allenai/bi-att-flow

View on GitHub

Bi-directional Attention Flow (BiDAF) network is a multi-stage hierarchical process that represents context at different levels of granularity and uses a bi-directional attention flow mechanism to achieve a query-aware context representation without early summarization.

Stars: 1.5k
Forks: 667
Open beginner issues: 0
Indexed issues: 66

Dominant language: Python
License: Apache-2.0
Last GitHub push: Jan 25, 2018
Latest indexed: Sep 13, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed
PR merge metrics: No merged PRs in 30d

Open All

66 open issues indexed

basic_cnn (vs basic, tree) Open

allenai/bi-att-flow#6 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Compatibility with tf-serving Open

allenai/bi-att-flow#11 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
how to predict answers for custom question and context by reusing loaded model. Open

allenai/bi-att-flow#13 · 33 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ForwardEvaluator Bug Open

allenai/bi-att-flow#18 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
how to handle large context by Machine Comprehension(bi-att-flow) &Resource Exhausted error Open

allenai/bi-att-flow#21 · 6 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
process_tokens() in utils.py Open

allenai/bi-att-flow#30 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
How to generate the save file in the save fold? Open

allenai/bi-att-flow#40 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Error when running this tensorflow-1.1.0 version Open

allenai/bi-att-flow#41 · 19 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Is there any document for this code? Open

allenai/bi-att-flow#42 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
KeyError: 'contextss' Open

allenai/bi-att-flow#43 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ValueError when running training with Multi-GPU Open

allenai/bi-att-flow#44 · 7 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
nltk tokenize doesn't work? Open

allenai/bi-att-flow#45 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
The difference between basic and basic_cnn? Open

allenai/bi-att-flow#48 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
3.1 Loading from pre-trained weights "download zip" Open

allenai/bi-att-flow#50 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Problem: sum of class instances when downgrading to python 2.7 Open

allenai/bi-att-flow#51 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Looking for tester - install and run routine bash script Open

allenai/bi-att-flow#52 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ImportError: /usr/local/lib/python3.4/dist-packages/tensorflow/python/_pywrap_tensorflow.so: invalid ELF header Open

allenai/bi-att-flow#53 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Variable model_1/loss/ExponentialMovingAverage/ does not exist Open

allenai/bi-att-flow#54 · 13 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
What does aug mean？ Open

allenai/bi-att-flow#56 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Why does the model work when turn 400 to 740 Open

allenai/bi-att-flow#57 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Use the pre-trained model to make new predictions Open

allenai/bi-att-flow#58 · 0 comments · 2 reactions · 0 assignees · Indexed Sep 13, 2026
CNN+SQuAD Open

allenai/bi-att-flow#59 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
f1 and em score is less than official result by 0.3% Open

allenai/bi-att-flow#60 · 5 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
What does softsel mean？ Open

allenai/bi-att-flow#62 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
how faster about the inference time? Open

allenai/bi-att-flow#63 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ValueError when setting finetune to True Open

allenai/bi-att-flow#64 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
confusion about var_ema Open

allenai/bi-att-flow#65 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
W tensorflow/core/framework/op_kernel.cc:1158] Not found: Key main/g0/fw/basic_lstm_cell/bias/ExponentialMovingAverage not found in checkpoint Open

allenai/bi-att-flow#66 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ce_loss with loss_mask however ce_loss2 without Open

allenai/bi-att-flow#67 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Dev branch fails with TF r1.4 Open

allenai/bi-att-flow#69 · 25 comments · 4 reactions · 0 assignees · Indexed Sep 13, 2026
FileNotFoundError: [Errno 2] No such file or directory: 'inter_single/eval.json' Open

allenai/bi-att-flow#70 · 4 comments · 2 reactions · 0 assignees · Indexed Sep 13, 2026
Predict empty answer with high score Open

allenai/bi-att-flow#71 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
How to train the model on just 500 Questions or less to run it on a CPU? Open

allenai/bi-att-flow#73 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Which Glove pre-trained weights were used to train basic-18000 Open

allenai/bi-att-flow#76 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Trained model for CNN\Daily Mail Open

allenai/bi-att-flow#79 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
output layer code does not match paper Open

allenai/bi-att-flow#80 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ImportError: cannot import name 'bidirectional_rnn' Open

allenai/bi-att-flow#81 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Training Bidaf on CNN dataset fails Open

allenai/bi-att-flow#83 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
cannot import name _linear in nn.py Open

allenai/bi-att-flow#84 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
The error comes up when running python -m basic.cli --mode train --noload Open

allenai/bi-att-flow#85 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Strange issue : OOV when testing, while training is solid Open

allenai/bi-att-flow#90 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
_linear() got an unexpected keyword argument 'bias_start' Open

allenai/bi-att-flow#93 · 8 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
absl.flags._exceptions.UnrecognizedFlagError: Unknown command line flag 'new_emb_mat' Open

allenai/bi-att-flow#96 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
run-demo.py doesn't exist Open

allenai/bi-att-flow#98 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
How do you get results for dev_data with squad official evaluator? Open

allenai/bi-att-flow#99 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Question: why use TensorFlow BasicLSTMCell instead of other kind of LSTM cell in TF? Open

allenai/bi-att-flow#100 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
Pre-trained weights (save.zip) are not available in the CodaLab worksheet Open

allenai/bi-att-flow#107 · 6 comments · 7 reactions · 0 assignees · Indexed Sep 13, 2026
ValueError: Variable highway/highway_network/layer_0/kernel already exists, disallowed. Did you mean to set reuse=True in VarScope? Open

allenai/bi-att-flow#108 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 13, 2026
ValueError: Dimensions must be equal, but are 900 and 300 for 'model_0/prepro/prepro/fw/fw/while/basic_lstm_cell/MatMul_6' (op: 'MatMul') with input shapes: [?,900], [300,400]. Open

allenai/bi-att-flow#109 · 3 comments · 2 reactions · 0 assignees · Indexed Sep 13, 2026
Question: Can not download the TQA data when I want to use the tqa branch Open

allenai/bi-att-flow#110 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 13, 2026

allenai/bi-att-flow

Get new issues in your inbox