Upernetconvnext Finetuned Segments Food Oct 14

🚀 upernetconvnext-finetuned-segments-food-oct-14

This model is a fine - tuned version of [openmmlab/upernet - convnext - small](https://huggingface.co/openmmlab/upernet - convnext - small) on the EduardoPacheco/FoodSeg103 dataset. It can be used for image segmentation tasks in the food domain, providing evaluation results on key metrics such as loss, mean Iou, and various accuracy values.

🚀 Quick Start

This section provides a brief introduction to the model and its evaluation results. For more detailed usage, please refer to the following sections.

✨ Features

• Fine - tuned Model: Based on the [openmmlab/upernet - convnext - small](https://huggingface.co/openmmlab/upernet - convnext - small) model, it is fine - tuned on the EduardoPacheco/FoodSeg103 dataset.
• Comprehensive Evaluation Metrics: Provides multiple evaluation metrics on the evaluation set, including loss, mean Iou, mean accuracy, overall accuracy, and accuracy for different food categories.

📚 Documentation

Model Performance

This model is a fine - tuned version of [openmmlab/upernet - convnext - small](https://huggingface.co/openmmlab/upernet - convnext - small) on the EduardoPacheco/FoodSeg103 dataset. It achieves the following results on the evaluation set:

• Loss: 2.7723
• Mean Iou: 0.0671
• Mean Accuracy: 0.1792
• Overall Accuracy: 0.1867
• Accuracy Background: nan
• Accuracy Candy: nan
• Accuracy Egg tart: nan
• Accuracy French fries: 0.1073
• Accuracy Chocolate: nan
• Accuracy Biscuit: nan
• Accuracy Popcorn: nan
• Accuracy Pudding: nan
• Accuracy Ice cream: 0.1838
• Accuracy Cheese butter: nan
• Accuracy Cake: 0.0
• Accuracy Wine: nan
• Accuracy Milkshake: nan
• Accuracy Coffee: nan
• Accuracy Juice: 0.0
• Accuracy Milk: nan
• Accuracy Tea: nan
• Accuracy Almond: nan
• Accuracy Red beans: nan
• Accuracy Cashew: nan
• Accuracy Dried cranberries: nan
• Accuracy Soy: nan
• Accuracy Walnut: nan
• Accuracy Peanut: nan
• Accuracy Egg: 0.0
• Accuracy Apple: 0.0
• Accuracy Date: nan
• Accuracy Apricot: nan
• Accuracy Avocado: nan
• Accuracy Banana: 0.0
• Accuracy Strawberry: 0.1034
• Accuracy Cherry: nan
• Accuracy Blueberry: nan
• Accuracy Raspberry: nan
• Accuracy Mango: nan
• Accuracy Olives: nan
• Accuracy Peach: nan
• Accuracy Lemon: 0.2544
• Accuracy Pear: nan
• Accuracy Fig: nan
• Accuracy Pineapple: 0.0
• Accuracy Grape: nan
• Accuracy Kiwi: nan
• Accuracy Melon: nan
• Accuracy Orange: 0.5418
• Accuracy Watermelon: nan
• Accuracy Steak: 0.0
• Accuracy Pork: 0.0
• Accuracy Chicken duck: 0.3444
• Accuracy Sausage: 0.0
• Accuracy Fried meat: nan
• Accuracy Lamb: nan
• Accuracy Sauce: 0.2308
• Accuracy Crab: nan
• Accuracy Fish: 0.0
• Accuracy Shellfish: nan
• Accuracy Shrimp: nan
• Accuracy Soup: nan
• Accuracy Bread: 0.9865
• Accuracy Corn: nan
• Accuracy Hamburg: nan
• Accuracy Pizza: 0.0
• Accuracy hanamaki baozi: nan
• Accuracy Wonton dumplings: nan
• Accuracy Pasta: nan
• Accuracy Noodles: nan
• Accuracy Rice: 0.4019
• Accuracy Pie: 0.0
• Accuracy Tofu: nan
