logs17:RL test len equals 2 is the best

Log Type	Detail
1: What specific output am I working on right now?	Follow up on logs16:RL test shorter reply is better. See if RL is working.
2: Thinking out loud - hypotheses about the current problem - what to work on next - how can I verify	Run this and see if it converges avg_len == 2
3: A record of currently ongoing runs along with a short reminder of what question each run is supposed to answer	should
4: Results of runs and conclusion	It got reward=-1 for all input and leads 0 standardized reward. So the training didn't go anywhere. 💩💩💩💩💩💩💩💩💩💩💩💩💩💩💩💩💩💩ますますます！！！！💩💩💩
5: Next steps	I think -1 reward is really bad. It should be a bit random.
6: mega.nz	rl_test_20180517111211

RL

hparam| src: {'machine': 'client1', 'batch_size': 64, 'num_units': 512, 'num_layers': 2, 'vocab_size': 5000, 'embedding_size': 256, 'learning_rate': 0.5, 'learning_rate_decay': 0.99, 'use_attention': True, 'encoder_length': 28, 'decoder_length': 28, 'max_gradient_norm': 5.0, 'beam_width': 0, 'num_train_steps': 22, 'model_path': 'model/tweet_large'} dst {'machine': 'client1', 'batch_size': 64, 'num_units': 512, 'num_layers': 2, 'vocab_size': 5000, 'embedding_size': 256, 'learning_rate': 0.1, 'learning_rate_decay': 0.99, 'use_attention': True, 'encoder_length': 28, 'decoder_length': 28, 'max_gradient_norm': 5.0, 'beam_width': 0, 'num_train_steps': 1560, 'model_path': 'model/tweet_large_rl'}