AttributeError: 'numpy.ndarray' object has no attribute 'atom_features'

[mukherjeesutanu]

❓ Questions & Help

Hi! I am new to deepchem and I am trying to build a classification model based on HIV.csv dataset. But during the fitting of the model I got the following error: AttributeError: 'numpy.ndarray' object has no attribute 'atom_features'. The corresponding code is given below:

import deepchem as dc
import pandas as pd
import os
df = pd.read_csv("HIV.csv")

smiles = df["smiles"][:100]

labels = df["HIV_active"][:100]

import deepchem as dc

featurizer = dc.feat.CircularFingerprint()

loader = dc.data.InMemoryLoader(tasks=["task1"], featurizer=featurizer)

dataset = loader.create_dataset(zip(smiles, labels), shard_size=2)

len(dataset)

from deepchem.splits.splitters import RandomSplitter
split_dataset = RandomSplitter()

train_data,valid_data,test_data=split_dataset.train_valid_test_split(dataset, frac_train= 0.8, frac_valid= 0.1, frac_test= 0.1, seed=23)

transformer = dc.trans.BalancingTransformer(dataset=dataset)
dataset = transformer.transform(dataset)

model = dc.models.GraphConvModel(n_tasks=1, mode='classification', graph_conv_layers=[128,128], dropout=0.20)

model.fit(train_data, nb_epoch=100)

[arunppsg]

For GraphConvModel, I think ConvMolFeaturizer should be used instead of CircularFingerprint. Could you retry the same with ConvMolFeaturizer?