The kaleido executable is required by the kaleido Python library, but it was not included

[Antone4628]

I installed Kaleido today using 'conda install -c conda-forge python-kaleido'. The environment I installed it to is running python 3.8.5.

When I attempted to export a Plotly plot I received the following error:

−2
−1.5
−1
−0.5
0
0.5
1
1.5
2
Dupin's Cyclide with Color Corresponding to Mean Curvature

ValueError Traceback (most recent call last)
Cell In [19], line 24
21 if not os.path.exists("images"):
22 os.mkdir("images")
---> 24 sig.write_image("images/sig1.svg")

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/plotly/basedatatypes.py:3829, in BaseFigure.write_image(self, *args, **kwargs)
3769 """
3770 Convert a figure to a static image and write it to a file or writeable
3771 object
(...)
3825 None
3826 """
3827 import plotly.io as pio
-> 3829 return pio.write_image(self, *args, **kwargs)

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/plotly/io/_kaleido.py:267, in write_image(fig, file, format, scale, width, height, validate, engine)
251 raise ValueError(
252 """
253 Cannot infer image type from output path '{file}'.
(...)
261 )
262 )
264 # Request image
265 # -------------
266 # Do this first so we don't create a file if image conversion fails
--> 267 img_data = to_image(
268 fig,
269 format=format,
270 scale=scale,
271 width=width,
272 height=height,
273 validate=validate,
274 engine=engine,
275 )
277 # Open file
278 # ---------
279 if path is None:
280 # We previously failed to make sense of file as a pathlib object.
281 # Attempt to write to file as an open file descriptor.

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/plotly/io/_kaleido.py:144, in to_image(fig, format, width, height, scale, validate, engine)
141 # Validate figure
142 # ---------------
143 fig_dict = validate_coerce_fig_to_dict(fig, validate)
--> 144 img_bytes = scope.transform(
145 fig_dict, format=format, width=width, height=height, scale=scale
146 )
148 return img_bytes

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/kaleido/scopes/plotly.py:153, in PlotlyScope.transform(self, figure, format, width, height, scale)
142 raise ValueError(
143 "Invalid format '{original_format}'.\n"
144 " Supported formats: {supported_formats_str}"
(...)
148 )
149 )
151 # Transform in using _perform_transform rather than superclass so we can access the full
152 # response dict, including error codes.
--> 153 response = self._perform_transform(
154 figure, format=format, width=width, height=height, scale=scale
155 )
157 # Check for export error, later can customize error messages for plotly Python users
158 code = response.get("code", 0)

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/kaleido/scopes/base.py:293, in BaseScope._perform_transform(self, data, **kwargs)
284 """
285 Transform input data using the current scope, returning dict response with error code
286 whether successful or not.
(...)
290 :return: Dict of response from Kaleido executable, whether successful or not
291 """
292 # Ensure that kaleido subprocess is running
--> 293 self._ensure_kaleido()
295 # Perform export
296 export_spec = self._json_dumps(dict(kwargs, data=data)).encode('utf-8')

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/kaleido/scopes/base.py:176, in BaseScope._ensure_kaleido(self)
170 self._std_error = io.BytesIO()
172 # Launch kaleido subprocess
173 # Note: shell=True seems to be needed on Windows to handle executable path with
174 # spaces. The subprocess.Popen docs makes it sound like this shouldn't be
175 # necessary.
--> 176 proc_args = self._build_proc_args()
177 self._proc = subprocess.Popen(
178 proc_args,
179 stdin=subprocess.PIPE,
(...)
182 shell=sys.platform == "win32"
183 )
185 # Set up thread to asynchronously collect standard error stream

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/kaleido/scopes/base.py:123, in BaseScope._build_proc_args(self)
116 def _build_proc_args(self):
117 """
118 Build list of kaleido command-line arguments based on current values of
119 the properties specified by self._scope_flags and self.chromium_args
120
121 :return: list of flags
122 """
--> 123 proc_args = [self.executable_path(), self.scope_name]
124 for k in self._scope_flags:
125 v = getattr(self, k)

