2020-dl-training-program / lab3-image-caption Goto Github PK
View Code? Open in Web Editor NEWLicense: MIT License
License: MIT License
助教您好:
想請問一下我在執行$ python build_vocab.py之後會顯示 Resource punkt not found
但當我執行以下的command之後
import nltk
nltk.download('punkt')
會顯示 "Error loading punkt : <urlopen error [Error 110] Connection timed out"
應該是工作站在連網上有一定限制的關係
想請問一下手動下載punkt比較詳細的步驟?
感謝
助教您好
請問在DIY_LSTM.py裡,forward的部分,需要檢查input_seq是不是一個packed的sequence呢
這邊說的packed 是 pack_padded_sequence 做的那件事嗎
助教Report中的Demo是隨便輸入自己給的圖片嗎
然後demo中的question要寫什麼呢??
助教您好,
我在執行python3 train.py的時候會遇到 undefined symbol: _Py_ZeroStruct這樣的Error
看了以下的網頁我試著把Readme裡面第一步的makefile裡面的python改成python3
然後執行了步驟一
但還是會遇到相同的問題
想請問一下助教這樣的問題有沒有排解的方法?
Hi 助教您好,
想請問一下關於我們需要實作的 LSTM 的輸出格式,
由於範例 code 所呼叫的 nn.LSTM 傳進去的是 packed sequence,
故 output 至下一層的 dense layer 時也是 packed sequence 的 shape (即非 [B, T, C] or [T, B, C] 的 shape)
因此想請問我們自己實作的 LSTM, 按照 comment 裡的 TO-DO,
流程上應該會先 unpack packed sequence,
再透過 nn.Linear 操作完 LSTM 的運算再輸出
但是這樣的流程下, 最後輸出並不會是原本 sample 直接 run nn.LSTM 的輸出 type (也就是 packed sequence),
導致後續在計算 loss 的時候會有 Error 產生, 那這部份有需要將 LSTM 輸出的結果再轉回 packed sequence 嗎?
因為如果將輸出再轉回去 packed sequence, loss 雖然會下降的跟 sample code 接近, 但 testing 的結果會很奇怪,
不確定是否輸出完又轉回 packed sequence 會讓 output 輸出意義不同, 所以也想詢問這部分是否有需要做其他調整...
(例如調整 target 的 type or shape)
謝謝~
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.