• Accuracy Eggplant: nan
• Accuracy Potato: 0.1621
• Accuracy Garlic: nan
• Accuracy Cauliflower: 0.0
• Accuracy Tomato: 0.5529
• Accuracy Kelp: nan
• Accuracy Seaweed: nan
• Accuracy Spring onion: nan
• Accuracy Rape: nan
• Accuracy Ginger: nan
• Accuracy Okra: nan
• Accuracy Lettuce: nan
• Accuracy Pumpkin: nan
• Accuracy Cucumber: nan
• Accuracy White radish: nan
• Accuracy Carrot: 0.9551
• Accuracy Asparagus: 0.0012
• Accuracy Bamboo shoots: nan
• Accuracy Broccoli: 0.9085
• Accuracy Celery stick: nan
• Accuracy Cilantro mint: 0.0
• Accuracy Snow peas: nan
• Accuracy cabbage: nan
• Accuracy Bean sprouts: nan
• Accuracy Onion: 0.0
• Accuracy Pepper: 0.0
• Accuracy Green beans: nan
• Accuracy French beans: 0.0001
• Accuracy King oyster mushroom: nan
• Accuracy Shiitake: nan
• Accuracy Enoki mushroom: nan
• Accuracy Oyster mushroom: nan
• Accuracy White button mushroom: nan
• Accuracy Salad: nan
• Accuracy Other ingredients: 0.0
• Iou Background: 0.0
• Iou Candy: nan
• Iou Egg tart: nan
• Iou French fries: 0.0801
• Iou Chocolate: nan
• Iou Biscuit: nan
• Iou Popcorn: nan
• Iou Pudding: nan
• Iou Ice cream: 0.0147
• Iou Cheese butter: nan
• Iou Cake: 0.0
• Iou Wine: nan
• Iou Milkshake: nan
• Iou Coffee: nan
• Iou Juice: 0.0
• Iou Milk: nan
• Iou Tea: nan
• Iou Almond: nan
• Iou Red beans: nan
• Iou Cashew: nan
• Iou Dried cranberries: nan
• Iou Soy: nan
• Iou Walnut: nan
• Iou Peanut: nan
• Iou Egg: 0.0
• Iou Apple: 0.0
• Iou Date: nan
• Iou Apricot: nan
• Iou Avocado: nan
• Iou Banana: 0.0
• Iou Strawberry: 0.1005
• Iou Cherry: nan
• Iou Blueberry: nan
• Iou Raspberry: nan
• Iou Mango: nan
• Iou Olives: nan
• Iou Peach: nan
• Iou Lemon: 0.2328
• Iou Pear: nan
• Iou Fig: nan
• Iou Pineapple: 0.0
• Iou Grape: nan
• Iou Kiwi: nan
• Iou Melon: nan
• Iou Orange: 0.3032
• Iou Watermelon: nan
• Iou Steak: 0.0
• Iou Pork: 0.0
• Iou Chicken duck: 0.0980
• Iou Sausage: 0.0
• Iou Fried meat: nan
• Iou Lamb: nan
• Iou Sauce: 0.2163
• Iou Crab: nan
• Iou Fish: 0.0
• Iou Shellfish: nan
• Iou Shrimp: nan
• Iou Soup: nan
• Iou Bread: 0.2062
• Iou Corn: nan
• Iou Hamburg: nan
• Iou Pizza: 0.0
• Iou hanamaki baozi: nan
• Iou Wonton dumplings: nan
• Iou Pasta: nan
• Iou Noodles: nan
• Iou Rice: 0.1747
• Iou Pie: 0.0
• Iou Tofu: nan
• Iou Eggplant: nan
• Iou Potato: 0.1605
• Iou Garlic: nan
• Iou Cauliflower: 0.0
• Iou Tomato: 0.0681
• Iou Kelp: nan
• Iou Seaweed: nan
• Iou Spring onion: nan
• Iou Rape: nan
• Iou Ginger: nan
• Iou Okra: nan
• Iou Lettuce: 0.0
• Iou Pumpkin: nan
• Iou Cucumber: nan
• Iou White radish: nan
• Iou Carrot: 0.4083
• Iou Asparagus: 0.0011
• Iou Bamboo shoots: nan
• Iou Broccoli: 0.2155
• Iou Celery stick: nan
• Iou Cilantro mint: 0.0
• Iou Snow peas: nan
• Iou cabbage: nan
• Iou Bean sprouts: nan
• Iou Onion: 0.0
• Iou Pepper: 0.0
• Iou Green beans: nan
• Iou French beans: 0.0001
• Iou King oyster mushroom: nan
• Iou Shiitake: nan
• Iou Enoki mushroom: nan
• Iou Oyster mushroom: nan
• Iou White button mushroom: nan
• Iou Salad: nan
• Iou Other ingredients: 0.0

