def cluster_analysis(clean_feature_file,
                     feature_names,
                     output_dir,
                     method='ward'):
    print datetime.now().strftime('starting:%Y-%m-%d %H:%M:%S')

    df_f = pd.read_csv(clean_feature_file)

    # all_feature_names=gl_feature_names
    print "There are %d neurons in this dataset" % df_f.shape[0]

    REMOVE_OUTLIERS = 1  #clipping the dataset
    if REMOVE_OUTLIERS > 0:
        postfix = "_ol_removed"
    else:
        postfix = "_ol_clipped_5_glonly"

    if method == "ap" or method == "all":
        fc.run_affinity_propagation(df_f, feature_names, output_dir, postfix)

    num_clusters = 1000
    if method == "ward" or method == "all":
        fc.run_ward_cluster(df_features=df_f,
                            feature_names=feature_names,
                            num_clusters=num_clusters,
                            output_dir=output_dir,
                            output_postfix=postfix,
                            experiment_type='bigneuron',
                            low=500,
                            high=1500,
                            plot_heatmap=0,
                            RemoveOutliers=REMOVE_OUTLIERS)

    print datetime.now().strftime('end:%Y-%m-%d %H:%M:%S')
    return
def cluster_analysis(clean_feature_file,feature_names,output_dir,feature_set_type,  method='ward',swc_path = None):
    print datetime.now().strftime('starting:%Y-%m-%d %H:%M:%S')
    if (swc_path == None):
        swc_path = "./SWC"

    df_f = pd.read_csv(clean_feature_file)


   # all_feature_names=gl_feature_names
    print "There are %d neurons in this dataset" % df_f.shape[0]


    REMOVE_OUTLIERS = 1 #clipping the dataset
    if REMOVE_OUTLIERS > 0:
        postfix = "_ol_removed_"+feature_set_type
    else:
        postfix = "_ol_clipped_5_glonly"


    if method == "ap" or method == "all":
        fc.run_affinity_propagation(df_f, feature_names, output_dir, postfix,swc_path,REMOVE_OUTLIERS)


    num_clusters = 12
    if method == "ward" or method == "all":
        fc.run_ward_cluster(df_features=df_f, feature_names=feature_names, num_clusters=num_clusters,
                            output_dir= output_dir,
                            output_postfix=postfix,experiment_type='ivscc', low=8, high = 35, plot_heatmap=0,
                            RemoveOutliers=REMOVE_OUTLIERS, swc_path=swc_path)

    print datetime.now().strftime('end:%Y-%m-%d %H:%M:%S')
    return
def cluster_analysis(clean_feature_file,feature_names,output_dir, method='ward'):
    print datetime.now().strftime('starting:%Y-%m-%d %H:%M:%S')

    df_f = pd.read_csv(clean_feature_file)


   # all_feature_names=gl_feature_names
    print "There are %d neurons in this dataset" % df_f.shape[0]


    REMOVE_OUTLIERS = 1 #clipping the dataset
    if REMOVE_OUTLIERS > 0:
        postfix = "_ol_removed"
    else:
        postfix = "_ol_clipped_5_glonly"


    if method == "ap" or method == "all":
        fc.run_affinity_propagation(df_f, feature_names, output_dir, postfix)


    num_clusters = 1000
    if method == "ward" or method == "all":
        fc.run_ward_cluster(df_features=df_f, feature_names=feature_names, num_clusters=num_clusters,output_dir= output_dir,
                                          output_postfix=postfix,experiment_type='bigneuron', low=500, high = 1500, plot_heatmap=0, RemoveOutliers=REMOVE_OUTLIERS)

    print datetime.now().strftime('end:%Y-%m-%d %H:%M:%S')
    return
示例#4
0
def cluster_analysis(clean_feature_file,
                     feature_names,
                     output_dir,
                     feature_set_type,
                     method='ward',
                     swc_path=None):
    print datetime.now().strftime('starting:%Y-%m-%d %H:%M:%S')
    if (swc_path == None):
        swc_path = "./SWC"

    df_f = pd.read_csv(clean_feature_file)

    # all_feature_names=gl_feature_names
    print "There are %d neurons in this dataset" % df_f.shape[0]

    REMOVE_OUTLIERS = 1  #clipping the dataset
    if REMOVE_OUTLIERS > 0:
        postfix = "_ol_removed_" + feature_set_type
    else:
        postfix = "_ol_clipped_5_glonly"

    if method == "ap" or method == "all":
        fc.run_affinity_propagation(df_f, feature_names, output_dir, postfix,
                                    swc_path, REMOVE_OUTLIERS)

    num_clusters = 12
    if method == "ward" or method == "all":
        fc.run_ward_cluster(df_features=df_f,
                            feature_names=feature_names,
                            num_clusters=num_clusters,
                            output_dir=output_dir,
                            output_postfix=postfix,
                            experiment_type='ivscc',
                            low=8,
                            high=35,
                            plot_heatmap=0,
                            RemoveOutliers=REMOVE_OUTLIERS,
                            swc_path=swc_path)

    print datetime.now().strftime('end:%Y-%m-%d %H:%M:%S')
    return