STT
STT copied to clipboard
Bug: lm_optimize fails due to check failing.
Welcome to the 🐸STT project! We are excited to see your interest, and appreciate your support!
This repository is governed by the Contributor Covenant Code of Conduct. For more details, see the CODE_OF_CONDUCT.md file.
If you've found a bug, please provide the following information:
Describe the bug When attempting to run the optimizer python script to determine the best alpha and beta for a customer scorer, it errors with the following output on an M1 Macbook 2020 running macOS Monterey 12.5.1.
python3 lm_optimizer.py
2022-09-30 12:37:43.198725: F tensorflow/core/lib/monitoring/sampler.cc:42] Check failed: bucket_limits_[i] > bucket_limits_[i - 1] (0 vs. 10)
qemu: uncaught target signal 6 (Aborted) - core dumped
Aborted
To Reproduce Steps to reproduce the behavior:
- Run
docker pull ghcr.io/coqui-ai/stt-train:v0.10.0-alpha.4
- Run
docker run -it --name stt-test --entrypoint /bin/bash ghcr.io/coqui-ai/stt-train:v0.10.0-alpha.4
- Make file at /code/data/text.txt with following data
up
down
left
right
velocity
distance
forward
backward
faster
slower
stop
start
take off
land
launch
- Run
python3 generate_lm.py --input_txt /code/data/text.txt --output_dir /code/data/ --top_k 500000 --kenlm_bins /code/kenlm/build/bin/ --arpa_order 5 --max_arpa_memory "85%" --arpa_prune "0|0|1" --binary_a_bits 255 --binary_q_bits 8 --binary_type trie --discount_fallback
- Run
python3 lm_optimizer.py
- See error
Expected behavior To receive an optimized alpha and beta
Environment (please complete the following information):
-
OS Platform and Distribution (e.g., Linux Ubuntu 16.04): macOS Monterey 12.5.1
-
TensorFlow installed from (our builds, or upstream TensorFlow): Whatever build that is within the docker
-
TensorFlow version (use command below): Whatever that is within the docker
-
Python version: Whatever that is within the docker
-
Bazel version (if compiling from source): Whatever that is within the docker
-
GCC/Compiler version (if compiling from source): Whatever that is within the docker
-
CUDA/cuDNN version: Whatever that is within the docker
-
GPU model and memory: No GPU being utilized
-
Exact command to reproduce:
Additional context Add any other context about the problem here.