Training Procedure

Training Hyperparameters

The following hyperparameters were used during training:

• learning_rate: 6e - 05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
• lr_scheduler_type: linear
• num_epochs: 5

Training Results

| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Background | Accuracy Candy | Accuracy Egg tart | Accuracy French fries | Accuracy Chocolate | Accuracy Biscuit | Accuracy Popcorn | Accuracy Pudding | Accuracy Ice cream | Accuracy Cheese butter | Accuracy Cake | Accuracy Wine | Accuracy Milkshake | Accuracy Coffee | Accuracy Juice | Accuracy Milk | Accuracy Tea | Accuracy Almond | Accuracy Red beans | Accuracy Cashew | Accuracy Dried cranberries | Accuracy Soy | Accuracy Walnut | Accuracy Peanut | Accuracy Egg | Accuracy Apple | Accuracy Date | Accuracy Apricot | Accuracy Avocado | Accuracy Banana | Accuracy Strawberry | Accuracy Cherry | Accuracy Blueberry | Accuracy Raspberry | Accuracy Mango | Accuracy Olives | Accuracy Peach | Accuracy Lemon | Accuracy Pear | Accuracy Fig | Accuracy Pineapple | Accuracy Grape | Accuracy Kiwi | Accuracy Melon | Accuracy Orange | Accuracy Watermelon | Accuracy Steak | Accuracy Pork | Accuracy Chicken duck | Accuracy Sausage | Accuracy Fried meat | Accuracy Lamb | Accuracy Sauce | Accuracy Crab | Accuracy Fish | Accuracy Shellfish | Accuracy Shrimp | Accuracy Soup | Accuracy Bread | Accuracy Corn | Accuracy Hamburg | Accuracy Pizza | Accuracy hanamaki baozi | Accuracy Wonton dumplings | Accuracy Pasta | Accuracy Noodles | Accuracy Rice | Accuracy Pie | Accuracy Tofu | Accuracy Eggplant | Accuracy Potato | Accuracy Garlic | Accuracy Cauliflower | Accuracy Tomato | Accuracy Kelp | Accuracy Seaweed | Accuracy Spring onion | Accuracy Rape | Accuracy Ginger | Accuracy Okra | Accuracy Lettuce | Accuracy Pumpkin | Accuracy Cucumber | Accuracy White radish | Accuracy Carrot | Accuracy Asparagus | Accuracy Bamboo shoots | Accuracy Broccoli | Accuracy Celery stick | Accuracy Cilantro mint | Accuracy Snow peas | Accuracy cabbage | Accuracy Bean sprouts | Accuracy Onion | Accuracy Pepper | Accuracy Green beans | Accuracy French beans | Accuracy King oyster mushroom | Accuracy Shiitake | Accuracy Enoki mushroom | Accuracy Oyster mushroom | Accuracy White button mushroom | Accuracy Salad | Accuracy Other ingredients | Iou Background | Iou Candy | Iou Egg tart | Iou French fries | Iou Chocolate | Iou Biscuit | Iou Popcorn | Iou Pudding | Iou Ice cream | Iou Cheese butter | Iou Cake | Iou Wine | Iou Milkshake | Iou Coffee | Iou Juice | Iou Milk | Iou Tea | Iou Almond | Iou Red beans | Iou Cashew | Iou Dried cranberries | Iou Soy | Iou Walnut | Iou Peanut | Iou Egg | Iou Apple | Iou Date | Iou Apricot | Iou Avocado | Iou Banana | Iou Strawberry | Iou Cherry | Iou Blueberry | Iou Raspberry | Iou Mango | Iou Olives | Iou Peach | Iou Lemon | Iou Pear | Iou Fig | Iou Pineapple | Iou Grape | Iou Kiwi | Iou Melon | Iou Orange | Iou Watermelon | Iou Steak | Iou Pork | Iou Chicken duck | Iou Sausage | Iou Fried meat | Iou Lamb | Iou Sauce | Iou Crab | Iou Fish | Iou Shellfish | Iou Shrimp | Iou Soup | Iou Bread | Iou Corn | Iou Hamburg | Iou Pizza | Iou hanamaki baozi | Iou Wonton dumplings | Iou Pasta | Iou Noodles | Iou Rice | Iou Pie | Iou Tofu | Iou Eggplant | Iou Potato | Iou Garlic | Iou Cauliflower | Iou Tomato | Iou Kelp | Iou Seaweed | Iou Spring onion | Iou Rape | Iou Ginger | Iou Okra | Iou Lettuce | Iou Pumpkin | Iou Cucumber | Iou White radish | Iou Carrot | Iou