DeepMol
DeepMol copied to clipboard
BioSystemsUM
DeepMol: A Machine and Deep Learning Framework for Computational Chemistry
saving and loading models pose problems when there are dots in the absolute path of the model
When computing non-tabular features, like one-hot-encoding, the `to_csv` method in the `Dataset` class fails. I think this is the correct behavior as pandas can only save tabular data to CSV...
Sometimes when fitting the Tabulat Transformer model it fails and raises: ### Node: 'model_1/embedding_1/embedding_lookup' indices[24,5] = 222 is not in [0, 208) [[{{node model_1/embedding_1/embedding_lookup}}]] [Op:__inference_train_function_9384] tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,...
Add a `leaderboard` property or method to get a table with all (or top x) results of the pipeline optimization procedure. This table should contain the steps, respective parameters, and...
Add splitter to hyperparameter optimization with cross-validation
"robust_multitask_classifier_model" works well except when loading the model back in (probably related with the custom_objects) (SAME WITH robust_multitask_regressor_model)
"sc_score_model" not working (Input 0 of layer "dense" is incompatible with the layer: expected axis -1 of input shape to have value 2048, but received input with shape (100, 1))
"progressive_multitask_classifier_model" not working (ValueError: Index out of range using input dim 1; input has only 1 dims for '{{node strided_slice_1}} ...) (SAME WITH progressive_multitask_classifier_model) in [src/deepmol/pipeline_optimization/_utils.py](https://github.com/BioSystemsUM/DeepMol/pull/86/files/9d4e48237192ac59fdffb1a745bea1f942247bea#diff-351e8e34b992507ca33b45dfef244c025c428a24c3a031239946fa2102a5e6bb)
"megnet_model" is not working (error with torch_geometric (extra_requirement))
"multitask_irv_classifier_model" not working (needs 1D featurizer + irv transformer but it is not working) in [src/deepmol/pipeline_optimization/_utils.py](https://github.com/BioSystemsUM/DeepMol/pull/86/files/9d4e48237192ac59fdffb1a745bea1f942247bea#diff-351e8e34b992507ca33b45dfef244c025c428a24c3a031239946fa2102a5e6bb)
Metadata
Owner
Metadata
DeepMol: A Machine and Deep Learning Framework for Computational Chemistry