kerasformers/resnet152_tv_in1k

See our collection for all versions of ResNet.

Run ResNet with Keras 3: JAX, PyTorch, or TensorFlow

kerasformers/resnet152_tv_in1k

Paper: Deep Residual Learning for Image Recognition (arXiv:1512.03385) · HF Papers

ResNet is the residual CNN backbone that introduced skip connections. Use ResNetImageClassify for ImageNet logits or ResNetModel (optionally as_backbone=True) for feature maps.

For more details on the model, please go to the upstream model card.

Pure-Keras 3 conversion of timm/resnet152.tv_in1k for kerasformers. One implementation runs unmodified on TensorFlow / Torch / JAX.

This is an image-classification / backbone checkpoint (ResNetImageClassify / ResNetModel).

✨ Quick start

import os
os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"

from PIL import Image
import numpy as np
from kerasformers.models.resnet import ResNetImageClassify, ResNetModel

model = ResNetImageClassify.from_weights("kerasformers/resnet152_tv_in1k")
backbone = ResNetModel.from_weights(
    "kerasformers/resnet152_tv_in1k", as_backbone=True
)

image = Image.open("your_image.jpg").convert("RGB")
image = image.resize((224, 224))
x = np.asarray(image, dtype="float32")[None]  # (1, H, W, 3)
print(model(x).shape)  # (1, num_classes)
feats = backbone(x)
print(len(feats), [tuple(f.shape) for f in feats])

Load any ResNet variant the same way with from_weights("kerasformers/<variant>"):

Variant	Hub
resnet101_a1_in1k	kerasformers/resnet101_a1_in1k
resnet101_gluon_in1k	kerasformers/resnet101_gluon_in1k
resnet101_tv_in1k	kerasformers/resnet101_tv_in1k
resnet152_a1_in1k	kerasformers/resnet152_a1_in1k
resnet152_gluon_in1k	kerasformers/resnet152_gluon_in1k
resnet152_tv_in1k	kerasformers/resnet152_tv_in1k
resnet50_a1_in1k	kerasformers/resnet50_a1_in1k
resnet50_gluon_in1k	kerasformers/resnet50_gluon_in1k
resnet50_tv_in1k	kerasformers/resnet50_tv_in1k

Tips

• Set KERAS_BACKEND before importing Keras / kerasformers.
• ResNetImageClassify returns class logits; ResNetModel returns features (as_backbone=True for multi-scale stages).
• See docs and Loading Weights.
• Upstream / timm checkpoints: ResNetImageClassify.from_weights("hf:timm/resnet152.tv_in1k").

Special Thanks

A huge thank you to the ResNet authors and the timm / Hub communities for creating and releasing these models.

License: see YAML license (usually matches the upstream checkpoint).