Asparagus | Iou Bamboo shoots | Iou Broccoli | Iou Celery stick | Iou Cilantro mint | Iou Snow peas | Iou cabbage | Iou Bean sprouts | Iou Onion | Iou Pepper | Iou Green beans | Iou French beans | Iou King oyster mushroom | Iou Shiitake | Iou Enoki mushroom | Iou Oyster mushroom | Iou White button mushroom | Iou Salad | Iou Other ingredients | |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:-------------------:|:--------------:|:-----------------:|:---------------------:|:------------------:|:----------------:|:----------------:|:----------------:|:------------------:|:----------------------:|:-------------:|:-------------:|:------------------:|:---------------:|:--------------:|:-------------:|:------------:|:---------------:|:------------------:|:---------------:|:--------------------------:|:------------:|:---------------:|:---------------:|:------------:|:--------------:|:-------------:|:----------------:|:----------------:|:---------------:|:-------------------:|:---------------:|:------------------:|:------------------:|:--------------:|:---------------:|:--------------:|:--------------:|:-------------:|:------------:|:------------------:|:--------------:|:-------------:|:--------------:|:---------------:|:-------------------:|:--------------:|:-------------:|:---------------------:|:----------------:|:-------------------:|:-------------:|:--------------:|:-------------:|:-------------:|:------------------:|:---------------:|:-------------:|:--------------:|:-------------:|:----------------:|:--------------:|:------------------------:|:-------------------------:|:--------------:|:----------------:|:-------------:|:------------:|:-------------:|:-----------------:|:---------------:|:---------------:|:--------------------:|:---------------:|:-------------:|:----------------:|:---------------------:|:-------------:|:---------------:|:-------------:|:----------------:|:----------------:|:-----------------:|:---------------------:|:---------------:|:------------------:|:----------------------:|:-----------------:|:---------------------:|:----------------------:|:------------------:|:-----------------:|:---------------------:|:--------------:|:---------------:|:--------------------:|:---------------------:|:-----------------------------:|:-----------------:|:-----------------------:|:------------------------:|:------------------------------:|:--------------:|:--------------------------:|:--------------:|:---------:|:------------:|:----------------:|:-------------:|:-----------:|:-----------:|:-----------:|:-------------:|:-----------------:|:--------:|:--------:|:-------------:|:----------:|:---------:|:--------:|:-------:|:----------:|:-------------:|:----------:|:---------------------:|:-------:|:----------:|:----------:|:-------:|:---------:|:--------:|:-----------:|:-----------:|:----------:|:--------------:|:----------:|:-------------:|:-------------:|:---------:|:----------:|:---------:|:---------:|:--------:|:-------:|:-------------:|:---------:|:--------:|:---------:|:----------:|:--------------:|:---------:|:--------:|:----------------:|:-----------:|:--------------:|:--------:|:---------:|:--------:|:--------:|:-------------:|:----------:|:--------:|:---------:|:--------:|:-----------:|:---------:|:-------------------:|:--------------------:|:---------:|:-----------:|:--------:|:-------:|:--------:|:------------:|:----------:|:----------:|:---------------:|:------:|

📄 License

This model is released under the MIT license.