-
我使用4张2080ti（11G）卡训练，batch_size_per_card只能设置成16，但是每张gpu显存占用才3432M，batch_size_per_card设置成32就报错：
-------------------------------------- …
-
## Description
A clean expo-dev-client app crashes on startup in android simulator (working fine on iOS)
this issue occurs on version 2.3.0 and above and it's OK on lower versions
### E…
-
您好，问题如标题。下面是启动训练后生成的配置文件： （使用的是rec_icdar15_train.yml，其中name:Cosine，是参考配置文件的参数添加的）
> Global:
> debug: false
> use_gpu: true
> epoch_num: 100
> log_smooth_window: 20
> print_batch_step…
-
Hi,
so I trained Kaldi using your (old) s5 script, and as a sanity check I tried to decode the training data.
When I compared the text file from the training data to my results, I noticed that there…
-
I just trained JSUT and finished Training
I try to decode one wav file for testing purpose
But it just start downloading tedlium2.transformer.v1
How do I decode using trained JSUT?
*Trained Mode…
-
Hi, I am newly in speechbrain. Since I want to train a ASR model in different language, I can not use pretrained models; thus I have to train my model from start (Tokenizer, LM, ASR). In order to lear…
-
请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem
- 系统环境/System Environment：Ubuntu 16.04.6 LTS
- 版本号/Version：Paddle 2.2.1： PaddleOCR latest： 问题相关组件/Related …
-
您好，我使用的rec_r34_vd_none_bilstm_ctc.yml配置在V100上进行训练，数据集为360W的《中文文档文字识别》。训练了100个epoch，acc仍未0.006左右，请问一下这种情况正常吗？还有没有比较继续训练？
-
## Bug Description
Hit two compilation bugs with an unreleased version of pytorch (commit compiled linked below) on an almost certainly "unsupported" configuration. Figured I'd share the changes …
-
#2470 问题下已经逐条排查过了，github上能搜到的#2080 #2782 #2873也都排查了
参数已对齐，检测还算准确，但是识别有很大问题
还有什么需要排查的点吗
举例来说：
python tools/infer/predict_system.py --image_dir="./try2/" --det_model_dir="./inference/db_mv3" --re…