proxy Predictor Using Emotion Issues

[Ipassanti]

Thanks for sharing your Facial_Details_Synthesis code.
I got a Problem on Proxy Predictor by using emotion prior.

---

Error Information:

D:\file_Qian\environment\released_v0.1>python proxyPredictor.py -i ../imagefolder -o ../results --emotion 1
Using TensorFlow backend.
2020-05-05 08:38:40.403957: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2020-05-05 08:38:43.591496: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2020-05-05 08:38:43.640138: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce GTX 1650 computeCapability: 7.5
coreClock: 1.56GHz coreCount: 16 deviceMemorySize: 4.00GiB deviceMemoryBandwidth: 119.24GiB/s
2020-05-05 08:38:43.649424: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2020-05-05 08:38:43.685206: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2020-05-05 08:38:43.723567: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2020-05-05 08:38:43.744489: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2020-05-05 08:38:43.798967: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2020-05-05 08:38:43.827650: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2020-05-05 08:38:43.904791: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2020-05-05 08:38:43.910319: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2020-05-05 08:38:43.915798: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2020-05-05 08:38:43.922650: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce GTX 1650 computeCapability: 7.5
coreClock: 1.56GHz coreCount: 16 deviceMemorySize: 4.00GiB deviceMemoryBandwidth: 119.24GiB/s
2020-05-05 08:38:43.933750: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2020-05-05 08:38:43.938871: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2020-05-05 08:38:43.944149: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2020-05-05 08:38:43.950168: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2020-05-05 08:38:43.955576: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2020-05-05 08:38:43.961239: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2020-05-05 08:38:43.967510: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2020-05-05 08:38:43.973514: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2020-05-05 08:38:46.827408: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-05-05 08:38:46.833201: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0
2020-05-05 08:38:46.835981: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N
2020-05-05 08:38:46.842118: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 2915 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1650, pci bus id: 0000:01:00.0, compute capability: 7.5)
Loaded emotion model from disk
Attempting to read from directory: D:\file_Qian\environment\imagefolder
Loading the model
Model loaded
Starting tracking
D:\file_Qian\environment\imagefolder\sample2.jpg
===> Landmarks detection done.

===> estimating proxy of D:\file_Qian\environment\imagefolder\sample2.jpg
Using expression piror from emotion features.

Traceback (most recent call last):
File "proxyPredictor.py", line 198, in
main(args)
File "proxyPredictor.py", line 170, in main
features_emotion = inferanceOneImg(args, img_name, feature_model)
File "proxyPredictor.py", line 45, in inferanceOneImg
return feature_model.predict(img).reshape((-1))
File "C:\Users\LEGION\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 1441, in predict
x, _, _ = self._standardize_user_data(x)
File "C:\Users\LEGION\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 579, in _standardize_user_data
exception_prefix='input')
File "C:\Users\LEGION\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training_utils.py", line 135, in standardize_input_data
'with shape ' + str(data_shape))
ValueError: Error when checking input: expected conv2d_1_input to have 4 dimensions, but got array with shape (1, 64, 64, 2, 1)

---

---

Issue:1

ValueError: Error when checking input: expected conv2d_1_input to have 4 dimensions, but got array with shape (1, 64, 64, 2, 1)

it was like somehow my code dismissed dimension.

I used a 700 X 700 jpg as one input. And firstly i thought I set a wrong environment.
And reinstalled the environment.seems like it didn't work.

Issue:2

And i have another question,on another issues, I saw it can don't use downsample to get a finer output.But i'm not familiar to the code. I want to ask which part should i change to not downsample.