hwy893747147
commented
2 years ago Thank you for your team's sharing. I think this is a very valuable project. I can run it successfully now, but there are still some details, mainly as follows Question a In the example of running "tods/examples/sk_examples/DeepLog_test.py" #print("Prediction Score\n", prediction_score) my run results are "Prediction Labels [[0] [0] [0] ... [0] [0] [0]] Prediction Score" Refer to the documentation "tods/examples/Demo Notebook/TODS Official Demo Notebook.ipynb" for the correct run result " [[0. ] [0.3569443 ] [0.3569443 ] ... [0.77054234] [0.4575615 ] [0.17499346]]" Question b In the example of running "tods/examples/sk_examples/DeepLog_test.py" my run results are "d3m.primitives.tods.detection_algorithm.LSTMODetector: Primitive is not providing a description through its docstring." Refer to the documentation "tods/examples/Demo Notebook/TODS Official Demo Notebook.ipynb" for the correct run result "Primitive: d3m.primitives.tods.detection_algorithm.telemanom(hyperparams=Hyperparams({'contamination': 0.1, 'window_size': 1, 'step_size': 1, 'return_subseq_inds': False, 'use_columns': (), 'exclude_columns': (), 'return_result': 'new', 'use_semantic_types': False, 'add_index_columns': False, 'error_on_no_input': True, 'return_semantic_type': 'https://metadata.datadrivendiscovery.org/types/Attribute', 'smoothing_perc': 0.05, 'windowsize': 100, 'error_buffer': 50, 'batch_size': 70, 'dropout': 0.3, 'validation_split': 0.2, 'optimizer': 'Adam', 'lstm_batch_size': 64, 'loss_metric': 'mean_squared_error', 'layers': [10, 10], 'epochs': 1, 'patience': 10, 'min_delta': 0.0003, 'l_s': 2, 'n_predictions': 1, 'p': 0.05}), random_seed=0)" Thank you very much for your selfless help
Hello, thank you very much for using our project, for question A, are you getting no output at all for the Prediction Scores? For question B, I think this could be an installation issue with d3m. Without knowing enough details on your testing environment I would not be able to fix it at the moment, however, I would like to invite you to check out our latest dev branch and see if it has been updated after you cloned our project. Furthermore, we have recently implemented the example notebooks on Google Colab and is now available to use for everyone, you can simply run the notebook without configuring a local environment. It can be found at HERE. Please feel free to test it out and let us know of any new errors. Thank you very much!
Thank you for your team's sharing. I think this is a very valuable project. I can run it successfully now, but there are still some details, mainly as follows Question a In the example of running "tods/examples/sk_examples/DeepLog_test.py"
print("Prediction Score\n", prediction_score)
my run results are "Prediction Labels [[0] [0] [0] ... [0] [0] [0]] Prediction Score" Refer to the documentation "tods/examples/Demo Notebook/TODS Official Demo Notebook.ipynb" for the correct run result " [[0. ] [0.3569443 ] [0.3569443 ] ... [0.77054234] [0.4575615 ] [0.17499346]]" Question b In the example of running "tods/examples/sk_examples/DeepLog_test.py" my run results are "d3m.primitives.tods.detection_algorithm.LSTMODetector: Primitive is not providing a description through its docstring." Refer to the documentation "tods/examples/Demo Notebook/TODS Official Demo Notebook.ipynb" for the correct run result "Primitive: d3m.primitives.tods.detection_algorithm.telemanom(hyperparams=Hyperparams({'contamination': 0.1, 'window_size': 1, 'step_size': 1, 'return_subseq_inds': False, 'use_columns': (), 'exclude_columns': (), 'return_result': 'new', 'use_semantic_types': False, 'add_index_columns': False, 'error_on_no_input': True, 'return_semantic_type': 'https://metadata.datadrivendiscovery.org/types/Attribute', 'smoothing_perc': 0.05, 'windowsize': 100, 'error_buffer': 50, 'batch_size': 70, 'dropout': 0.3, 'validation_split': 0.2, 'optimizer': 'Adam', 'lstm_batch_size': 64, 'loss_metric': 'mean_squared_error', 'layers': [10, 10], 'epochs': 1, 'patience': 10, 'min_delta': 0.0003, 'l_s': 2, 'n_predictions': 1, 'p': 0.05}), random_seed=0)" Thank you very much for your selfless help