All Products
Search

This Product

    PolarDB:Inferensi dan interaksi data Qwen LLM

    Document Center

    PolarDB:Inferensi dan interaksi data Qwen LLM

    Last Updated:Jun 04, 2026

    PolarDB for AI mengintegrasikan Qwen LLM Alibaba Cloud ke dalam kluster PolarDB. Panggil Qwen LLM bawaan melalui SQL untuk melakukan inferensi dan interaksi terhadap data PolarDB Anda.

    Sebagai contoh, gunakan model analisis sentimen (_polar4ai_tongyi_sa) untuk menganalisis sentimen dalam data tabel, atau model chat (_polar4ai_tongyi) untuk menghasilkan jawaban dari input tertentu.

    Sentiment analysis model

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_sa, SELECT content FROM textinfo LIMIT 1) WITH ();
-- Konten dari tabel database: Barang ini hanya terlihat lumayan; pengalaman aktualnya sangat buruk. Saya tidak merekomendasikan pembeliannya.
    Output: Negative

    Chat model

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi, SELECT 'Who are you') WITH ();
    Output: I am a large language model from Alibaba Cloud. My name is Qwen.

    Scenarios

    Model

    Scenario

    Sentiment analysis model

    Menganalisis sentimen kalimat.

    Chat model

    Menghasilkan jawaban dari konten input.

    Summarization model

    Meringkas konten artikel.

    Translation model

    Menerjemahkan bahasa Tionghoa ke bahasa Inggris, atau bahasa lain (termasuk bahasa Inggris) ke bahasa Tionghoa.

    Positive/negative comment model

    Menghasilkan komentar positif atau negatif dari konten input.

    Limitations

    • Inferensi online memproses satu record dalam satu waktu. Inferensi offline mendukung pemrosesan batch.

    • Qwen LLM mendukung panjang konteks maksimum 8.000 token. Permintaan akan timeout setelah 10 detik karena batasan sumber daya komputasi. Tambahkan node AI untuk meningkatkan performa.

    Usage

    Syntax

    Panggil Qwen LLM bawaan untuk inferensi online atau offline menggunakan sintaks berikut. Detail sintaks lengkap tersedia di Run inference.

    --Online inference
/*polar4ai*/SELECT * FROM PREDICT (MODEL modelname, SELECT columnname FROM tablename) with ()
/*polar4ai*/SELECT * FROM PREDICT (MODEL modelname, SELECT text) with ()

--Offline inference
/*polar4ai*/SELECT * FROM PREDICT (MODEL modelname, SELECT columnname FROM tablename) with (mode='async')
/*polar4ai*/SELECT * FROM PREDICT (MODEL modelname, SELECT text) with (mode='async')

    Parameters

    Parameter

    Description

    modelname

    Nama model. Tentukan nama berdasarkan Scenarios. Nilai yang valid:

    • Sentiment analysis model: _polar4ai_tongyi_sa

    • Positive/negative comment model:

      • Positive comment model: _polar4ai_tongyi_p_comment

      • Negative comment model: _polar4ai_tongyi_n_comment

    • Summarization model: _polar4ai_tongyi_summarize

    • Translation model:

      • Translate to Chinese: _polar4ai_tongyi_tran_2_zh

      • Translate to English: _polar4ai_tongyi_tran_2_en

    • Chat model: _polar4ai_tongyi

    tablename

    Nama tabel.

    columnname

    Nama kolom tabel, digunakan sebagai input model.

    text

    Teks biasa, digunakan sebagai input model.

    Berjalan langsung tanpa tabel data. Cocok untuk skenario panggilan API.

    mode

    Mode inferensi.

    • Atur ke async untuk inferensi offline.

    • Abaikan untuk inferensi online.

    Examples

    1. Prepare the environment

    1. Tambahkan node AI dan konfigurasikan akun database-nya: Enable the PolarDB for AI feature

      Jika Anda telah menambahkan node AI saat pembelian kluster, konfigurasikan akun database secara langsung.

    2. Gunakan Cluster Endpoint untuk menghubungkan ke kluster PolarDB: Log on to PolarDB for AI

    2. Prepare data

    1. Buat tabel bernama textinfo.

      CREATE TABLE IF NOT EXISTS textinfo (
    id INT NOT NULL,
    content TEXT NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

