Menafsirkan hasil prediksi dari model ekstraksi entity teks
Tetap teratur dengan koleksi
Simpan dan kategorikan konten berdasarkan preferensi Anda.
Setelah meminta prediksi, Vertex AI akan menampilkan hasil berdasarkan
tujuan model Anda. Prediksi dari model ekstraksi entity akan menampilkan anotasi untuk setiap dokumen, seperti lokasi entity yang terdeteksi, label yang ditetapkan, dan skor tingkat keyakinan.
Tingkat keyakinan menunjukkan seberapa yakin model Anda telah mengidentifikasi dan melabeli setiap entity secara akurat. Makin tinggi angkanya, makin tinggi tingkat keyakinan model terhadap ketepatan prediksi.
Contoh output prediksi batch
Sampel berikut adalah hasil yang diprediksi untuk model ekstraksi entity yang dilatih untuk mendeteksi penyakit. Offset (offset karakter awal dan akhir) menentukan lokasi tempat model mendeteksi entity dalam dokumen, dan kolom content menunjukkan entity yang terdeteksi.
Nama tampilan menunjukkan label yang diasosiasikan oleh model dengan setiap entity, misalnya SpecificDisease atau DiseaseClass. Label dipetakan ke segmen teks secara berurutan.
[[["Mudah dipahami","easyToUnderstand","thumb-up"],["Memecahkan masalah saya","solvedMyProblem","thumb-up"],["Lainnya","otherUp","thumb-up"]],[["Sulit dipahami","hardToUnderstand","thumb-down"],["Informasi atau kode contoh salah","incorrectInformationOrSampleCode","thumb-down"],["Informasi/contoh yang saya butuhkan tidak ada","missingTheInformationSamplesINeed","thumb-down"],["Masalah terjemahan","translationIssue","thumb-down"],["Lainnya","otherDown","thumb-down"]],["Terakhir diperbarui pada 2025-09-04 UTC."],[],[],null,["# Interpret prediction results from text entity extraction models\n\n| Starting on September 15, 2024, you can only customize classification, entity extraction, and sentiment analysis objectives by moving to Vertex AI Gemini prompts and tuning. Training or updating models for Vertex AI AutoML for Text classification, entity extraction, and sentiment analysis objectives will no longer be available. You can continue using existing Vertex AI AutoML Text models until June 15, 2025. For a comparison of AutoML text and Gemini, see [Gemini for AutoML text users](/vertex-ai/docs/start/automl-gemini-comparison). For more information about how Gemini offers enhanced user experience through improved prompting capabilities, see [Introduction to tuning](/vertex-ai/generative-ai/docs/models/tune-gemini-overview). To get started with tuning, see [Model tuning for Gemini text models](/vertex-ai/generative-ai/docs/models/tune_gemini/tune-gemini-learn)\n\nAfter requesting a prediction, Vertex AI returns results based on your\nmodel's objective. Predictions from entity extraction models return annotations\nfor each document, such as the location of detected entities, the assigned\nlabels, and confidence scores.\n\nThe confidence communicates how confident your model accurately identified and\nlabeled each entity. The higher the number, the higher the model's confidence in\nthe correctness of the prediction.\n\nExample batch prediction output\n-------------------------------\n\nThe following sample is the predicted result for an entity extraction model that\nwas trained to detect diseases. The offsets (start and end character offsets)\nspecify the location where the model detected an entity in the document, and the\n`content` field shows the detected entity.\n\nThe display names show the labels that the model associated with each entity,\nsuch as `SpecificDisease` or `DiseaseClass`. The labels map to the text segments\nin order.\n\n\n| **Note**: The following JSON Lines example includes line breaks for\n| readability. In your JSON Lines files, line breaks are included only after each\n| each JSON object.\n\n\u003cbr /\u003e\n\n\n```\n{\n \"key\": 1,\n \"predictions\": {\n \"ids\": [\n \"1234567890123456789\",\n \"2234567890123456789\",\n \"3234567890123456789\"\n ],\n \"displayNames\": [\n \"SpecificDisease\",\n \"DiseaseClass\",\n \"SpecificDisease\"\n ],\n \"textSegmentStartOffsets\": [13, 40, 57],\n \"textSegmentEndOffsets\": [29, 51, 75],\n \"confidences\": [\n 0.99959725141525269,\n 0.99912621492484128,\n 0.99935531616210938\n ]\n }\n}\n```"]]