From 62d4c24d69e074bf503a9772c29ef9ed798e161b Mon Sep 17 00:00:00 2001
From: dohe0342 <kimdohe1070@gmail.com>
Date: Tue, 11 Apr 2023 15:54:47 +0900
Subject: [PATCH] from local

---
 .../.data2vec_audio.py.swp                    | Bin 40960 -> 40960 bytes
 .../data2vec_audio.py                         |   6 ------
 2 files changed, 6 deletions(-)

diff --git a/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/.data2vec_audio.py.swp b/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/.data2vec_audio.py.swp
index 66d3265fab6da22401bd87291d5af70ae2b423b4..817e39bf52020fd1c8a20dd9bb136c3053fb741f 100644
GIT binary patch
delta 206
zcmZoTz|?SnNi@kI%+puFQqO<^2m}}y_*qO-ayE)q3otrwo+wbP&-;vpfq{<=BI!I?
z(BZkQ2L}VgX?6yNDj==|Vjm#(24Zy}ehgGO2Z%#~7^LehD+9wRAYKl{l|US{+0;>%
zWs;jLmvO9ut*t^xQE4iWH`&8g5J<5>1W{NlATglGW)IgG#>w_-+>@`lxlb-~ub#~1
P$30ov!+7%~kGyaI;?*vp

delta 287
zcmZoTz|?SnNi@kI%+puFQqO<^2m}}y*jY?dTsMkV3ov?ao+wbP&-<F0fkB%UBI!9<
z(BZjlG6w^L0}!(S@k4e7h7=$+17dz4-p9tkFdc}e0dXb}X8^Ga5HkSrPM|48n@t^M
zStc)VW#4S$=FK?S&|PY>k($KhN$&2Gjoc-56p%ncQD$C=MnO@2Zb3<WL1IcuW?p)H
zZenq^UU6nss)pv|lkVMO5JhRJi6x~)sl`wk&B=d#B__vv7@L4hK$m1dvO@vc5~%LY
IO&)dO08JoEH2?qr

diff --git a/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py b/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py
index ad9222398..e6dfff769 100644
--- a/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py
+++ b/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py
@@ -492,15 +492,9 @@ class Data2VecAudioModel(BaseFairseqModel):
         if prompt is not None:
             #features = torch.cat([features, prompt])
             prompt = prompt.expand((features.size()[0], prompt.size()[0], prompt.size()[1]))
-            print(prompt.size())
-            print(features.size())
             features = torch.cat([prompt, features], dim=1)
-            print(features.size())
             prompt_padding_mask = torch.zeros(prompt.size()[0], prompt.size()[1]).type(torch.BoolTensor).to(features.device)
-            print(prompt_padding_mask.size())
-            print(padding_mask.size())
             padding_mask = torch.cat([prompt_padding_mask, padding_mask], dim=1)
-            print(padding_mask.size())
 
         features = self.layer_norm(features)