    2. Tambahkan data berikut ke tabel textinfo.

      INSERT INTO textinfo (id,content) VALUES (1,"这个东西只是看着还行，实际体验上非常的不好，不推荐大家购买。");
INSERT INTO textinfo (id,content) VALUES (2,"华为最新手机Mate 60 Pro自开卖后销售火爆，目前已缺货。知名分析师郭明錤周一发文称，大幅提升Mate 60 Pro的出货量预期，华为对于产业和股市的影响力不可忽视，这种影响力正在回归中。");
INSERT INTO textinfo (id,content) VALUES (3,"近年来，中国电影市场逐渐倾向于现实题材。2018年的《我不是药神》以31亿票房成为暑期档冠军。随后，《扫毒2》、《怒火·重案》和《中国医生》等影片也取得了成功。今年暑期档，现实题材如《消失的她》、《八角笼中》和《孤注一掷》等电影大受欢迎，反映了观众对贴近生活内容的兴趣。");
INSERT INTO textinfo (id,content) VALUES (4,"你是谁");
INSERT INTO textinfo (id,content) VALUES (5,"This item only looks decent; the actual experience is very poor. I do not recommend purchasing it.");

    3. Call the Qwen LLM

    Online inference

    Inferensi online hanya memproses satu record dalam satu waktu.

    Sentiment analysis model

    Menganalisis sentimen kalimat. Kasus penggunaan umum: analisis ulasan e-commerce, analisis bias berita, dan pemantauan opini publik.

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_sa, SELECT content FROM textinfo WHERE id = 1) WITH ();

    Hasil yang dikembalikan:

    Negative

    Positive and negative comment model

    Menghasilkan komentar positif atau negatif berdasarkan sepotong konten.

    • Hasilkan komentar positif:

      /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_p_comment, SELECT content FROM textinfo WHERE id = 2) WITH ();

      Hasil yang dikembalikan:

      The Huawei Mate 60 Pro has been selling extremely well since its launch, which demonstrates strong consumer recognition and support for the Huawei brand. As a well-known technology company in China, Huawei has always attracted significant attention in the smartphone market. The post by Ming-Chi Kuo further confirms Huawei's influence in the industry and stock market, indicating that Huawei is gradually regaining its rightful position. Huawei's influence extends beyond product sales to its role in driving the entire tech industry and stock market. We should encourage and support excellent companies like Huawei to promote the continuous development of China's technology industry.

    • Hasilkan komentar negatif:

      /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_n_comment, SELECT content FROM textinfo WHERE id = 2) WITH ();

      Hasil yang dikembalikan:

      While the sales of the Huawei Mate 60 Pro may seem optimistic, from the perspective of market diversity and competition, the popularity of a single product might obscure consumer demand for other options. Over-concentration on one product could limit the market's drive for innovation, causing other potentially excellent products to not receive enough attention and development opportunities.

    Summarization model

    Meringkas teks yang diberikan.

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_summarize, SELECT content FROM textinfo WHERE id = 3) WITH ();

    Hasil yang dikembalikan:

    In recent years, the Chinese film market has increasingly favored realistic themes. In 2018, "Dying to Survive" became the summer box office champion with a revenue of 3.1 billion. Later, films like "The White Storm 2: Drug Lords," "Raging Fire," and "Chinese Doctors" also achieved great success. This summer, realistic films such as "Lost in the Stars," "Octagonal," and "No More Bets" have been very popular with audiences, reflecting a strong interest in content that reflects real life.

    Translation model

    • Terjemahkan sepotong konten Tionghoa ke bahasa Inggris.

      /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_tran_2_en, SELECT content FROM textinfo WHERE id = 1) WITH ();

      Hasil yang dikembalikan:

      This item only looks decent; the actual experience is very poor. I do not recommend purchasing it.

    • Terjemahkan bahasa lain (termasuk bahasa Inggris) ke bahasa Tionghoa.

      /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_tran_2_zh, SELECT content FROM textinfo WHERE id = 5) WITH ();

      Hasil yang dikembalikan:

      这件商品看起来还不错，实际体验却很差。我不推荐购买。

    Chat model

    Menghasilkan jawaban dari konten input, termasuk prompt eksplisit.

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi, SELECT content FROM textinfo WHERE id = 4) WITH ();

    Hasil yang dikembalikan:

    I am a large language model from Alibaba Cloud. My name is Qwen.

    Anda juga dapat memasukkan text biasa secara langsung.

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi, SELECT 'Who are you') WITH ();

    Hasil yang dikembalikan:

    I am a large language model from Alibaba Cloud. My name is Qwen.

    Offline inference

    Inferensi offline mendukung pemrosesan batch beberapa record. Contoh ini menggunakan sentiment analysis model:

    /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_sa, SELECT content FROM textinfo) WITH (mode='async');

    Hasil berikut dikembalikan:

    +--------------------------------------+
| task_id                              |
+--------------------------------------+
| 6227479c-1e82-11f0-b7cb-xxxxxxxxxxxx |
+--------------------------------------+

    Kueri mengembalikan task ID. Gunakan ID ini untuk memeriksa status tugas dan mengunduh hasilnya.

    /*polar4ai*/SHOW TASK `6227479c-1e82-11f0-b7cb-xxxxxxxxxxxx`;

    Ketika taskStatus menunjukkan finish, tugas telah selesai.