File ~/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/kaleido/scopes/base.py:99, in BaseScope.executable_path(cls)
97 path = os.environ.get("PATH", os.defpath)
98 formatted_path = path.replace(os.pathsep, "\n ")
---> 99 raise ValueError(
100 """
101 The kaleido executable is required by the kaleido Python library, but it was not included
102 in the Python package and it could not be found on the system PATH.
103
104 Searched for included kaleido executable at:
105 {vendored_executable_path}
106
107 Searched for executable 'kaleido' on the following system PATH:
108 {formatted_path}\n""".format(
109 vendored_executable_path=vendored_executable_path,
110 formatted_path=formatted_path,
111 )
112 )
114 return executable_path

ValueError:
The kaleido executable is required by the kaleido Python library, but it was not included
in the Python package and it could not be found on the system PATH.

Searched for included kaleido executable at:
/Users/antonechacartegui/opt/anaconda3/envs/pltenv/lib/python3.8/site-packages/kaleido/executable/kaleido

Searched for executable 'kaleido' on the following system PATH:
/Users/antonechacartegui/opt/anaconda3/bin
/Users/antonechacartegui/opt/anaconda3/condabin
/usr/local/bin
/usr/bin
/bin
/usr/sbin
/sbin
/Library/TeX/texbin
/opt/X11/bin

[Bewinxed]

Having this issue as well, suddenly i'm getting this error for scripts that worked fine before

[Antone4628]

I ended up installing Orca and using that engine to export my plots. It worked great for me. I'm not sure if this would help you for what you're doing but you might give it a try.

[Bewinxed]

For me, I uninstalled plotly and kaleido

reinstalled latest versions

then installed "urbansim" as some vague comment on StackOverflow suggested.

Now everything works :D

[gregorywaynepower]

For me, I uninstalled plotly and kaleido

reinstalled latest versions

then installed "urbansim" as some vague comment on StackOverflow suggested.

Now everything works :D

I did this, then attempting to install via pip as a last-ditch effort. But it ended up returning "kaleido already satisfied". And now I can generate images.

[sigSquared2]

I'm using mac 10.14.6 with python 3.10 in a conda environment whose kernel is mapped to a running jupyter notebook from a separate base environment. After installing kaleido via conda install -c conda-forge python-kaleido into my conda environment and attempting to render a static image via plotly, I encountered the same error message referenced by the OP. What I found, however, is that an a executible was installed with the kaleido package under /Users/user_name/anaconda3/envs/env_name/bin. There are two other executibles in this same /bin directory, one is under /KaleidoApp, and the second under /KaleidoApp/bin. I'm unaware of the differences between all three, besides they each have different file sizes.

I generated a copy of the kaleido executible in the /env_name/bin directory and placed it in /Users/user_name/anaconda3/envs/env_name/lib/python3.10/site-packages/kaleido/executable, restarted my env_name kernel and everything was resolved. Note that I had to create the executable directory because it didn't exist. Perhaps this is a Kaleido issue, or maybe it has something to do with conda's package manager?

[niketdoshi]

I'm using mac 10.14.6 with python 3.10 in a conda environment whose kernel is mapped to a running jupyter notebook from a separate base environment. After installing kaleido via conda install -c conda-forge python-kaleido into my conda environment and attempting to render a static image via plotly, I encountered the same error message referenced by the OP. What I found, however, is that an a executible was installed with the kaleido package under /Users/user_name/anaconda3/envs/env_name/bin. There are two other executibles in this same /bin directory, one is under /KaleidoApp, and the second under /KaleidoApp/bin. I'm unaware of the differences between all three, besides they each have different file sizes.

I generated a copy of the kaleido executible in the /env_name/bin directory and placed it in /Users/user_name/anaconda3/envs/env_name/lib/python3.10/site-packages/kaleido/executable, restarted my env_name kernel and everything was resolved. Note that I had to create the executable directory because it didn't exist. Perhaps this is a Kaleido issue, or maybe it has something to do with conda's package manager?

