Python package / GNU Linux terminal utility for porting machine learning algorithms to FPGA.
Project description
MANUAL:
- “-t” (training) parameter:
Accepts a string with arguments for training the model.
- Training tasks executes in parallel processes. Each training task
generates: “.sv” file for FPGA, “.wd” file and “.tar” archive with model data for prospective classification task.
- Training rules:
“training_file.csv” must contain classes in the first column.
“training_file.csv” must not contain a column head.
The character - separator of “training_file.csv” must be “,”.
You can check available training algorithms with option “–info”.
- “-c” (classification) parameter:
Accepts a string with arguments for classification the passed data set.
- Classification tasks might be executed parallel on several FPGAs.
Several classification tasks, addressed to FPGA, execute consistently. Each classification task generates “.csv” file with predicted answers.
- Classification rules:
“classification_file.csv” must not contain a column head.
The character - separator of “training_file.csv” must be “,”.
The “word_dict.wd” might match to firmware of FPGA.
- You can check available training algorithms and USB serial ports
with option “–info”.
- “-s” (split) flag:
- In case the “-t” parameter is passed, it breaks the file into a training
and test sample in the ratio of 70/30.
- “-v” (verbose) flag:
Enables verbose output of all errors.
All additional notifications are highlighted in yellow.
- “-i” (information) flag:
Displays a message about the available algorithms and serial ports.
- “-a” (accuracy) flag:
Calculates classification accuracy.
- The sample passed for classification must contain the classes
in the first column.
Examples:
This example starts the training task:
$ fpga4p -t “training.csv,nbc_t”
This example starts the classification task:
$ fpga4p -c “class.csv,nbc_c,training.wd,/dev/ttyUSB0”
- This example prints available modules for training and
classification, as well as available serial USB ports:
$ fpga4p -i
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file fpga4p-2.1.tar.gz.
File metadata
- Download URL: fpga4p-2.1.tar.gz
- Upload date:
- Size: 18.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.1 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.6.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f0a90a16730af85d45e5d622ccf0540833b4798007d4425906138c180208c146
|
|
| MD5 |
b31a922195c2bc0558dda8201dd72148
|
|
| BLAKE2b-256 |
2c8a878424dfdf4e4b9805bf6f08a128213288934d8f2e38574f20924e1aefec
|
File details
Details for the file fpga4p-2.1-py3-none-any.whl.
File metadata
- Download URL: fpga4p-2.1-py3-none-any.whl
- Upload date:
- Size: 25.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.1 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.6.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1e1b9d440215adb484070b7eb024788a5934cd56ba805fbc230b4d309b7c5859
|
|
| MD5 |
b9c0203f1be8611756b83aee59a67889
|
|
| BLAKE2b-256 |
029218c851367d5848c5b10a675873c34e01bbda29190f93af761c2962e9e4c9
|