Will you provide the pretrained model?

[GuideWsp]

No description provided.

[xiezw5]

Yes, it will be released soon.

[aligoglos]

Please upload on drive.

[ABDOELSHEMY]

please upload the Pretrained models on drive or something easy to download from, thank you

[zhihongp]

Ran into problems with pre-trained models when run ./test_models_pc.sh cdc_x2_test ./CDC_test.py ./models/HGSR-MHR_X2_CDC.pth 2

Could you help double check? BTW, the x4 version seems to work, and x3 has load_state_dict issue too.

Initializing DataSet, image list: ./DRealSR/Test_x2 ...
Found 83 HR 83 LR ...
Build Generator Net...
Loading from DataParallel module......
Traceback (most recent call last):
File "../TorchTools/TorchNet/tools.py", line 178, in load_weights
model.load_state_dict(model_weights, strict=strict)
File "./lib/python3.6/site-packages/torch/nn/modules/module.py", line 845, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for HourGlassNetMultiScaleInt:
Missing key(s) in state_dict: "flat_map.2.1.weight", "flat_map.2.1.bias", "flat_map.4.0.weight", "flat_map.4.0.bias", "edge_map.2.1.weight", "edge_map.2.1.bias", "edge_map.4.0.weight", "edge_map.4.0.bias", "corner_map.2.1.weight", "corner_map.2.1.bias", "corner_map.4.0.weight", "corner_map.4.0.bias", "upsample_flat.2.1.weight", "upsample_flat.2.1.bias", "upsample_flat.4.0.weight", "upsample_flat.4.0.bias", "upsample_edge.2.1.weight", "upsample_edge.2.1.bias", "upsample_edge.4.0.weight", "upsample_edge.4.0.bias", "upsample_corner.2.1.weight", "upsample_corner.2.1.bias", "upsample_corner.4.0.weight", "upsample_corner.4.0.bias".
Unexpected key(s) in state_dict: "flat_map.2.0.weight", "flat_map.2.0.bias", "edge_map.2.0.weight", "edge_map.2.0.bias", "corner_map.2.0.weight", "corner_map.2.0.bias", "upsample_flat.2.0.weight", "upsample_flat.2.0.bias", "upsample_edge.2.0.weight", "upsample_edge.2.0.bias", "upsample_corner.2.0.weight", "upsample_corner.2.0.bias".
size mismatch for flat_map.3.0.weight: copying a param with shape torch.Size([3, 64, 1, 1]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for flat_map.3.0.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for edge_map.3.0.weight: copying a param with shape torch.Size([3, 64, 1, 1]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for edge_map.3.0.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for corner_map.3.0.weight: copying a param with shape torch.Size([3, 64, 1, 1]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for corner_map.3.0.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for upsample_flat.3.0.weight: copying a param with shape torch.Size([3, 64, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for upsample_flat.3.0.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for upsample_edge.3.0.weight: copying a param with shape torch.Size([3, 64, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for upsample_edge.3.0.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for upsample_corner.3.0.weight: copying a param with shape torch.Size([3, 64, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for upsample_corner.3.0.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([64]).