It worked for me as well

[creative-resort]

The trouble is with the custom conda environment not being activated upon kernel launch, by standard.
I've created an issue toward addressing this:
[ENH] - include conda env activation for kernel with custom environment

[njrhodes]

I had to do two things to solve this problem fully:

First, when I tried running a minimally reproducible example using save_image() which calls kaleido, I prefaced with:

reticulate::py_run_string("import sys")

but this threw the error

ModuleNotFoundError: No module named 'rpytools'

I had to copy the rpytools folder from the reticulate installation (located at: C:\Users\nrhode\Documents\R\win-library\4.1\reticulate\python) to my virtual environment folder listed as number [8] below:

library(reticulate)
sys <- import("sys", convert = TRUE)
sys$path 
[1] ""                                                                                      
[2] "C:\\Users\\nrhode\\AppData\\Local\\r-miniconda\\envs\\r-reticulate"                    
[3] "C:\\Users\\nrhode\\AppData\\Local\\r-miniconda\\envs\\r-reticulate\\python39.zip"      
[4] "C:\\Users\\nrhode\\AppData\\Local\\r-miniconda\\envs\\r-reticulate\\DLLs"              
[5] "C:\\Users\\nrhode\\AppData\\Local\\r-miniconda\\envs\\r-reticulate\\lib"               
[6] "C:\\Program Files\\RStudio\\resources\\app\\bin"                                       
[7] "C:\\Users\\nrhode\\AppData\\Local\\r-miniconda\\envs\\r-reticulate"                    
[8] "C:\\Users\\nrhode\\AppData\\Local\\r-miniconda\\envs\\r-reticulate\\lib\\site-packages"

A clue came from: rstudio/keras#615 (comment)

This solved the first error.

>library(reticulate)
>sys <- import("sys", convert = TRUE)
>sys$path 
>import("rpytools")
Module(rpytools)

Next, I attempted a minimal example which threw an error when trying to find the kaleido executable:

p <- plotly::plot_ly(x = 1:10)
reticulate::py_run_string("import sys")
plotly::save_image(p, "./pic.png")

Error:

reticulate::py_last_error()

-- Python Exception Message ----------------------------------------------------------------
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate\lib\site-packages\kaleido\scopes\plotly.py", line 103, in transform
    response = self._perform_transform(
  File "C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate\lib\site-packages\kaleido\scopes\base.py", line 280, in _perform_transform
    self._ensure_kaleido()
  File "C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate\lib\site-packages\kaleido\scopes\base.py", line 166, in _ensure_kaleido
    proc_args = self._build_proc_args()
  File "C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate\lib\site-packages\kaleido\scopes\base.py", line 113, in _build_proc_args
    proc_args = [self.executable_path(), self.scope_name]
  File "C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate\lib\site-packages\kaleido\scopes\base.py", line 89, in executable_path
    raise ValueError(
ValueError: 
The kaleido executable is required by the kaleido Python library, but it was not included
in the Python package and it could not be found on the system PATH.

Searched for included kaleido executable at:
    C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate\lib\site-packages\kaleido\executable\kaleido 

Searched for executable 'kaleido' on the following system PATH:
...

I finally figured out where the Kaleido executables were located on my system (path below) AND that I had to add this path location explicitly to my Windows 10 Environmental variable path:

C:\Users\nrhode\AppData\Local\r-miniconda\pkgs\kaleido-core-0.1.0-h8ffe710_0\Library\bin\KaleidoApp

This solved the problem. @mnneely @Siel

sessionInfo()
R version 4.1.2 (2021-11-01)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19044)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252

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

other attached packages:
[1] plotly_4.10.4 lubridate_1.9.3 forcats_1.0.0 stringr_1.5.1 dplyr_1.1.4
[6] purrr_1.0.2 readr_2.1.5 tidyr_1.3.1 tibble_3.2.1 ggplot2_3.5.0
[11] tidyverse_2.0.0 Pmetrics_2.1.1

