XinhaoMei
XinhaoMei
same question!
> Hi, I am getting this error: `FileNotFoundError: [Errno 2] No such file or directory: 'data/Clotho/pickles/456/train_keywords_dict_pred_5.p'` > > what could cause this ? there is no train_keywords_dict_pred_x.p file. Is it...
> We use the training data to get the vocabulary, which contains all possible words the model can generate. At each time step, we get a logit from the model...
Hi, do you mean cross-entropy training for the first step? The default setting is using PANNs as encoder and a two-layer Transformer as decoder and training on Clotho. You can...
You are welcome. By the way, the [ACT](https://github.com/XinhaoMei/ACT) used AudioCaps, and I uploaded the dataset in that repository. You can have a look at it. Thanks!
Hi, you could try something like: ``zip -F train.zip --out ac_train.zip`` ``unzip ac_train.zip``
have you also downloaded other zip files, like train.z01 to train.z04?
Hello, To solve the out of memory issue, you could try to reduce the batch size. Of coure, running on a large gpu memory would be better. We did not...
Hello, since you've made significant changes to the original setup, it's challenging for me to pinpoint the exact reason for the discrepancies in results. So sorry for this.
Softmax is applied in self.loss().在 2023年9月2日，21:40，Vedanshi Shah ***@***.***> 写道： Could you explain the significance of mask in the NT-Xent loss function? mask = labels.expand(n, n).eq(labels.expand(n, n).t()).to(a2t.device) mask_diag = mask.diag() mask_diag...