    +------------+---------------------------------------------------------------------------------------------------------------------------------------------+---------+----------------+----------------+--------+--------------+----------+
| taskStatus | filePath                                                                                                                                    | results | startTime      | endTime        | errMsg | successBatch | allBatch |
+------------+---------------------------------------------------------------------------------------------------------------------------------------------+---------+----------------+----------------+--------+--------------+----------+
| finish     | ["http://db4ai-xxx.aliyuncs.com/final_results/_polar4ai_tongyi_sa/xxx.csv?security-token=xxx&OSSAccessKeyId=xxx&Expires=xxx&Signature=xxx"] |         | 2025-04-21 xxx | 2025-04-21 xxx |        | 5            | 5        |
+------------+---------------------------------------------------------------------------------------------------------------------------------------------+---------+----------------+----------------+--------+--------------+----------+
    Catatan

    filePath kedaluwarsa setelah 100 menit. Periksa parameter Expires dalam filePath untuk memverifikasi masa berlakunya. Unduh hasil sebelum kedaluwarsa.

    4. (Optional) Write inference results to the database

    Untuk mengurangi penulisan ke database, PolarDB menyimpan hasil inferensi AI sebagai tabel eksternal PolarDB for AI Augmented Table (AAT) di OSS. Anda harus mengaktifkan arsip data dingin terlebih dahulu.

    1. Enable cold data archiving.

      Buka PolarDB console dan navigasi ke halaman Configuration and Management > Data and Lifecycle > Data Archive (Cold Data) untuk mengaktifkan fitur arsip data dingin.

    2. Buat database dan berikan izin ke akun tersebut.

      Buat database bernama polar4ai dengan set karakter utf8mb4. Sistem menyimpan AAT yang dihasilkan secara otomatis di database polar4ai. Berikan izin baca dan tulis ke akun database node AI.

    3. Gunakan sentiment analysis model untuk melakukan inferensi offline dan secara otomatis menghasilkan AAT.

      /*polar4ai*/SELECT * FROM PREDICT (MODEL _polar4ai_tongyi_sa, SELECT content FROM textinfo) WITH (x_cols='content', mode='async', into_type='db') INTO tongyi_sa_predict_in_db;

      Hasil berikut dikembalikan:

      +--------------------------------------+
| task_id                              |
+--------------------------------------+
| e61aefa8-42af-11f0-92d0-xxxxxxxxxxxx |
+--------------------------------------+

      Kueri mengembalikan task ID. Gunakan ID ini untuk memeriksa status tugas.

      /*polar4ai*/SHOW TASK `e61aefa8-42af-11f0-92d0-xxxxxxxxxxxx`;

      Ketika taskStatus menunjukkan finish, tugas telah selesai.

      +------------+------------+---------+----------------+----------------+--------+--------------+----------+
| taskStatus | filePath   | results | startTime      | endTime        | errMsg | successBatch | allBatch |
+------------+------------+---------+----------------+----------------+--------+--------------+----------+
| finish     | []         |         | 2025-04-21 xxx | 2025-04-21 xxx |        | 5            | 5        |
+------------+------------+---------+----------------+----------------+--------+--------------+----------+

    4. Lihat informasi tentang AAT yang dihasilkan secara otomatis di database polar4ai.

      SHOW CREATE TABLE polar4ai.tongyi_sa_predict_in_db;

      Hasil berikut dikembalikan:

      +--------------------------+-------------------------------------------------------------------------------------------------------------+
| Table                    | Create Table                                                                                                |
+--------------------------+-------------------------------------------------------------------------------------------------------------+
| tongyi_sa_predict_in_db  | CREATE TABLE `tongyi_sa_predict_in_db` (
  `content` text,
  `content_result` text
) ENGINE=CSV DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci /*!99990 800010128 NULL_MARKER='NULL' */ CONNECTION='default_oss_server' |
+--------------------------+-------------------------------------------------------------------------------------------------------------+

    5. Lihat data di AAT.

      SELECT * FROM polar4ai.tongyi_sa_predict_in_db;

      Hasil berikut dikembalikan:

      +-------------------------------------------------------------+----------------+
| content                                                     | content_result |
+-------------------------------------------------------------+----------------+
| 这个东西只是看着还行，实际体验上非常的不好，不推荐大家购买。          | Negative       |
| 华为最新手机Mate 60 Pro自开卖后销售火爆，目前已缺货。知名分析师郭... | Positive       |
| 近年来，中国电影市场逐渐倾向于现实题材。2018年的《我不是药神》以31... | Neutral        |
| 你是谁                                                       | Neutral        |
| This item only looks decent; the actual experience is ve... | Negative       |
+-------------------------------------------------------------+-----------------+