LSTMEVAL(1)

🦋 LSTMEVAL(1) — lstmeval, but make it group-chat cute 🦋

🌸 NAME 🌸

lstmeval — evaluation program for LSTM-based networks. That's the whole name, bestie. Lowkey iconic.

---

📜 SYNOPSIS 📜

lstmeval --model
lang.lstm|modelname_checkpoint|modelname_N.NN_NN_NN.checkpoint
[--traineddata lang/lang.traineddata] --eval_listfile
lang.eval_files.txt [--verbosity N] [--max_image_MB NNNN]

---

💖 DESCRIPTION 💖

Okay so lstmeval(1) evaluates LSTM-based networks — it's giving “vibe check, but for your model.” You can give it either a recognition model or a training checkpoint as input for evaluation, along with a list of lstmf files. If you're evaluating a training checkpoint, --traineddata should also be specified. Intermediate training checkpoints can also be used, fr fr.

The synopsis is basically choose-your-fighter: lang.lstm, modelname_checkpoint, or modelname_N.NN_NN_NN.checkpoint. No cap, all three are valid forms in the original page.

---

✨ OPTIONS ✨

• --model FILE — Name of model file (training or recognition). (type:string default:)
• --traineddata FILE — If model is a training checkpoint, then traineddata must be the traineddata file that was given to the trainer. (type:string default:)
• --eval_listfile FILE — File listing sample files in lstmf training format. (type:string default:)
• --max_image_MB INT — Max memory to use for images. (type:int default:2000)
• --verbosity INT — Amount of diagnosting information to output (0-2). (type:int default:1)

Yes, the manpage says “diagnosting.” We keep the facts exactly as they are, bestie. 🫶

---

🕰️ HISTORY 🕰️

lstmeval(1) was first made available for tesseract4.00.00alpha. Baby's first era, iykyk. 💅

---

🔗 RESOURCES 🔗

• Main web site: https://github.com/tesseract-ocr
• Information on training tesseract LSTM: https://tesseract-ocr.github.io/tessdoc/TrainingTesseract-4.00.html

---

👀 SEE ALSO 👀

tesseract(1) — go say hi to the main character.

---

📜 COPYING 📜

Copyright (C) 2012 Google, Inc. Licensed under the Apache License, Version 2.0. Slay legally, bestie. ⚖️✨

---

👑 AUTHOR 👑

The Tesseract OCR engine was written by Ray Smith and his research groups at Hewlett Packard (1985-1995) and Google (2006-2018). Periodt. 💖

---

🫶 That's the whole page, bestie — dated 2026-09-28, sitting pretty in the LSTMEVAL(1) manual. Byeee! 🌈💋