RuntimeError: Evaluation error: ValueError: None values not supported..

[ThomasWolf0701]

Describe the bug

I ran the cell sample from the github page and till the compile step everything is fine.
When running the compile step I get:

Error in py_call_impl(callable, dots$args, dots$keywords) :
RuntimeError: Evaluation error: ValueError: None values not supported..

A clear and concise description of what the bug is as
Detailed traceback:

Error in py_call_impl(callable, dots$args, dots$keywords) :
RuntimeError: Evaluation error: ValueError: None values not supported..

File "C:\Users\THOMAS1\DOCUME1\CONDA1\envs\R-RETI1\lib\site-packages\tensorflow_core\python\ops\math_ops.py", line 1422, in range
limit = ops.convert_to_tensor(limit, name="limit")
File "C:\Users\THOMAS1\DOCUME1\CONDA1\envs\R-RETI1\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1184, in convert_to_tensor
return convert_to_tensor_v2(value, dtype, preferred_dtype, name)
File "C:\Users\THOMAS1\DOCUME1\CONDA1\envs\R-RETI1\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1242, in convert_to_tensor_v2
as_ref=False)
File "C:\Users\THOMAS1\DOCUME1\CONDA1\envs\R-RETI1\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1296, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "C:\Users\THOMAS1\DOCUME1\CONDA1\envs\R-RETI1\lib\site-packages\tensorflow_c

The output after running yolo3

2020-10-17 19:37:03.572520: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_100.dll
2020-10-17 19:37:06.629388: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2020-10-17 19:37:06.667244: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: GeForce GTX 1070 major: 6 minor: 1 memoryClockRate(GHz): 1.645
pciBusID: 0000:01:00.0
2020-10-17 19:37:06.667676: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
2020-10-17 19:37:06.668985: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2020-10-17 19:37:06.669443: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2020-10-17 19:37:06.672431: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: GeForce GTX 1070 major: 6 minor: 1 memoryClockRate(GHz): 1.645
pciBusID: 0000:01:00.0
2020-10-17 19:37:06.672801: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
2020-10-17 19:37:06.674102: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2020-10-17 19:37:07.358505: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1159] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-10-17 19:37:07.358893: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1165] 0
2020-10-17 19:37:07.359069: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 0: N
2020-10-17 19:37:07.360054: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1304] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6794 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1)

The code i ran
library(tidyverse)
library(platypus)
library(abind)

BCCD_path <- "BCCD/"
annot_path <- file.path(BCCD_path, "Annotations/")
blood_labels <- c("Platelets", "RBC", "WBC")
n_class <- length(blood_labels)
net_h <- 416 # Must be divisible by 32
net_w <- 416 # Must be divisible by 32
anchors_per_grid <- 3

blood_anchors <- generate_anchors(
anchors_per_grid = anchors_per_grid, # Number of anchors (per one grid) to generate
annot_path = annot_path, # Annotations directory
labels = blood_labels, # Class labels
n_iter = 10, # Number of k-means++ iterations
annot_format = "pascal_voc", # Annotations format
seed = 55, # Random seed
centroid_fun = mean # Centroid function
)

blood_yolo <- yolo3(
net_h = net_h, # Input image height
net_w = net_w, # Input image width
grayscale = FALSE, # Should images be loaded as grayscale or RGB
n_class = n_class, # Number of object classes (80 for COCO dataset)
anchors = blood_anchors # Anchor boxes
)
blood_yolo %>% load_darknet_weights("yolov3.weights") # Optional

YOUR CODE HERE

Expected behavior
A clear and concise description of what you expected to happen.

Screenshots
If applicable, add screenshots to help explain your problem.

Session information (please complete the following information):

OS: [e.g. iOS]:
R version:
Python version:
TensorFlow (Python) version (tensorflow::tf_version()):
R session information (sessionInfo()):

[ThomasWolf0701]

Now i did a reinstall of keras and now get the error:
after:

blood_yolo %>%

• fit_generator(
  

•     generator = blood_yolo_generator,
  

•     epochs = 1000,
  

•     steps_per_epoch = 19,
  

•     validation_data = valid_blood_yolo_generator,
  

•     validation_steps = 5,
  

•     callbacks = list(callback_model_checkpoint("development/BCCD/blood_w.hdf5",
  

•                                                save_best_only = TRUE,
  

•                                                save_weights_only = TRUE)
  

