JaidedAI/EasyOCR

Issues

• RuntimeError: Mismatched state_dict keys when loading custom ResNet model into EasyOCR due to inconsistent layer naming

  #1343 opened 19 days ago by nrosto
  3

• problem with arabic lines

  #1338 opened a month ago by zMuh
  3

• Getting "Could not initialize NNPACK! Reason: Unsupported hardware." warning even though NNPACK is enabled

  #1352 opened 4 days ago by kirillmeisser
  0

• Can I install it without other language models except English model?

  #1348 opened 6 days ago by Septemberlemon
  1

• FileNotFoundError in `download_and_unzip` when running multiple easyocr's concurrently

  #1335 opened a month ago by starpit
  2

• Does Easyocr have metal (Mac GPU) support?

  #1351 opened 5 days ago by edegp
  1

• EasyOCR not reading Japanese text

  #1350 opened 5 days ago by TLiley2
  0

• Can you give tutorial How train and add language by myself ?

  #1349 opened 6 days ago by Hellomik2002
  0

• Problems with the unit-tests

  #1347 opened 9 days ago by UnitedMarsupials
  0

• numpy 2

  #1301 opened 4 months ago by Napolitain
  1

• How to limit VRam GPU usage

  #1326 opened 2 months ago by Pabloferex
  2

• Annual Patch and Release Policy Inquiry

  #1346 opened 14 days ago by gorigu
  0

• Using CPU. Note: This module is much faster with a GPU.

  #1340 opened 22 days ago by parvinders347
  1

• Fine-tuned CRAFT model works much slower on CPU than default one.

  #1324 opened 2 months ago by romanvelichkin
  1

• Runtime cuDNN error when training custom non-Latin-character model

  #1345 opened 14 days ago by S0mbre
  0

• Fine-Tuning EasyOCR on Urdu Textbooks

  #1342 opened 19 days ago by mujeeb-merwat
  0

• size mismatch for Prediction.bias Error!

  #1315 opened 2 months ago by Chiyan200
  2

• training not giving the expected results

  #1321 opened 2 months ago by RamadanHussein
  8

• Inconsistent loading of images

  #1339 opened a month ago by matskv
  0

• Easyocr still does not work well with Urdu language, in case of CNIC cards, its result is below average. How to accomplish this? Need guidance. And, also what if I want to prepare multilingual OCR including Urdu, English, Special characters, Roman Numerals, Numeric counting. How to achieve this in a single model. I have made alot of research but could not file any source to accomplish this.

  #1337 opened a month ago by Tahira2910
  0

• Adding support for Hebrew (2024 update)

  #1334 opened a month ago by danhertztech
  0

• Is this project abandoned?

  #1290 opened 5 months ago by ogencoglu
  3

• offline using

  #1332 opened a month ago by paul-yangmy
  5

• Some areas worthy of improvement

  #1331 opened 2 months ago by theonlybigmiao
  0

• Neither CUDA nor MPS are available - defaulting to CPU. Note: This module is much faster with a GPU. Illegal instruction

  #1323 opened 2 months ago by luisjose1996
  3

• Custom Model Doesn't Recognize Spaces

  #1328 opened 2 months ago by nrosto
  1

• OSError

  #1329 opened 2 months ago by ananitai440
  1

• installing easyocr without pip upgrade

  #1299 opened 4 months ago by deepakttwo
  0

• EasyOCR Training Issue

  #1327 opened 2 months ago by nrosto
  1

• Empty result for custom model

  #1325 opened 2 months ago by Spidrre
  2

• FutureWarning: You are using `torch.load` with `weights_only=False`

  #1297 opened 4 months ago by Brainstorrrm
  2

• Using GPU in the training

  #1320 opened 2 months ago by RamadanHussein
  3

• requirements.txt with all versions

  #1317 opened 2 months ago by Melanee-Melanee
  1

• Fa (Persian) mismatch

  #1316 opened 2 months ago by mohamadkenway
  0

• How do I control maximum and mimimum font size to be read?

  #1292 opened 5 months ago by ishandutta2007
  2

• bangla text issue

  #1313 opened 3 months ago by Automatorbd
  0

• "name 'F' is not defined" error when training a custom model

  #1312 opened 3 months ago by pasan148
  0

• Hailo Support

  #1310 opened 3 months ago by bash-sanka1p
  0

• `easyocr.Reader.readtext(...)` in rare occasions returns bounding box with `float` coordinates, and not `int`

  #1307 opened 3 months ago by ScheiBig
  2

• annot write mode CMYK as PNG

  #1308 opened 3 months ago by cxycxm
  0

• 段错误 (核心已转储)

  #1305 opened 3 months ago by SunNoJJ
  1

• Unable to install

  #1304 opened 3 months ago by elnath78
  2

• Fine-tuning the model for reading both English and Arabic

  #1303 opened 4 months ago by mariam-alsaleh
  0

• Bulgarian language produces garbage on Italic texts

  #1302 opened 4 months ago by varna9000
  0

• [BUG] Resize operation in CRAFT post processing

  #1300 opened 4 months ago by ash2703
  0

• Load a finetuned and pretrain model into same reader instance

  #1298 opened 4 months ago by hyder28
  0

• Add better support for Brazilian Portuguese

  #1296 opened 4 months ago by insinfo
  0

• Zero Accuracy in Trainer

  #1295 opened 4 months ago by init1mhd
  0

• pyinstaller error

  #1291 opened 5 months ago by yancccc
  2

• IndexError: index 54 is out of bounds for axis 0 with size 39

  #1287 opened 5 months ago by KPLinux
  1