     | 0.73     | -0.056        | -0.043               | -0.076            |
| 97 | Watermelon    | 0.714    | 0.833  | 0.625     | 0.73     | -0.058        | -0.161               | 0.074             |
| 80 | Radish        | 0.607    | 0.708  | 0.531     | 0.73     | -0.06         | -0.238               | 0.12              |
| 0  | Apple         | 0.545    | 0.429  | 0.75      | 0.73     | -0.07         | 0.083                | -0.142            |
| 2  | Avocado       | 0.698    | 0.88   | 0.579     | 0.73     | -0.071        | -0.291               | 0.19              |
| 59 | Olives        | 0.564    | 0.524  | 0.611     | 0.73     | -0.079        | -0.01                | -0.143            |
| 36 | Grape         | 0.727    | 0.632  | 0.857     | 0.73     | -0.083        | 0.084                | -0.218            |
| 9  | Blueberries   | 0.829    | 0.81   | 0.85      | 0.73     | -0.088        | -0.107               | -0.07             |
| 43 | Honey         | 0.809    | 0.76   | 0.864     | 0.73     | -0.093        | -0.056               | -0.125            |
| 28 | Dumpling      | 0.687    | 0.92   | 0.548     | 0.73     | -0.1          | -0.226               | 0.12              |
| 34 | Fries         | 0.444    | 0.32   | 0.727     | 0.73     | -0.101        | 0.265                | -0.347            |
| 21 | Coconut       | 0.794    | 0.781  | 0.806     | 0.73     | -0.104        | -0.111               | -0.099            |
| 18 | Cherry        | 0.48     | 0.632  | 0.387     | 0.73     | -0.113        | -0.28                | 0.099             |
| 39 | Green onion   | 0.678    | 0.8    | 0.588     | 0.73     | -0.122        | -0.132               | -0.1              |
| 95 | Tomato sauce  | 0.703    | 0.812  | 0.619     | 0.73     | -0.127        | -0.227               | -0.003            |
| 56 | Nuts          | 0.771    | 0.931  | 0.659     | 0.73     | -0.143        | -0.23                | -0.01             |
| 75 | Popcorn       | 0.757    | 0.875  | 0.667     | 0.73     | -0.162        | -0.228               | -0.069            |
| 48 | Lime          | 0.533    | 0.545  | 0.522     | 0.73     | -0.177        | -0.107               | -0.27             |
| 64 | Papaya        | 0.696    | 0.762  | 0.64      | 0.73     | -0.181        | -0.253               | -0.1              |
| 65 | Pasta         | 0.489    | 0.44   | 0.55      | 0.73     | -0.188        | 0.012                | -0.473            |
| 68 | Pear          | 0.455    | 0.385  | 0.556     | 0.73     | -0.196        | -0.181               | -0.198            |
| 1  | Artichoke     | 0.778    | 0.7    | 0.875     | 0.73     | -0.198        | -0.077               | -0.3              |
| 51 | Meat ball     | 0.453    | 0.6    | 0.364     | 0.73     | -0.214        | -0.303               | -0.067            |
| 62 | Orange juice  | 0.417    | 0.312  | 0.625     | 0.73     | -0.281        | -0.057               | -0.402            |
| 67 | Peach         | 0.167    | 0.143  | 0.2       | 0.73     | -0.486        | -0.527               | -0.45             |

And their mean differences:

Differences:
      F1-score: 0.0023100000000000026
      Precision-score: 0.019199999999999995
      Recall-score: -0.00014999999999999015

Notice how the values aren't very large at all.

This is most likely because many of the duplicates were removed and they would've been artificially increasing the previous results.

So my thinking is saying that the new model should be more robust in the field even though the results aren't that different, they are more aligned with what they should truly be now (since the data has had a good clean).

And the F1 scores have been reshuffled a little to before:

Future updates will automate this process of training and retraining and evaluating models.

Next I'm going to look into data versioning/tracking to keep track of what data was used in what model.

And since the model is performing quite well now, I'll see how to display information on the webpage for whatever food is detected. It may be good to get the top 3 classes showing up too.