deepchem | Democratizing Deep-Learning for Drug Discovery | Machine Learning library
kandi X-RAY | deepchem Summary
kandi X-RAY | deepchem Summary
Website | Documentation | Colab Tutorial | Discussion Forum | Gitter. DeepChem aims to provide a high quality open-source toolchain that democratizes the use of deep-learning in drug discovery, materials science, quantum chemistry, and biology.
Support
Quality
Security
License
Reuse
Top functions reviewed by kandi - BETA
- Benchmark classification
- Reshard data to new shards
- Create a dataset
- Construct a pandas dataframe
- Perform hyperparameters search
- Runs the model
- Compute the range of the optimization parameters
- Convert a dictionary to a filename
- Install miniconda
- Predict for the given generator
- Generates Pose objects for a molecule
- Create sluice model
- Load USPTO
- Featurize a molecule
- Construct a feed dictionary
- Loads the PDBBind coordinates for the given fragment
- Compute the metric
- Load PDBBIND dataset
- Load QM9 dataset
- Load a Zinc15 dataset
- Connects the graph
- Train the model
- Interpolate a sequence of motifs and deepLifts
- Generate a set of Pose objects
- Loads theroiterberg ANI format
- Build the graph
deepchem Key Features
deepchem Examples and Code Snippets
Community Discussions
Trending Discussions on deepchem
QUESTION
import deepchem as dc
import pandas as pd
import numpy as np
import os, glob
tasks, datasets, transformers = dc.molnet.load_hiv(featurizer='GraphConv')
train_dataset, valid_dataset, test_dataset = datasets
print(datasets)
n_tasks = len(tasks)
model = dc.models.GraphConvModel(n_tasks, mode='classification')
hist = model.fit(train_dataset, nb_epoch=50)
metric = dc.metrics.Metric(dc.metrics.roc_auc_score)
print('Training set score:', model.evaluate(train_dataset, [metric], transformers))
print('Test set score:', model.evaluate(test_dataset, [metric], transformers))
model.save("HIV_test1.h5")
...ANSWER
Answered 2021-May-24 at 15:48The docs for deepchem model.save()
don't actually say much. But you can't provide a filename to it (it takes no additional arguments so implicitly, just the model
).
It turns out that in order to "save", you need to specify a model_dir
when initialising your model. I haven't checked but apparently your model will be saved to that location after certain steps, like .fit()
.
Using this example and applying to your code:
QUESTION
I am using Deepchem to create features for the my GraphConvolution model as follows.
...ANSWER
Answered 2019-Oct-08 at 09:50Why not look at the source code for the ConvMol object?
The output of the featurizer returns an array of ConvMol objects (one for each rdkit molecule input) i.e. deepchem.feat.mol_graphs.ConvMol, what you actually want to inspect is the first element of the array, mol_object[0].
Looking at the source code you can then understand what information about the molecule is contained such as the atom features which can be accessed ConvMol.atom_features, or in your case mol_object[0].atom_features
QUESTION
I am using Deepchem wrapper for GraphConvolution model as follows. I have my smiles data in .csv
which consists of 5 molecules with their smiles representation and their respective activities. The data can be accessed from here directly.
Importing the libraries:
...ANSWER
Answered 2019-Oct-10 at 11:33As per your previous question dataset_train.X is an array of ConvMol objects. These ConvMol objects are a container for the features of each of your input molecules. The features are not represented like they are for your targets 'train_dataset.y' as they are more complex graph features. look at the source code for the ConvMol object again and look at the source code for the ConvMolFeaturizer. You can then determine how you want to interpret these features:
QUESTION
I have a simple CNN model written in the tf.keras framework, which I wish to use with variable input size.
According to this "documentation" I can use variable input size by setting input_shape=(None, None, n_channels)
, and I have used a GlobalMaxPooling2D
layer before my dense layer to standardize the input to the dense layer.
Yet when I train the model with one size of image and try to predict on a different size I get the error:
...ANSWER
Answered 2019-Mar-01 at 14:08The input shape should be in the first layer of your model, but you are putting it in the second. So Keras is assuming a shape from your training data.
Community Discussions, Code Snippets contain sources that include Stack Exchange Network
Vulnerabilities
No vulnerabilities reported
Install deepchem
tensorflow - just cuda installed
pytorch - https://pytorch.org/get-started/locally/#start-locally
jax - https://github.com/google/jax#pip-installation-gpu-cuda
The DeepChem project maintains an extensive collection of tutorials. All tutorials are designed to be run on Google colab (or locally if you prefer). Tutorials are arranged in a suggested learning sequence which will take you from beginner to proficient at molecular machine learning and computational biology more broadly. After working through the tutorials, you can also go through other examples. To apply deepchem to a new problem, try starting from one of the existing examples or tutorials and modifying it step by step to work with your new use-case. If you have questions or comments you can raise them on our gitter.
Support
Reuse Trending Solutions
Find, review, and download reusable Libraries, Code Snippets, Cloud APIs from over 650 million Knowledge Items
Find more librariesStay Updated
Subscribe to our newsletter for trending solutions and developer bootcamps
Share this Page