From 02c4eefc38c6f6fabdde47422f209ed77bf868a1 Mon Sep 17 00:00:00 2001 From: Karan Jajoria Date: Tue, 22 Jul 2025 20:01:34 +0530 Subject: [PATCH] I have tried to resolve an issue based on #182 Someone raised an Issue (#182 ), I tried to resolve it by manual installation of patch-ng which is connect with the error. --- README.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/README.md b/README.md index 87f1016edd..a05f367d57 100644 --- a/README.md +++ b/README.md @@ -71,6 +71,13 @@ Follow these instructions to integrate word beam search decoding: 2. Compile and install by running `pip install .` at the root level of the CTCWordBeamSearch repository 3. Specify the command line option `--decoder wordbeamsearch` when executing `main.py` to actually use the decoder +**If you are having issue with** +`pip install -r requirement.txt` +**then this might be a problem while wheel preparation step for the lmbd==1.0.0 package.**
+**Try installing the patch-ng mannually by**
+`pip install patch-ng`
+`pip install -r requirement.txt` + The dictionary is automatically created in training and validation mode by using all words contained in the IAM dataset (i.e. also including words from validation set) and is saved into the file `data/corpus.txt`. Further, the manually created list of word-characters can be found in the file `model/wordCharList.txt`. Beam width is set to 50 to conform with the beam width of vanilla beam search decoding.