[Hackathon 7th] 修复 panns 中 predict.py 对于 pir 的 json 模型路径
#3914
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
Update 20241127模型推理的问题已经解决， |
zxcd
reviewed
Nov 28, 2024
| @@ -74,12 +75,18 @@ def __init__(self, | |||
| self.batch_size = batch_size | |||
|
|
|||
| model_file = os.path.join(model_dir, "inference.pdmodel") | |||
| if not os.path.exists(model_file): | |||
There was a problem hiding this comment.
建议直接判断是否有.json文件，如果有model_file=.json，没有model_file=.pdmodel，报错统一由85行来
zxcd
reviewed
Dec 3, 2024
| @@ -73,13 +74,18 @@ def __init__(self, | |||
| enable_mkldnn=False): | |||
| self.batch_size = batch_size | |||
|
|
|||
| model_file = os.path.join(model_dir, "inference.pdmodel") | |||
| if os.path.exists(os.path.join(model_dir, "inference.json")): | |||
|
@zxcd 关于 PR 里面去掉 另外，这里貌似也不需要判断 feat = paddle.transpose(feat, perm=[1, 0])
if feat.dim() == 1:
feat = feat.unsqueeze(0)如果 feat 可以 transpose 的话，那么 dim 不可能为 1，所以加这个判断也没啥用 ... ... |
zxcd
reviewed
Dec 4, 2024
| @@ -55,8 +58,7 @@ def extract_features(files: str, **kwargs): | |||
|
|
|||
| feature_extractor = LogMelSpectrogram(sr, **kwargs) | |||
| feat = feature_extractor(paddle.to_tensor(waveforms[i])) | |||
| feat = paddle.transpose(feat, perm=[1, 0]).unsqueeze(0) | |||
|
|
|||
| feat = paddle.transpose(feat, perm=[1, 0]) | |||
There was a problem hiding this comment.
single audio still need unsqueeze, suggest adding an if judgment.
There was a problem hiding this comment.
这里应该不需要 unsqueeze ~ 后面代码用的是 np.stack ，而 np.stack 会自动在 axis=0 加一个维度，也就是咱们需要的 batch ～ 如果这里加了 unsqueeze ，那后面就需要用 np.vstack ，不然会多出来一个维度 ～
>>> import numpy as np
>>> a = np.random.rand(2, 3)
>>> b = [a]
>>> np.stack(b, axis=0)
array([[[0.67850623, 0.57210335, 0.21218978],
[0.10639948, 0.49831181, 0.45858706]]])
>>> np.stack(b, axis=0).shape
(1, 2, 3)
>>> b = [a, a]
>>> np.stack(b, axis=0).shape
(2, 2, 3)
>>> np.vstack(b)
array([[0.67850623, 0.57210335, 0.21218978],
[0.10639948, 0.49831181, 0.45858706],
[0.67850623, 0.57210335, 0.21218978],
[0.10639948, 0.49831181, 0.45858706]])
>>> np.vstack(b).shape
(4, 3)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
Bug fixes
PR changes
Others
Describe
修复
panns中predict.py对于 pir 的 json 模型路径这里修复了两个问题：
paddle（这个文件当时运行过吗？？？）注意：
PaddleSpeech/examples/esc50测试$ CUDA_VISIBLE_DEVICES=0 ./run.sh 4 cpu ./export ~/datasets/5-9032-A-0.wav过程中发现此问题，但是，修改了这里的两个问题之后，模型仍然不能进行推理，仍需定位问题 ～
@zxcd @Liyulingyue