drorhilman

import os
import subprocess
from moviepy.editor import VideoFileClip
from moviepy.video.io.ffmpeg_tools import ffmpeg_extract_subclip
from tqdm import tqdm
import imageio
from rich.progress import track, SpinnerColumn, Progress
from rich import print

drorhilman

# -- consts --

ECHARTS_CDN = "https://cdn.jsdelivr.net/npm/[email protected]/dist/echarts.min"
ECHARTS_REQUIREJS_CONF = f"requirejs.config({{paths: {{echarts: '{ECHARTS_CDN}',}}}});"
ECHARTS_TEMPLATE = f"""
    <div id="{{ID}}" style="width: {{WIDTH}};height:{{HEIGHT}};"></div>
    <script type="module">
    {ECHARTS_REQUIREJS_CONF}    
    requirejs(["echarts"], (echarts) => {{
                    let myChart = echarts.init(document.getElementById('{{ID}}'));

drorhilman

drorhilman

from keras.layers import Dense, Dropout, Input, BatchNormalization
from keras.models import Model
def get_nn_model(dropout=0.6):
    inp = Input(shape = (90,))
    x = Dense(100, activation='relu')(inp)
    x = BatchNormalization()(x)
    x = Dropout(dropout)(x)
    
    x = Dense(50, activation='relu')(x)
    x = BatchNormalization()(x)

drorhilman

def is_nuc(nuc):
    try:
        n1, n2 = nuc.split(':')
        if not n1 in 'ATGC' and n2 in 'ATGC':
            return np.nan
        else:
            return nuc
    except: 
        return np.nan 
    

drorhilman

from sklearn.metrics import confusion_matrix
import seaborn as sns
forest.fit(x,y)
predict = forest.predict(x)
cm = pd.DataFrame(confusion_matrix(y, predict), columns=forest.classes_, index=forest.classes_)
sns.heatmap(cm)

drorhilman

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score
forest = RandomForestClassifier(n_estimators=30, class_weight='balanced')

x = encoded[[c for c in encoded.columns if 'snp' in c]].values
y = classes.values

scores = cross_val_score(forest, x, y, cv=5)
scores.mean(), scores.std()