|
1 | | -.. highlight:: shell |
2 | | - |
3 | 1 | ============ |
4 | 2 | Installation |
5 | 3 | ============ |
@@ -43,36 +41,10 @@ To install with pdf output: |
43 | 41 |
|
44 | 42 | $ pip install pyhdx==0.4.0b6[pdf] |
45 | 43 |
|
46 | | -.. |
47 | | - From sources |
48 | | - ------------ |
49 | | -
|
50 | | - 1. Download or ``git clone`` the master branch of the PyHDX repository |
51 | | - |
52 | | - 2. Create a ``conda`` environment |
53 | | - |
54 | | - .. code-block:: rst |
55 | | -
|
56 | | - conda create --name <name> python=3.8 |
57 | | -
|
58 | | - 3. Activate conda environment |
59 | | - |
60 | | - .. code-block:: rst |
61 | | -
|
62 | | - conda activate <name> |
63 | | -
|
64 | | - 4. Install the dependencies |
65 | | - |
66 | | - ``conda install -c conda-forge pyhdx --only-deps`` |
67 | 44 |
|
68 | | - 5. Building wheels for the project |
69 | 45 |
|
70 | | - ``python setup.py sdist bdist_wheel`` |
71 | | - |
72 | | - 6. Installing the wheels (should be generated in the dist folder) |
73 | | - |
74 | | - ``pip install dist/PyHDX-version.whl`` |
75 | 46 |
|
| 47 | + |
76 | 48 |
|
77 | 49 | Running the web server |
78 | 50 | ---------------------- |
@@ -104,9 +76,69 @@ This will start a Dask cluster on the scheduler address as specified in the PyHD |
104 | 76 | (user dir / .pyhdx folder) |
105 | 77 |
|
106 | 78 |
|
| 79 | +From sources |
| 80 | +------------ |
| 81 | + |
| 82 | +1. Clone the PyHDX repository and cd into the project's root directory: |
| 83 | + .. code-block:: rst |
| 84 | +
|
| 85 | + git clone https://github.com/Jhsmit/PyHDX.git |
| 86 | + cd PyHDX |
| 87 | +
|
| 88 | +
|
| 89 | +2. Create a ``conda`` environment |
| 90 | + |
| 91 | + .. code-block:: rst |
| 92 | +
|
| 93 | + conda create --name <name> python=3.8 |
| 94 | +
|
| 95 | +3. Activate conda environment |
| 96 | + |
| 97 | + .. code-block:: rst |
| 98 | +
|
| 99 | + conda activate <name> |
| 100 | +
|
| 101 | +4. Install PyTorch |
| 102 | + |
| 103 | +If you would like a specific PyTorch version to use with PyHDX (ie CUDA/ROCm support), you should install this first. |
| 104 | +Installation instructions are on the Pytorch_ website. |
| 105 | + |
| 106 | +5. Install other dependencies |
| 107 | + |
| 108 | + .. code-block:: rst |
| 109 | +
|
| 110 | + conda install -c conda-forge pyhdx=0.4.0b6 --only-deps`` |
| 111 | +
|
| 112 | +This install dependencies only for base PyHDX. To install web application dependencies, |
| 113 | +run the `_requirements.py` file in the PyHDX root folder. This generates `reqs-<extra>.txt` files which lists |
| 114 | +requirements. |
| 115 | + |
| 116 | + .. code-block:: rst |
| 117 | + python _requirements.py |
| 118 | + pip install -r reqs-base.txt -r req-web.txt -r req-web.txt |
| 119 | +
|
| 120 | +Or |
| 121 | + |
| 122 | + .. code-block:: rst |
| 123 | + python _requirements.py |
| 124 | + conda install --file reqs-base.txt --file req-web.txt --file req-web.txt |
| 125 | +
|
| 126 | +
|
| 127 | +6. Install PyHDX in editable / development mode |
| 128 | + |
| 129 | + .. code-block:: rst |
| 130 | +
|
| 131 | + conda develop . |
| 132 | +
|
| 133 | + .. code-block:: rst |
| 134 | +
|
| 135 | + pip install -e . |
| 136 | +
|
107 | 137 | Dependencies |
108 | 138 | ------------ |
109 | 139 |
|
110 | 140 | The requirements for PyHDX are listed in setup.cfg |
111 | 141 |
|
112 | 142 | .. _Github repo: https://github.com/Jhsmit/pyhdx |
| 143 | + |
| 144 | +.. _Pytorch: https://pytorch.org/ |
0 commit comments