loaded via a namespace (and not attached):
[1] nlme_3.1-153 fs_1.6.3 usethis_2.1.6
[4] devtools_2.4.5 webshot_0.5.5 progress_1.2.3
[7] httr_1.4.7 rprojroot_2.0.3 numDeriv_2016.8-1.1
[10] backports_1.4.1 tools_4.1.2 profvis_0.3.8
[13] utf8_1.2.4 R6_2.5.1 DT_0.32
[16] metafor_4.2-0 lazyeval_0.2.2 colorspace_2.1-0
[19] ggdist_3.2.1 withr_3.0.0 urlchecker_1.0.1
[22] tidyselect_1.2.0 gridExtra_2.3 prettyunits_1.2.0
[25] processx_3.8.3 curl_5.2.0 compiler_4.1.2
[28] cli_3.6.2 scales_1.3.0 checkmate_2.3.1
[31] mvtnorm_1.2-4 callr_3.7.5 rappdirs_0.3.3
[34] trelliscopejs_0.2.6 digest_0.6.34 rmarkdown_2.25
[37] base64enc_0.1-3 pkgconfig_2.0.3 htmltools_0.5.7
[40] sessioninfo_1.2.2 fastmap_1.1.1 htmlwidgets_1.6.4
[43] rlang_1.1.3 rstudioapi_0.14 shiny_1.8.0
[46] farver_2.1.1 generics_0.1.3 jsonlite_1.8.8
[49] crosstalk_1.2.1 mclust_6.1 distributional_0.3.2
[52] magrittr_2.0.3 metadat_1.2-0 patchwork_1.2.0
[55] Matrix_1.5-4 Rcpp_1.0.12 munsell_0.5.0
[58] fansi_1.0.6 reticulate_1.35.0 lifecycle_1.0.4
[61] yaml_2.3.8 stringi_1.8.3 mathjaxr_1.6-0
[64] pkgbuild_1.4.0 grid_4.1.2 promises_1.2.1
[67] crayon_1.5.2 miniUI_0.1.1.1 lattice_0.20-45
[70] hms_1.1.3 knitr_1.45 ps_1.7.6
[73] pillar_1.9.0 codetools_0.2-18 pkgload_1.3.2
[76] glue_1.7.0 evaluate_0.23 DistributionUtils_0.6-1
[79] data.table_1.15.0 remotes_2.4.2 png_0.1-8
[82] vctrs_0.6.5 tzdb_0.4.0 httpuv_1.6.14
[85] foreach_1.5.2 gtable_0.3.4 cachem_1.0.8
[88] xfun_0.42 mime_0.12 xtable_1.8-4
[91] later_1.3.2 autocogs_0.1.4 viridisLite_0.4.2
[94] iterators_1.0.14 memoise_2.0.1 timechange_0.3.0
[97] ellipsis_0.3.2 here_1.0.1

