tsc
https://github.com/aymericdamien/TensorFlow-Examples/blob/9e1bb504f5d0f209d000997ce2ad95bb891798ab/examples/3_NeuralNetworks/bidirectional_rnn.py#L81 I think "outputs [-1]" and "outputs [0]" are equivalent (reversed) in this line of code, but the former (89%) works better than the latter (86%). Why?
Why do you provide a model that can reach 100% top@1 in the training set?
The word embedding in this code does not seem to be obtained by word2vec? The parameter settings in the code are also different from the paper? Where is the ATTENTION...
I configured proxychains4, but these lines of code doesn't work: ```shell proxychains4 python -c "import os;os.system('curl baidu.com')" proxychains4 apt update proxychains4 pip install faker ``` However, the results of running...