app.py

import gradio as gr
import torch
from PIL import Image
from torchvision import transforms
import numpy as np
import random
import torch.nn as nn
from torch.utils.data import DataLoader, Dataset
from torchvision.models.resnet import ResNet50_Weights
from typing import Type, Any, Callable, Union, List, Optional
from torch import Tensor
from huggingface_hub import hf_hub_download

username = "leandrumartin"
model_repo = "assignment2model"
model_path = hf_hub_download(repo_id=f"{username}/{model_repo}", filename="clothing1m.pth")

CATEGORY_NAMES = ['T-Shirt', 'Shirt', 'Knitwear', 'Chiffon', 'Sweater', 'Hoodie', 'Windbreaker', 'Jacket', 'Downcoat', 'Suit', 'Shawl', 'Dress', 'Vest', 'Underwear']

def conv3x3(in_planes: int, out_planes: int, stride: int = 1, groups: int = 1, dilation: int = 1) -> nn.Conv2d:
    """3x3 convolution with padding"""
    return nn.Conv2d(
        in_planes,
        out_planes,
        kernel_size=3,
        stride=stride,
        padding=dilation,
        groups=groups,
        bias=False,
        dilation=dilation,
    )


def conv1x1(in_planes: int, out_planes: int, stride: int = 1) -> nn.Conv2d:
    """1x1 convolution"""
    return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)


class BasicBlock(nn.Module):
    expansion: int = 1

    def __init__(
            self,
            inplanes: int,
            planes: int,
            stride: int = 1,
            downsample: Optional[nn.Module] = None,
            groups: int = 1,
            base_width: int = 64,
            dilation: int = 1,
            norm_layer: Optional[Callable[..., nn.Module]] = None,
    ) -> None:
        super().__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        if groups != 1 or base_width != 64:
            raise ValueError("BasicBlock only supports groups=1 and base_width=64")
        if dilation > 1:
            raise NotImplementedError("Dilation > 1 not supported in BasicBlock")
        # Both self.conv1 and self.downsample layers downsample the input when stride != 1
        self.conv1 = conv3x3(inplanes, planes, stride)
        self.bn1 = norm_layer(planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes)
        self.bn2 = norm_layer(planes)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x: Tensor) -> Tensor:
        identity = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out


class Bottleneck(nn.Module):
    # Bottleneck in torchvision places the stride for downsampling at 3x3 convolution(self.conv2)
    # while original implementation places the stride at the first 1x1 convolution(self.conv1)
    # according to "Deep residual learning for image recognition"https://arxiv.org/abs/1512.03385.
    # This variant is also known as ResNet V1.5 and improves accuracy according to
    # https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch.

    expansion: int = 4

    def __init__(
            self,
            inplanes: int,
            planes: int,
            stride: int = 1,
            downsample: Optional[nn.Module] = None,
            groups: int = 1,
            base_width: int = 64,
            dilation: int = 1,
            norm_layer: Optional[Callable[..., nn.Module]] = None,
    ) -> None:
        super().__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        width = int(planes * (base_width / 64.0)) * groups
        # Both self.conv2 and self.downsample layers downsample the input when stride != 1
        self.conv1 = conv1x1(inplanes, width)
        self.bn1 = norm_layer(width)
        self.conv2 = conv3x3(width, width, stride, groups, dilation)
        self.bn2 = norm_layer(width)
        self.conv3 = conv1x1(width, planes * self.expansion)
        self.bn3 = norm_layer(planes * self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x: Tensor) -> Tensor:
        identity = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)
        out = self.relu(out)

        out = self.conv3(out)
        out = self.bn3(out)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out


class ResNet(nn.Module):
    def __init__(
            self,
            block: Type[Union[BasicBlock, Bottleneck]],
            layers: List[int],
            num_classes: int = 1000,
            show: bool = False,
            zero_init_residual: bool = False,
            groups: int = 1,
            width_per_group: int = 64,
            replace_stride_with_dilation: Optional[List[bool]] = None,
            norm_layer: Optional[Callable[..., nn.Module]] = None,
    ) -> None:
        super().__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        self._norm_layer = norm_layer

