50efccc10d7dd32caf913b7da3059551dd7739d6,deepchem/splits/__init__.py,StratifiedSplitter,__generate_required_hits,#StratifiedSplitter#Any#Any#,96

Before Change


        required_hit_dict = {}
        totalCount = len(y_df.index)
        for col in y_df:
            NaN_count = y_df[col].isnull().sum()
            notNaN = totalCount - NaN_count
            requiredNotNaN = frac_train * notNaN
            required_hit_dict[colIndex] = requiredNotNaN
            colIndex += 1
        return required_hit_dict

    def __generate_required_index(self, y_df, required_hit_dict):

After Change


    def __generate_required_hits(self, w, frac_train):
        required_hits = (w != 0).sum(0)//returns list of per column sum of non zero elements
        for colHits in required_hits:
          colHits = int(frac_train * colHits)
        return required_hits

    def __generate_required_index(self, w, required_hit_list):
        colIndex = 0
Italian Trulli
In pattern: SUPERPATTERN

Frequency: 4

Non-data size: 5

Instances


Project Name: deepchem/deepchem
Commit Name: 50efccc10d7dd32caf913b7da3059551dd7739d6
Time: 2016-08-01
Author: apappu97@gmail.com
File Name: deepchem/splits/__init__.py
Class Name: StratifiedSplitter
Method Name: __generate_required_hits


Project Name: scikit-learn-contrib/imbalanced-learn
Commit Name: 249a1a53fe07a903d01d59fb69437c300b59461f
Time: 2016-06-18
Author: glemaitre@visor.udg.edu
File Name: unbalanced_dataset/under_sampling/instance_hardness_threshold.py
Class Name: InstanceHardnessThreshold
Method Name: transform


Project Name: senarvi/theanolm
Commit Name: eb8b70c825ef3e02818cbfa95c0f8a9ec4184275
Time: 2015-12-25
Author: seppo.git@marjaniemi.com
File Name: theanolm/network.py
Class Name: Network
Method Name: __init__


Project Name: facebookresearch/ParlAI
Commit Name: bda270d6255bca6c9518f4f6e8d2cdee8cd6a40c
Time: 2018-08-29
Author: ahm@fb.com
File Name: parlai/agents/seq2seq/modules.py
Class Name: Encoder
Method Name: forward