Hi,guys.While using your model to inference my SVS from TCGA，I encountered a situation where the picture was garbled.I followed readme introduction，The inference environment and the generation environment were configured separately according to the requirements specified in the requirements.txt file.Model was fetched fromhttps://zenodo.org/records/15272883.
Below is the operational report. My device is RTX4090, with CUDA10.1 and CUDNN7.6.5 installed.
(AIgrading) G:\wl\Aigrading\AIgrading-main>python ./inference_slide/main_gp.py -d "G:\wl\Aigrading\SVS\output\cws_tiling" -o "G:\wl\Aigrading\SVS\Results\Masks_step1_output" -s "G:\wl\Aigrading\SVS\Results\Masks_step2_stitched" -sf "G:\wl\Aigrading\SVS\Results\Masks_step3_final"
2025-11-16 19:10:48.342582: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
Found 372 samples to process.
--- Processing sample 1/372: TCGA-05-4244-01Z.svs ---
C:\Users\1.conda\envs\AIgrading\lib\site-packages\tensorflow\python\keras\layers\core.py:1028: UserWarning: Res50PSCA is not loaded, but a Lambda layer uses it. It may cause errors.
warnings.warn('{} is not loaded, but a Lambda layer uses it. '
C:\Users\1.conda\envs\AIgrading\lib\site-packages\tensorflow\python\keras\layers\core.py:1028: UserWarning: Self3CrossAttention_Res507 is not loaded, but a Lambda layer uses it. It may cause errors.
warnings.warn('{} is not loaded, but a Lambda layer uses it. '
2025-11-16 19:10:52.209417: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2025-11-16 19:10:52.239688: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties:
pciBusID: 0000:81:00.0 name: NVIDIA GeForce RTX 4090 D computeCapability: 8.9
coreClock: 2.52GHz coreCount: 114 deviceMemorySize: 23.99GiB deviceMemoryBandwidth: 938.86GiB/s
2025-11-16 19:10:52.239809: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2025-11-16 19:10:52.247318: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2025-11-16 19:10:52.251373: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2025-11-16 19:10:52.253040: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2025-11-16 19:10:52.258452: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2025-11-16 19:10:52.261081: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2025-11-16 19:10:52.269106: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2025-11-16 19:10:52.269241: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2025-11-16 19:10:52.269589: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2025-11-16 19:10:52.283252: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x24ec7b740c0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2025-11-16 19:10:52.283649: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
2025-11-16 19:10:52.283850: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties:
pciBusID: 0000:81:00.0 name: NVIDIA GeForce RTX 4090 D computeCapability: 8.9
coreClock: 2.52GHz coreCount: 114 deviceMemorySize: 23.99GiB deviceMemoryBandwidth: 938.86GiB/s
2025-11-16 19:10:52.283943: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2025-11-16 19:10:52.284014: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2025-11-16 19:10:52.284086: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2025-11-16 19:10:52.284156: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2025-11-16 19:10:52.284225: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2025-11-16 19:10:52.284293: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2025-11-16 19:10:52.284355: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2025-11-16 19:10:52.284450: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2025-11-16 19:10:53.366490: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix:
2025-11-16 19:10:53.366602: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108] 0
2025-11-16 19:10:53.367286: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0: N
2025-11-16 19:10:53.367486: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 21559 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 4090 D, pci bus id: 0000:81:00.0, compute capability: 8.9)
2025-11-16 19:10:53.370872: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x24f3fa69f60 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2025-11-16 19:10:53.370973: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA GeForce RTX 4090 D, Compute Capability 8.9
G:\wl\Aigrading\SVS\output\cws_tiling\TCGA-05-4244-01Z.svs
2025-11-16 19:11:01.771649: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2025-11-16 19:11:26.050179: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
The final interpretation image is mostly garbled, and there are also pure black images.
Hi,guys.While using your model to inference my SVS from TCGA，I encountered a situation where the picture was garbled.I followed readme introduction，The inference environment and the generation environment were configured separately according to the requirements specified in the requirements.txt file.Model was fetched fromhttps://zenodo.org/records/15272883.
Below is the operational report. My device is RTX4090, with CUDA10.1 and CUDNN7.6.5 installed.
(AIgrading) G:\wl\Aigrading\AIgrading-main>python ./inference_slide/main_gp.py -d "G:\wl\Aigrading\SVS\output\cws_tiling" -o "G:\wl\Aigrading\SVS\Results\Masks_step1_output" -s "G:\wl\Aigrading\SVS\Results\Masks_step2_stitched" -sf "G:\wl\Aigrading\SVS\Results\Masks_step3_final"
2025-11-16 19:10:48.342582: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
Found 372 samples to process.
--- Processing sample 1/372: TCGA-05-4244-01Z.svs ---
C:\Users\1.conda\envs\AIgrading\lib\site-packages\tensorflow\python\keras\layers\core.py:1028: UserWarning: Res50PSCA is not loaded, but a Lambda layer uses it. It may cause errors.
warnings.warn('{} is not loaded, but a Lambda layer uses it. '
C:\Users\1.conda\envs\AIgrading\lib\site-packages\tensorflow\python\keras\layers\core.py:1028: UserWarning: Self3CrossAttention_Res507 is not loaded, but a Lambda layer uses it. It may cause errors.
warnings.warn('{} is not loaded, but a Lambda layer uses it. '
2025-11-16 19:10:52.209417: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2025-11-16 19:10:52.239688: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties:
pciBusID: 0000:81:00.0 name: NVIDIA GeForce RTX 4090 D computeCapability: 8.9
coreClock: 2.52GHz coreCount: 114 deviceMemorySize: 23.99GiB deviceMemoryBandwidth: 938.86GiB/s
2025-11-16 19:10:52.239809: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2025-11-16 19:10:52.247318: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2025-11-16 19:10:52.251373: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2025-11-16 19:10:52.253040: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2025-11-16 19:10:52.258452: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2025-11-16 19:10:52.261081: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2025-11-16 19:10:52.269106: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2025-11-16 19:10:52.269241: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2025-11-16 19:10:52.269589: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2025-11-16 19:10:52.283252: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x24ec7b740c0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2025-11-16 19:10:52.283649: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
2025-11-16 19:10:52.283850: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties:
pciBusID: 0000:81:00.0 name: NVIDIA GeForce RTX 4090 D computeCapability: 8.9
coreClock: 2.52GHz coreCount: 114 deviceMemorySize: 23.99GiB deviceMemoryBandwidth: 938.86GiB/s
2025-11-16 19:10:52.283943: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2025-11-16 19:10:52.284014: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2025-11-16 19:10:52.284086: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2025-11-16 19:10:52.284156: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2025-11-16 19:10:52.284225: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2025-11-16 19:10:52.284293: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2025-11-16 19:10:52.284355: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2025-11-16 19:10:52.284450: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2025-11-16 19:10:53.366490: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix:
2025-11-16 19:10:53.366602: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108] 0
2025-11-16 19:10:53.367286: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0: N
2025-11-16 19:10:53.367486: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 21559 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 4090 D, pci bus id: 0000:81:00.0, compute capability: 8.9)
2025-11-16 19:10:53.370872: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x24f3fa69f60 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2025-11-16 19:10:53.370973: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA GeForce RTX 4090 D, Compute Capability 8.9
G:\wl\Aigrading\SVS\output\cws_tiling\TCGA-05-4244-01Z.svs
2025-11-16 19:11:01.771649: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2025-11-16 19:11:26.050179: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
The final interpretation image is mostly garbled, and there are also pure black images.