search

Conchylicultor / DeepQA

My tensorflow implementation of "A neural conversational model", a Deep learning based chatbot
Apache License 2.0
2.93k stars 1.17k forks source link

repetition of words in preTrainedv2 when using Readme Instructions #139

Closed EMCP closed 7 years ago

EMCP commented 7 years ago

I'm trying to make sure I understand what's going on here..

I downloaded the sample zip, placked the .pkl file in the aformentioned folder.. and executed the following

~/git/Conchylicultor/DeepQA (master)$ ./main.py --modelTag pretrainedv2 --test interactive
Welcome to DeepQA v0.1 !

TensorFlow detected: v1.2.1
Training samples not found. Creating dataset...
Constructing full dataset...
Extract conversations: 100%|████████████████████████████████████████████████████████████████████| 83097/83097 [02:17<00:00, 605.58it/s]
Loaded cornell: 59755 words, 221282 QA
Filtering words (vocabSize = 40000 and wordCount > 1)...
Saving dataset...                                                                                                                      
Loaded cornell: 24643 words, 159657 QA
Model creation...
2017-08-10 20:22:52.742656: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2017-08-10 20:22:52.742677: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2017-08-10 20:22:52.742692: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2017-08-10 20:22:52.742696: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
Initialize variables...
WARNING: No previous model found, starting from clean directory: ~/git/Conchylicultor/DeepQA/save/model-pretrainedv2
Testing: Launch interactive mode:

Welcome to the interactive mode, here you can ask to Deep Q&A the sentence you want. Don't have high expectation. Type 'exit' or just press ENTER to quit the program. Have fun.
Q: hi who is this?
A: Scrunchie scrunchie scrunchie alias alias alias supper supper supper supper supper supper

Q: are you alright?
A: Schmidt talkative alias alias alias alias alias alias alias alias alias waterbed

Q: are you trained?
A: But.. alias alias alias alias alias alias alias alias alias jefferson jefferson

Did I do something wrong ? Feels like I need to run the command differently.. I've noticed it does this when I turn on interactive.. still grappling with why. Seems if you don't use interactive, you go through and repetitively train .. but how do you interact with the better trained version ?

This run starts to cook my Macbook Air so I stop it.. but I'm guessing I could run it in a toned down docker container?

~/git/Conchylicultor/DeepQA (master)$ ./main.py --modelTag pretrainedv2 
Welcome to DeepQA v0.1 !

TensorFlow detected: v1.2.1
Loading dataset from /Users/emcp/Dev/git/Conchylicultor/DeepQA/data/samples/dataset-cornell-length10-filter1-vocabSize40000.pkl
Loaded cornell: 24643 words, 159657 QA
Model creation...
2017-08-10 20:39:55.432684: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2017-08-10 20:39:55.432710: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2017-08-10 20:39:55.432715: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2017-08-10 20:39:55.432719: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
Initialize variables...
WARNING: No previous model found, starting from clean directory: ~/git/Conchylicultor/DeepQA/save/model-pretrainedv2
Start training (press Ctrl+C to save and exit)...

----- Epoch 1/30 ; (lr=0.002) -----
Shuffling the dataset...
Training:   1%|▌                                                                                     | 4/624 [00:25<1:05:31,  6.34s/it]
EMCP commented 7 years ago

oh nevermind.. I just realized I needed to unzip the entire contents into /DeepQA/save/model-pretrainedv2/

Now I'm getting the error just like #120