        self.show = show
        self.inplanes = 64
        self.dilation = 1
        if replace_stride_with_dilation is None:
            # each element in the tuple indicates if we should replace
            # the 2x2 stride with a dilated convolution instead
            replace_stride_with_dilation = [False, False, False]
        if len(replace_stride_with_dilation) != 3:
            raise ValueError(
                "replace_stride_with_dilation should be None "
                f"or a 3-element tuple, got {replace_stride_with_dilation}"
            )
        self.groups = groups
        self.base_width = width_per_group
        self.conv1 = nn.Conv2d(3, self.inplanes, kernel_size=7, stride=2, padding=3, bias=False)
        self.bn1 = norm_layer(self.inplanes)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, 64, layers[0])
        self.layer2 = self._make_layer(block, 128, layers[1], stride=2, dilate=replace_stride_with_dilation[0])
        self.layer3 = self._make_layer(block, 256, layers[2], stride=2, dilate=replace_stride_with_dilation[1])
        self.layer4 = self._make_layer(block, 512, layers[3], stride=2, dilate=replace_stride_with_dilation[2])
        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
        self.fc = nn.Linear(512 * block.expansion, num_classes)
        # self.fc1 = nn.Linear(512 * block.expansion, 512)
        # self.lu = nn.LeakyReLU(0.1, inplace=True)
        # self.fc2 = nn.Linear(512, num_classes)

        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight, mode="fan_out", nonlinearity="relu")
            elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)

        # Zero-initialize the last BN in each residual branch,
        # so that the residual branch starts with zeros, and each residual block behaves like an identity.
        # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677
        if zero_init_residual:
            for m in self.modules():
                if isinstance(m, Bottleneck) and m.bn3.weight is not None:
                    nn.init.constant_(m.bn3.weight, 0)  # type: ignore[arg-type]
                elif isinstance(m, BasicBlock) and m.bn2.weight is not None:
                    nn.init.constant_(m.bn2.weight, 0)  # type: ignore[arg-type]

    def _make_layer(
            self,
            block: Type[Union[BasicBlock, Bottleneck]],
            planes: int,
            blocks: int,
            stride: int = 1,
            dilate: bool = False,
    ) -> nn.Sequential:
        norm_layer = self._norm_layer
        downsample = None
        previous_dilation = self.dilation
        if dilate:
            self.dilation *= stride
            stride = 1
        if stride != 1 or self.inplanes != planes * block.expansion:
            downsample = nn.Sequential(
                conv1x1(self.inplanes, planes * block.expansion, stride),
                norm_layer(planes * block.expansion),
            )

        layers = []
        layers.append(
            block(
                self.inplanes, planes, stride, downsample, self.groups, self.base_width, previous_dilation, norm_layer
            )
        )
        self.inplanes = planes * block.expansion
        for _ in range(1, blocks):
            layers.append(
                block(
                    self.inplanes,
                    planes,
                    groups=self.groups,
                    base_width=self.base_width,
                    dilation=self.dilation,
                    norm_layer=norm_layer,
                )
            )

        return nn.Sequential(*layers)

    def _forward_impl(self, x: Tensor) -> Tensor:
        # See note [TorchScript super()]
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        x = self.maxpool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        x = self.avgpool(x)
        x = torch.flatten(x, 1)
        out = self.fc(x)
        # x = self.lu(self.fc1(x))
        # out = self.fc2(x)
        if self.show:
            return out, x
        else:
            return out

    def forward(self, x: Tensor) -> Tensor:
        return self._forward_impl(x)


def _resnet(
        block: Type[Union[BasicBlock, Bottleneck]],
        layers: List[int],
        num_classes,
        show,
        **kwargs: Any,
) -> ResNet:

    model = ResNet(block, layers, num_classes, show, **kwargs)

    return model




def resnet50(num_classes, show=False, **kwargs: Any) -> ResNet:
    """ResNet-50 from `Deep Residual Learning for Image Recognition <https://arxiv.org/pdf/1512.03385.pdf>`__.

    .. note::
       The bottleneck of TorchVision places the stride for downsampling to the second 3x3
       convolution while the original paper places it to the first 1x1 convolution.
       This variant improves the accuracy and is known as `ResNet V1.5
       <https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch>`_.

    Args:
        weights (:class:`~torchvision.models.ResNet50_Weights`, optional): The
            pretrained weights to use. See
            :class:`~torchvision.models.ResNet50_Weights` below for
            more details, and possible values. By default, no pre-trained
            weights are used.
        progress (bool, optional): If True, displays a progress bar of the
            download to stderr. Default is True.
        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``
            base class. Please refer to the `source code
            <https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py>`_
            for more details about this class.

