missing models drusen.h5 pigment.h5 and amd.h5

[crisguycabs]

Hi!

My name is Crisostomo Barajas, faculty member and researcher in UDI (Colombia)

I´m trying to use DeepSeeNet as part of my research, and I found a problem: there are no h5 models in this repository or in the readme. I reviewed the model.py script, especifically the method:

def run_inference_final_score(model_folder, image_folder, input_file, output_file, risk_factors=['drusen', 'pigment', 'amd'])

and lines:

for risk_factor in risk_factors:  # For each model
    model_path = os.path.join(model_folder, risk_factor+'.h5')
    model = load_model(model_path)

Now, in this repository there are no links for downloading the three h5 models. There are, however, three downloable files in the DeepSeeNet repository; but when integrated in the project they failed. This is my best try to paste the error output:

(deepseeplus) simulaciones@hdspgpu:~/hdsp/Cris/deepseenet-plus$ python model.py -i --model_folder=models/ --image_folder=examples/example_images/ --input_file=examples/example_input_file.csv --output_file=examples/example_output.csv
2025-05-27 17:25:25.062097: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.
2025-05-27 17:25:25.065760: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2025-05-27 17:25:25.065777: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Running inference with output file: examples/example_output.csv
2025-05-27 17:25:26.596034: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596090: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublas.so.11'; dlerror: libcublas.so.11: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596126: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublasLt.so.11'; dlerror: libcublasLt.so.11: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596160: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596193: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcurand.so.10'; dlerror: libcurand.so.10: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596227: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusolver.so.11'; dlerror: libcusolver.so.11: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596259: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusparse.so.11'; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory
2025-05-27 17:25:26.596402: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2025-05-27 17:25:26.596617: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.

Running inference for drusen
Traceback (most recent call last):
File "/home/simulaciones/hdsp/Cris/deepseenet-plus/model.py", line 142, in
run_inference_final_score(model_folder, image_folder, input_file, output_file)
File "/home/simulaciones/hdsp/Cris/deepseenet-plus/model.py", line 54, in run_inference_final_score
left_pred = model.predict(x_left)
File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/utils/traceback_utils.py", line 67, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/tmp/autograph_generated_file0h2y5685.py", line 15, in tf__predict_function
retval = ag_.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
ValueError: in user code:

File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/engine/training.py", line 1845, in predict_function  *
    return step_function(self, iterator)
File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/engine/training.py", line 1834, in step_function  **
    outputs = model.distribute_strategy.run(run_step, args=(data,))
File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/engine/training.py", line 1823, in run_step  **
    outputs = model.predict_step(data)
File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/engine/training.py", line 1791, in predict_step
    return self(x, training=False)
File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/utils/traceback_utils.py", line 67, in error_handler
    raise e.with_traceback(filtered_tb) from None
File "/home/simulaciones/miniconda3/envs/deepseeplus/lib/python3.9/site-packages/keras/engine/input_spec.py", line 264, in assert_input_compatibility
    raise ValueError(f'Input {input_index} of layer "{layer_name}" is '

ValueError: Input 0 of layer "model_2" is incompatible with the layer: expected shape=(None, 224, 224, 3), found shape=(None, 512, 512, 3)

As you can see, the h5 files dont match the DeepSeeNet-Plus scripts.

Now, there is a downloable zip file with 3 h5 models called NS.h5, PCTCOL.h5 and PCTPSC.h5; loading and reviewing their summaries I can see that they have only one classification as exit, instead of the 3 classes expected for drusen, for example.

I humbly require your assistance with providing the correct h5 files for DeepSeeNet-Plus.

[qingyu-qc]

Hi,

The models and codes are both updated, so it will not match to the previous models. If you use the previous models, please follow the set up configuration in the original repo.

For the latest models, please follow the instruction in the readme:
The models can be downloaded via https://ftp.ncbi.nlm.nih.gov/pub/lu/Suppl/deeplensnet/models.zip (the Models section of https://github.com/ncbi-nlp/deepseenet-plus.git)

Thanks!