•     )
  

• )
  

Fehler in py_call_impl(callable, dots$args, dots$keywords) :
RuntimeError: in user code:

C:\Users\THOMAS~1\DOCUME~1\CONDA~1\envs\R-RETI~1\lib\site-packages\tensorflow\python\keras\engine\training.py:571 train_function  *
    outputs = self.distribute_strategy.run(
C:\Users\THOMAS~1\DOCUME~1\CONDA~1\envs\R-RETI~1\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:951 run  **
    return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
C:\Users\THOMAS~1\DOCUME~1\CONDA~1\envs\R-RETI~1\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:2290 call_for_each_replica
    return self._call_for_each_replica(fn, args, kwargs)
C:\Users\THOMAS~1\DOCUME~1\CONDA~1\envs\R-RETI~1\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:2649 _call_for_each_replica
    return fn(*args, **kwargs)
C:\Users\THOMAS~1\DOCUME~1\CONDA~1\envs\R-RETI~1\lib\site-packages\tensorflow\python\keras\engine\training.py:533 train_step  **
    y, y_pred, sample_weight, regulariza

[maju116]

@ThomasWolf0701 there is a problem with custom R generators fitting with R keras package, plese see: #70
I'm still working on permanent fix. Please provide your session info, R version, TensorFlow version (python) ans system (windows, linux, Mac)

[ThomasWolf0701]

R version: 3.6.3
Tensorflow: 2.2
System: Windows

[ThomasWolf0701]

Is there a Tensorflow version for which this works ?

[maju116]

@ThomasWolf0701 what version of R packages (keras, tensorflow) do you have ?

[ThomasWolf0701]

@ThomasWolf0701 what version of R packages (keras, tensorflow) do you have ?

Here is the code i ran:

library(tidyverse)
library(platypus)
library(abind)

BCCD_path <- "development/BCCD/"
annot_path <- file.path(BCCD_path,"Annotations/")
blood_labels <- c("Platelets", "RBC", "WBC")
n_class <- length(blood_labels)
net_h <- 416 # Must be divisible by 32
net_w <- 416 # Must be divisible by 32
anchors_per_grid <- 3

blood_anchors <- generate_anchors(
anchors_per_grid = anchors_per_grid, # Number of anchors (per one grid) to generate
annot_path = annot_path, # Annotations directory
labels = blood_labels, # Class labels
n_iter = 10, # Number of k-means++ iterations
annot_format = "pascal_voc", # Annotations format
seed = 55, # Random seed
centroid_fun = mean # Centroid function
)

blood_yolo <- yolo3(
net_h = net_h, # Input image height
net_w = net_w, # Input image width
grayscale = FALSE, # Should images be loaded as grayscale or RGB
n_class = n_class, # Number of object classes (80 for COCO dataset)
anchors = blood_anchors # Anchor boxes
)

blood_yolo %>% load_darknet_weights("development/yolov3.weights") # Optional

blood_yolo %>% compile(
optimizer = optimizer_adam(),
loss = yolo3_loss(blood_anchors, n_class = n_class),
metrics = yolo3_metrics(blood_anchors, n_class = n_class)
)

Now i am back to getting the following error already after the compile step:
Fehler in py_call_impl(callable, dots$args, dots$keywords) :
RuntimeError: Evaluation error: ValueError: None values not supported.

Detailed traceback:
File "C:\Users\Thomas Wolf\anaconda3\lib\site-packages\tensorflow_core\python\ops\math_ops.py", line 1430, in range
limit = ops.convert_to_tensor(limit, dtype=dtype, name="limit")
File "C:\Users\Thomas Wolf\anaconda3\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1314, in convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "C:\Users\Thomas Wolf\anaconda3\lib\site-packages\tensorflow_core\python\framework\constant_op.py", line 317, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "C:\Users\Thomas Wolf\anaconda3\lib\site-packages\tensorflow_core\python\framework\constant_op.py", line 258, in constant
allow_broadcast=True)
File "C:\Users\Thomas Wolf\anaconda3\lib\site-packages\tensorflow_core\python\framework\constant_op.py", line 296, in _constant_impl
allow_bro

The session info:

sessionInfo()
R version 4.0.3 (2020-10-10)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19041)

Matrix products: default

locale:
[1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 LC_MONETARY=German_Germany.1252 LC_NUMERIC=C
[5] LC_TIME=German_Germany.1252

