|
2 | 2 |
|
3 | 3 | ## Installation problems |
4 | 4 |
|
5 | | -### Tensorflow dependency |
6 | | - |
7 | | -ML Agents requires TensorFlow; if you don't already have it installed, `pip` |
8 | | -will try to install it when you install the ml-agents package. |
9 | | - |
10 | | -If you see a message like this |
11 | | - |
12 | | -```console |
13 | | -ERROR: Could not find a version that satisfies the requirement tensorflow<2.0,>=1.7 (from mlagents) (from versions: none) |
14 | | -ERROR: No matching distribution found for tensorflow<2.0,>=1.7 (from mlagents) |
15 | | -``` |
16 | | - |
17 | | -it means that there is no version of TensorFlow for your python environment. |
18 | | -Some known potential causes are: |
19 | | - |
20 | | -- You're using 32-bit python instead of 64-bit. See the answer |
21 | | - [here](https://stackoverflow.com/a/1405971/224264) for how to tell which you |
22 | | - have installed. |
23 | | -- You have the `tensorflow-gpu` package installed. This is equivalent to |
24 | | - `tensorflow`, however `pip` doesn't recognize this. The best way to resolve |
25 | | - this is to update to `tensorflow==1.15.0` which provides GPU support in the |
26 | | - same package (see the |
27 | | - [release notes](https://github.com/tensorflow/tensorflow/issues/33374) for |
28 | | - more details.) |
29 | | -- You're on another architecture (e.g. ARM) which requires vendor provided |
30 | | - packages. |
31 | | - |
32 | | -In all of these cases, the issue is a pip/python environment setup issue. Please |
33 | | -search the tensorflow github issues for similar problems and solutions before |
34 | | -creating a new issue. |
35 | | - |
36 | | -#### Visual C++ Dependency (Windows Users) |
37 | | -When running `mlagents-learn`, if you see a stack trace with a message like this: |
38 | | - |
39 | | -```console |
40 | | -ImportError: DLL load failed: The specified module could not be found. |
41 | | -``` |
42 | | - |
43 | | -then either of the required DLLs, `msvcp140.dll` (old) or `msvcp140_1.dll` (new), are missing on your machine. The `import tensorflow` command will print this warning message. |
44 | | - |
45 | | -To solve it, download and install (with a reboot) the install [Microsoft Visual C++ Redistributable for Visual Studio 2015, 2017 and 2019](https://support.microsoft.com/en-my/help/2977003/the-latest-supported-visual-c-downloads). |
46 | | - |
47 | | -For more details, please see the [TensorFlow 2.1.0 release notes](https://github.com/tensorflow/tensorflow/releases/tag/v2.1.0) |
48 | | -and the [TensorFlow github issue](https://github.com/tensorflow/tensorflow/issues/22794#issuecomment-573297027). |
49 | | - |
50 | 5 | ## Environment Permission Error |
51 | 6 |
|
52 | 7 | If you directly import your Unity environment without building it in the editor, |
|
0 commit comments