onnxruntime推理失败-stage6

[zapjone]

执行examples/csmsc下的run.sh --stage 6 --stop-stage 6来进行onnxruntime推理时（stage0~stage5都是已成功执行完成），报以下错误，请问有谁遇到过吗？ [2023-06-14 08:23:03,244] [ INFO] - Already cached /home/icksys/.paddlenlp/models/bert-base-chinese/bert-base-chinese-vocab.txt [2023-06-14 08:23:03,254] [ INFO] - tokenizer config file saved in /home/icksys/.paddlenlp/models/bert-base-chinese/tokenizer_config.json [2023-06-14 08:23:03,254] [ INFO] - Special tokens file saved in /home/icksys/.paddlenlp/models/bert-base-chinese/special_tokens_map.json Building prefix dict from the default dictionary ... [2023-06-14 08:23:03,932] [ DEBUG] init.py:113 - Building prefix dict from the default dictionary ... Loading model from cache /tmp/jieba.cache [2023-06-14 08:23:03,933] [ DEBUG] init.py:132 - Loading model from cache /tmp/jieba.cache Loading model cost 0.624 seconds. [2023-06-14 08:23:04,556] [ DEBUG] init.py:164 - Loading model cost 0.624 seconds. Prefix dict has been built successfully. [2023-06-14 08:23:04,556] [ DEBUG] init.py:166 - Prefix dict has been built successfully. 2023-06-14 08:23:04.586965864 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Gather node. Name:'p2o.Gather.0' Status Message: indices element out of data bounds, idx=216 must be within the inclusive range [-208,207] Traceback (most recent call last): File "/home/icksys/tts/PaddleSpeech/paddlespeech/t2s/exps/fastspeech2/../ort_predict_e2e.py", line 235, in main() File "/home/icksys/tts/PaddleSpeech/paddlespeech/t2s/exps/fastspeech2/../ort_predict_e2e.py", line 231, in main ort_predict(args) File "/home/icksys/tts/PaddleSpeech/paddlespeech/t2s/exps/fastspeech2/../ort_predict_e2e.py", line 86, in ort_predict am_sess.run(None, input_feed=am_input_feed) File "/home/icksys/anaconda3/envs/paddle/lib/python3.9/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Non-zero status code returned while running Gather node. Name:'p2o.Gather.0' Status Message: indices element out of data bounds, idx=216 must be within the inclusive range [-208,207]

[zxcd]

目测是因为你训练的时候的配置和使用onnx推理的时候配置不一致导致的。 indices element out of data bounds, idx=216 must be within the inclusive range [-208,207] 建议检查一下

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[wmlgl]

Reference

从stage 0一直执行到后面配置没有修改过哟，不过我出现问题的是训练英文模型的时候，非onnx是正常的，转成onnx就报错了