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] abind_1.4-5 platypus_0.1.1 keras_2.3.0.0 tensorflow_2.2.0 forcats_0.5.0 stringr_1.4.0 dplyr_1.0.2 purrr_0.3.4
[9] readr_1.4.0 tidyr_1.1.2 tibble_3.0.4 ggplot2_3.3.2 tidyverse_1.3.0

loaded via a namespace (and not attached):
[1] progress_1.2.2 reticulate_1.16 tidyselect_1.1.0 haven_2.3.1 lattice_0.20-41 colorspace_1.4-1 vctrs_0.3.4
[8] generics_0.0.2 base64enc_0.1-3 XML_3.99-0.5 blob_1.2.1 rlang_0.4.8 pillar_1.4.6 glue_1.4.2
[15] withr_2.3.0 DBI_1.1.0 rappdirs_0.3.1 dbplyr_1.4.4 RColorBrewer_1.1-2 modelr_0.1.8 readxl_1.3.1
[22] lifecycle_0.2.0 munsell_0.5.0 gtable_0.3.0 cellranger_1.1.0 rvest_0.3.6 labeling_0.3 tfruns_1.4
[29] fansi_0.4.1 broom_0.7.1 Rcpp_1.0.5 scales_1.1.1 backports_1.1.10 jsonlite_1.7.1 farver_2.0.3
[36] fs_1.5.0 gridExtra_2.3 digest_0.6.25 hms_0.5.3 stringi_1.5.3 grid_4.0.3 cli_2.1.0
[43] tools_4.0.3 magrittr_1.5 crayon_1.3.4 whisker_0.4 pkgconfig_2.0.3 zeallot_0.1.0 ellipsis_0.3.1
[50] Matrix_1.2-18 prettyunits_1.1.1 xml2_1.3.2 reprex_0.3.0 lubridate_1.7.9 assertthat_0.2.1 httr_1.4.2
[57] rstudioapi_0.11 R6_2.4.1 compiler_4.0.3

The reticulate info:

library(reticulate)
py_config()
python: C:/Users/Thomas Wolf/anaconda3/python.exe
libpython: C:/Users/Thomas Wolf/anaconda3/python37.dll
pythonhome: C:/Users/Thomas Wolf/anaconda3
version: 3.7.6 (default, Jan 8 2020, 20:23:39) [MSC v.1916 64 bit (AMD64)]
Architecture: 64bit
numpy: C:/Users/Thomas Wolf/anaconda3/Lib/site-packages/numpy
numpy_version: 1.18.1
tensorflow: C:\Users\THOMAS1\ANACON1\lib\site-packages\tensorflow_init_.p

python versions found:
C:/Users/Thomas Wolf/anaconda3/python.exe
C:/Users/Thomas Wolf/anaconda3/envs/my-rdkit-env/python.exe
C:/Users/Thomas Wolf/anaconda3/envs/r-tf/python.exe

I ran the example from this page and it runs just fine:
https://tensorflow.rstudio.com/guide/keras/

[maju116]

@ThomasWolf0701 please install platypus package from yolo3_fix branch: remotes::install_github("maju116/platypus@yolo3_fix") and instead of fit_generator function use yolo3_fit_generator function like in the example below:
https://github.com/maju116/platypus/blob/yolo3_fix/examples/Blood%20Cell%20Detection/Blood-Cell-Detection.md

It's still work in progress, but I hope it will be enough for now. Let me know if it's working for you.

[ThomasWolf0701]

@ThomasWolf0701 please install platypus package from yolo3_fix branch: remotes::install_github("maju116/platypus@yolo3_fix") and instead of fit_generator function use yolo3_fit_generator function like in the example below:
https://github.com/maju116/platypus/blob/yolo3_fix/examples/Blood%20Cell%20Detection/Blood-Cell-Detection.md

It's still work in progress, but I hope it will be enough for now. Let me know if it's working for you.

Tested it with a tensorflow 2.2 backend without gpu support for two epochs, and it´s working now. Mayn thanks for your help. Will let you know when I tested with GPU support.

[maju116]

@ThomasWolf0701 In my settings I still have en error when using yolo3_fit_generator if in the yolo3_generator I set shuffle = FALSE so for the moment please use shuffle = TRUE when creating generators for yolo3.