reticulate::py_list_packages()
package version requirement channel
1 blosc 1.21.5 blosc=1.21.5 conda-forge
2 bzip2 1.0.8 bzip2=1.0.8 conda-forge
3 c-blosc2 2.13.2 c-blosc2=2.13.2 conda-forge
4 ca-certificates 2024.2.2 ca-certificates=2024.2.2 conda-forge
5 hdf5 1.14.3 hdf5=1.14.3 conda-forge
6 intel-openmp 2024.0.0 intel-openmp=2024.0.0 conda-forge
7 kaleido-core 0.1.0 kaleido-core=0.1.0 conda-forge
8 krb5 1.21.2 krb5=1.21.2 conda-forge
9 libaec 1.1.2 libaec=1.1.2 conda-forge
10 libblas 3.9.0 libblas=3.9.0 conda-forge
11 libcblas 3.9.0 libcblas=3.9.0 conda-forge
12 libcurl 8.5.0 libcurl=8.5.0 conda-forge
13 libffi 3.4.2 libffi=3.4.2 conda-forge
14 libhwloc 2.9.3 libhwloc=2.9.3 conda-forge
15 libiconv 1.17 libiconv=1.17 conda-forge
16 liblapack 3.9.0 liblapack=3.9.0 conda-forge
17 libsqlite 3.45.1 libsqlite=3.45.1 conda-forge
18 libssh2 1.11.0 libssh2=1.11.0 conda-forge
19 libxml2 2.12.5 libxml2=2.12.5 conda-forge
20 libzlib 1.2.13 libzlib=1.2.13 conda-forge
21 lz4-c 1.9.4 lz4-c=1.9.4 conda-forge
22 mathjax 2.7.7 mathjax=2.7.7 conda-forge
23 mkl 2024.0.0 mkl=2024.0.0 conda-forge
24 numexpr 2.7.3 numexpr=2.7.3 conda-forge
25 numpy 1.26.4 numpy=1.26.4 conda-forge
26 openssl 3.2.1 openssl=3.2.1 conda-forge
27 orca 1.8 orca=1.8 conda-forge
28 packaging 23.2 packaging=23.2 conda-forge
29 pandas 2.2.1 pandas=2.2.1 conda-forge
30 patsy 0.5.6 patsy=0.5.6 conda-forge
31 pip 24.0 pip=24.0 conda-forge
32 plotly 5.19.0 plotly=5.19.0 plotly
33 prettytable 3.10.0 prettytable=3.10.0 conda-forge
34 pthreads-win32 2.9.1 pthreads-win32=2.9.1 conda-forge
35 py-cpuinfo 9.0.0 py-cpuinfo=9.0.0 conda-forge
36 pytables 3.9.2 pytables=3.9.2 conda-forge
37 python 3.9.18 python=3.9.18 conda-forge
38 python-dateutil 2.8.2 python-dateutil=2.8.2 conda-forge
39 python-kaleido 0.1.0 python-kaleido=0.1.0 conda-forge
40 python-tzdata 2024.1 python-tzdata=2024.1 conda-forge
41 python_abi 3.9 python_abi=3.9 conda-forge
42 pytz 2024.1 pytz=2024.1 conda-forge
43 pyyaml 6.0.1 pyyaml=6.0.1 conda-forge
44 scipy 1.12.0 scipy=1.12.0 conda-forge
45 setuptools 69.1.1 setuptools=69.1.1 conda-forge
46 six 1.16.0 six=1.16.0 conda-forge
47 snappy 1.1.10 snappy=1.1.10 conda-forge
48 statsmodels 0.14.1 statsmodels=0.14.1 conda-forge
49 tbb 2021.11.0 tbb=2021.11.0 conda-forge
50 tenacity 8.2.2 tenacity=8.2.2 pkgs/main
51 tk 8.6.13 tk=8.6.13 conda-forge
52 toolz 0.12.1 toolz=0.12.1 conda-forge
53 tzdata 2024a tzdata=2024a conda-forge
54 ucrt 10.0.22621.0 ucrt=10.0.22621.0 conda-forge
55 urbansim 3.2 urbansim=3.2 conda-forge
56 vc 14.3 vc=14.3 conda-forge
57 vc14_runtime 14.38.33130 vc14_runtime=14.38.33130 conda-forge
58 vs2015_runtime 14.38.33130 vs2015_runtime=14.38.33130 conda-forge
59 wcwidth 0.2.13 wcwidth=0.2.13 conda-forge
60 wheel 0.42.0 wheel=0.42.0 conda-forge
61 xz 5.2.6 xz=5.2.6 conda-forge
62 yaml 0.2.5 yaml=0.2.5 conda-forge
63 zlib-ng 2.0.7 zlib-ng=2.0.7 conda-forge
64 zstd 1.5.5 zstd=1.5.5 conda-forge

reticulate::conda_list()
name
1 base
2 r-reticulate
python
1 C:\Users\nrhode\AppData\Local\r-miniconda/python.exe
2 C:\Users\nrhode\AppData\Local\r-miniconda\envs\r-reticulate/python.exe

[romaromaromann]

I'm using mac 10.14.6 with python 3.10 in a conda environment whose kernel is mapped to a running jupyter notebook from a separate base environment. After installing kaleido via conda install -c conda-forge python-kaleido into my conda environment and attempting to render a static image via plotly, I encountered the same error message referenced by the OP. What I found, however, is that an a executible was installed with the kaleido package under /Users/user_name/anaconda3/envs/env_name/bin. There are two other executibles in this same /bin directory, one is under /KaleidoApp, and the second under /KaleidoApp/bin. I'm unaware of the differences between all three, besides they each have different file sizes.

I generated a copy of the kaleido executible in the /env_name/bin directory and placed it in /Users/user_name/anaconda3/envs/env_name/lib/python3.10/site-packages/kaleido/executable, restarted my env_name kernel and everything was resolved. Note that I had to create the executable directory because it didn't exist. Perhaps this is a Kaleido issue, or maybe it has something to do with conda's package manager?

Worked for me, thanks a lot