PHP code example of syncfly / python-in-php

    
        
<?php
require_once('vendor/autoload.php');

/* Start to develop here. Best regards https://php-download.com/ */

    

syncfly / python-in-php example snippets

use py\json;
use py\datetime\datetime;

echo json::dumps(['hello' => 'world']); // {"hello": "world"}
echo datetime::now()->isoformat();       // 2024-01-15T12:34:56.789012

use py\numpy;

$arr = numpy::array([1, 2, 3, 4, 5]);
echo numpy::mean($arr); // 3.0

use py\requests;

$response = requests::get('https://httpbin.org/json');
$data = $response->json();  // method call
echo $response->status_code; // attribute access

Py::eval('2 ** 10');                    // 1024
Py::sum([1, 2, 3]);                     // 6  (Python sum(), not array_sum)
Py::sorted([3, 1, 2], reverse: true);   // [3, 2, 1]  — PHP named args become Python kwargs
Py::builtin('pow', 2, 8);               // 256  — call any builtin by name

use py\builtins;

// with open(...) as f:  — the callback receives the entered value
Py::with(builtins::open('/tmp/data.txt', 'a'), function ($f) {
    $f->write(' world');
});
// the file is closed automatically when the context exits

use py\builtins;

// A PHP closure invoked by Python's map()
$doubled = builtins::list(builtins::map(fn ($x) => $x * 2, [1, 2, 3]));
// [2, 4, 6]

// As a sorting key
$sorted = builtins::sorted(['ccc', 'a', 'bb'], key: fn ($w) => strlen($w));
// ['a', 'bb', 'ccc']

use py\functools;

$adder = functools::partial(fn ($a, $b) => $a + $b, 10);
echo $adder(5); // 15 — the PHP callback runs each time Python calls the partial

> $lengths = builtins::map(Py::callback('strlen'), ['a', 'bb', 'ccc']);
> 

use Python_In_PHP\PythonException;
use py\json;

try {
    json::loads('invalid json');
} catch (PythonException $e) {
    echo $e->getMessage();   // "Python error: Expecting value: line 1..."
    echo $e->traceback;      // full Python traceback
}

use py\transformers;
use py\torch;

$model_name = 'google/gemma-3-4b-it';

$tokenizer = transformers\AutoTokenizer::from_pretrained($model_name);

$model = transformers\AutoModelForCausalLM::from_pretrained(
    $model_name,
    torch_dtype: torch::$bfloat16,
    device_map: "auto"
);

$messages = [
    ['role' => 'user', 'content' => 'Why PHP is great?']
];

$input_ids = $tokenizer->apply_chat_template(
    $messages,
    return_tensors: 'pt',
    add_generation_prompt: true
);

$outputs = $model->generate($input_ids, max_new_tokens: 2048);
$result = $tokenizer->decode($outputs[0], skip_special_tokens: true);

use py\transformers;
use py\torch;

$pipe = transformers\pipeline(
    'text-generation',
    model: 'google/gemma-3-4b-it',
    torch_dtype: torch::$bfloat16,
    device_map: 'auto'
);

$messages = [['role' => 'user', 'content' => 'Why PHP is great?']];
$output = $pipe($messages, max_new_tokens: 2048);
$result = end($output[0]['generated_text'])['content'];

bash
composer