    .. autoclass:: torchvision.models.ResNet50_Weights
        :members:
    """

    return _resnet(Bottleneck, [3, 4, 6, 3], num_classes, show, **kwargs)



class Clothing1M(Dataset):
    def __init__(self, image, train=True, transform=None, target_transform=None, augment=False, mode='noisy'):
        self.image = image
        self.transform = transform
        self.target_transform = target_transform
        self.augment = augment
        self.train = False
        self.mode = mode

        self.data = [self.image]

    def __getitem__(self, index):
        img, target = self.data[index], 0

        # to return a PIL Image
        # img_origin = Image.open(img).convert('RGB')
        img_origin = Image.fromarray(img).convert('RGB')

        if self.transform is not None:
            img = self.transform(img_origin)
            if self.augment:
                img1 = self.transform(img_origin)

        if self.target_transform is not None:
            target = self.target_transform(target)

        return img, 0

    def __len__(self):
        return len(self.data)

def set_seed(seed):
    torch.manual_seed(seed)
    torch.cuda.manual_seed_all(seed)
    np.random.seed(seed)
    random.seed(seed)
    torch.backends.cudnn.deterministic = True
    torch.backends.cudnn.benchmark = False

def preprocess_image(image):
    pass

def classify_image(image):
    args = {
        'overwrite': False,
        'tqdm': 0,
        'config_file': 'configs/clothing1m.yaml',
        'dataset': 'clothing1M',
        'root': './data',
        'noise_type': 'clean',
        'noise_rate': 0.0,
        'save_dir': None,
        'gpus': '0',
        'num_workers': 8,
        'grad_bound': 0.0,
        'seed': 233,
        'backbone': 'res50',
        'optimizer': 'sgd',
        'momentum': 0.9,
        'nesterov': False,
        'pretrained': True,
        'ssl_pretrained': None,
        'resume': model_path,
        'lr': 0.01,
        'scheduler': 'cos',
        'milestones': None,
        'gamma': None,
        'weight_decay': 0.0001,
        'batch_size': 128,
        'start_epoch': None,
        'epochs': 100,
        'warmup': 0,
        'ema': False,
        'beta': 1.0,
        'num_classes': 14,
    }

    device = 'cpu'
    set_seed(args['seed'])

    MEAN = (0.485, 0.456, 0.406)
    STD = (0.229, 0.224, 0.225)

    test_loader = DataLoader(
        dataset=Clothing1M(
            image=image,
            train=False,
            transform=transforms.Compose([
                transforms.Resize(256),
                transforms.CenterCrop(224),
                transforms.ToTensor(),
                transforms.Normalize(MEAN, STD)]
            )),
        batch_size=256,
        shuffle=False,
        pin_memory=True,
        num_workers=args['num_workers'])

    model = resnet50(num_classes=args['num_classes'], show=True)
    nFeat = 2048

    state_dict = ResNet50_Weights.IMAGENET1K_V2.get_state_dict(progress=True)
    state_dict = {k:v for k,v in state_dict.items() if 'fc' not in k}
    missing, unexpected = model.load_state_dict(state_dict, strict=False)
    print('Loading ImageNet pretrained model')
    print('Model missing keys:\n', missing)
    print('Model unexpected keys:\n', unexpected)

    checkpoint = torch.load(args['resume'], map_location=torch.device(device))
    state_dict = checkpoint['model_state_dict']
    for key in list(state_dict.keys()):
        if 'ema_model' in key:
            state_dict[key.replace('ema_model.', '')] = state_dict[key]
            del state_dict[key]
        else:
            del state_dict[key]
    model.load_state_dict(state_dict)
    epoch = checkpoint['epoch']
    if args['start_epoch'] is None:
        args['start_epoch'] = epoch + 1

    model = model.to(device)

    loader_x, loader_y = None, None
    for x, y in test_loader:
        print(x)
        print(y)
        loader_x, loader_y = x.to(device), y.to(device)
        break
    z, _ = model(loader_x)
    pred = torch.argmax(z, 1)
    prediction_label = CATEGORY_NAMES[pred.item()]
    return f'Predicted label: {prediction_label}'

# Example image query (optional but recommended for demonstration)
example_image = "./examples/image_0.jpg"  # Ensure this image is available in the repo
example_image_2 = "./examples/image_7.jpg"

# Create Gradio interface
interface = gr.Interface(
    fn=classify_image,
    inputs=gr.Image(),
    outputs=gr.Text(),
    examples=[example_image, example_image_2]  # Include an example input for users -- you will want to find a relevant image to include and push it to your HuggingFace Space
)

if __name__ == "__main__":
    interface.launch()