-
Make sure
datafolder in the same directory of the notebooks. -
Run any notebook using Jupyter Notebook.
Based on 20% validation, time taken based on single Tesla V100 32GB VRAM.
| name | accuracy | time taken (s) |
|---|---|---|
| 1. basic-rnn | 0.68 | 1.3219 |
| 2. basic-rnn-hinge | 0.65 | 1.2455 |
| 3. basic-rnn-huber | 0.68 | 1.2468 |
| 4. basic-rnn-bidirectional | 0.71 | 3.8174 |
| 5. basic-rnn-bidirectional-hinge | 0.68 | 2.5127 |
| 6. basic-rnn-bidirectional-huber | 0.63 | 3.5095 |
| 7. lstm-rnn | 0.73 | 2.69683 |
| 8. lstm-rnn-hinge | 0.72 | 8.2088 |
| 9. lstm-rnn-huber | 0.73 | 10.1754 |
| 10. lstm-rnn-bidirectional | 0.71 | 11.0388 |
| 11. lstm-rnn-bidirectional-huber | 0.71 | 5.5258 |
| 12. lstm-rnn-dropout-l2 | 0.74 | 3.2420 |
| 13. gru-rnn | 0.72 | 3.16123 |
| 14. gru-rnn-hinge | 0.72 | 6.71951 |
| 15. gru-rnn-huber | 0.70 | 7.93373 |
| 16. gru-rnn-bidirectional | 0.73 | 2.91590 |
| 17. gru-rnn-bidirectional-hinge | 0.72 | 5.66385 |
| 18. gru-rnn-bidirectional-huber | 0.70 | 18.01133 |
| 19. lstm-cnn-rnn | 0.65 | 4.42849 |
| 20. kmax-cnn | 0.73 | 18.89667 |
| 21. lstm-cnn-rnn-highway | 0.68 | 3.23122 |
| 22. lstm-rnn-attention | 0.75 | 13.97496 |
| 23. dilated-rnn-lstm | 0.25 | 24.54002 |
| 24. lnlstm-rnn | 0.68 | 24.86363 |
| 25. only-attention | 0.74 | 2.63291 |
| 26. multihead-attention | 0.69 | 9.033228 |
| 27. neural-turing-machine | ||
| 28. lstm-seq2seq | 0.72 | 9.63291 |
| 29. lstm-seq2seq-luong | ||
| 30. lstm-seq2seq-bahdanau | ||
| 31. lstm-seq2seq-beam | ||
| 32. lstm-seq2seq-birnn | ||
| 33. pointer-net | ||
| 34. lstm-rnn-bahdanau | 0.71 | 9.81993 |
| 35. lstm-rnn-luong | 0.66 | 27.73932 |
| 36. lstm-rnn-bahdanau-luong | 0.69 | 36.97628 |
| 37. lstm-birnn-bahdanau-luong | 0.70 | 38.86009 |
| 38. bytenet | ||
| 39. fast-slow-lstm | ||
| 40. siamese-network | 0.52 | 7.13535 |
| 41. estimator | ||
| 42. capsule-rnn-lstm | ||
| 43. capsule-seq2seq-lstm | ||
| 44. capsule-birrn-seq2seq-lstm | ||
| 45. nested-lstm | ||
| 46. lstm-seq2seq-highway | ||
| 47. triplet-loss-lstm | 0.50 | |
| 48. dnc | 0.68 | 85.98529 |
| 49. convlstm | 0.69 | 2.66726 |
| 50. temporalconvd | 0.66 | 11.90590 |
| 51. batch-all-triplet-loss-lstm | 0.70 | |
| 52. fast-text | 0.76 | 0.49499 |
| 53. gated-convolution-network | 0.67 | 3.37712 |
| 54. simple-recurrent-units | 0.65 | 3.12624 |
| 55. lstm-han | 0.50 | 3.47965 |
| 56. bert | 0.73 | 6.31015 |
| 57. dynamic-memory-network | 0.71 | 3.25820 |
| 58. entity-network | 0.74 | 1.10458 |
| 59. memory-network | 0.58 | 1.157306 |
| 60. char-sparse | 0.76 | 2.350096 |
| 61. residual-network | 0.72 | 9.557085 |
| 62. residual-network-bahdanau | 0.71 | 11.53799 |
| 63. deep-pyramid-cnn | 0.68 | 6.980528 |
| 64. transformer-xl | 0.51 | 38.66338 |
| 65. transfer-learning-gpt2 | 0.79 | 178.0716 |
| 66. quasi-rnn | 0.66 | 166.1456 |
| 67. tacotron | 0.74 | 360.5551 |
| 68. slice-gru | 0.72 | 10.140633 |
| 69. slice-gru-bahdanau | 0.70 | 20.247409 |
| 70. wavenet | 0.59 | 101.293274 |
| 71. transfer-learning-bert | 0.81 | 887.590460 |
| 72. transfer-learning-xlnet-large | 0.846 | 340.7679 |
| 73. lstm-birnn-max-avg | 0.7552 | 9.35624 |
| 74. transfer-learning-bert-base-6 | 0.7655 | 494.169 |
| 75. transfer-learning-bert-large-12 | 0.80 | 1365.30 |
| 76. transfer-learning-xlnet-base | 0.820441 | 240.262 |
| 77. transfer-learning-albert-base | 0.799053 | 61.8179 |
| 78. transfer-learning-electra-base | 0.836336 | 66.0257 |
| 79. transfer-learning-electra-large | 0.875248 | 195.37280 |