1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
|
# Baler on Windows Native+CUDA (GPU) - Experimental Guide
This documentation provides a step-by-step guide on running Baler with Windows Native+CUDA (GPU). Please follow the instructions carefully to ensure successful setup and execution.
## Prerequisites:
- Windows OS (optional: with CUDA compatible GPU installed).
- Ensure you have Git installed on your system for cloning the Baler project.
## Setup:
### STEP 1: Install Python3.10
Download and install Python 3.10 from the official website using the following link:
[Python 3.10.11](https://www.python.org/ftp/python/3.10.11/python-3.10.11-amd64.exe)
### STEP 2: Install Poetry
Once Python is installed, open your terminal or command prompt and run the following command to install Poetry:
```console
python -m pip install poetry
```
### STEP 3: Clone Baler Project
Refer to the primary README documentation of the Baler project for detailed instructions on cloning the repository using Git.
### STEP 4: Prepare Baler Project for GPU
Navigate to the cloned Baler directory. If you find a `poetry.lock` file in the directory, delete it.
Now, open the `pyproject.toml` file using your preferred text editor.
### STEP 5: (Optional) Update PyTorch Version for CUDA
In the `pyproject.toml` file, find the torch dependency version and update it with the following URL:
Replace this:
```
torch = ">=2.0.0, !=2.0.1"
```
With this:
```
torch = { url = "https://download.pytorch.org/whl/cu118/torch-2.0.0%2Bcu118-cp310-cp310-win_amd64.whl#sha256=5ee2b7c19265b9c869525c378fcdf350510b8f3fc08af26da1a2587a34cea8f5"}
```
> **Note**: This step is particularly important if you want to run your training on GPUs. Please ensure that the version you are using (in this case: PyTorch 2.0.0 with CUDA 11.8 (cu118) for Python 3.10 (cp310)) corresponds to the version you want to run.
### STEP 5a: (Optional) Disable Energy Profiling
If you wish to avoid profiling the energy usage (especially when encountering memory issues with GPU acceleration), open the `baler/baler.py` file and comment out lines 79 and 80. These lines should look like:
```python
@pytorch_profile
@energy_profiling(project_name="baler_training", measure_power_secs=1)
```
To comment them, simply place a `#` in front of each line:
```python
# @pytorch_profile
# @energy_profiling(project_name="baler_training", measure_power_secs=1)
```
> **Note**: This step is particularly important if you face out-of-memory issues when utilizing GPU acceleration or encounter memory-related problems in general.
### STEP 6: Install Project Dependencies
Inside the Baler directory, execute the following command to install the required dependencies:
```console
python -m poetry install
```
### STEP 7: Run Baler with Poetry
Once all dependencies are installed, you can run Baler with the following command:
```console
python -m poetry run baler baler [-h] --mode MODE --project WORKSPACE PROJECT [--verbose]
```
Replace `MODE`, `WORKSPACE`, and `PROJECT` with appropriate values as per your requirement.
---
Thank you for using Baler! If you face any issues, please refer to the project's GitHub issues section or contact the project maintainers.
|
