added app and run

app.py

@@ -0,0 +1,30 @@
+import streamlit as st
+from time import time, ctime
+import tiger_trainer as stt
+import run as trun
+import pandas as pd 
+import os, shutil, glob
+
+
+def run_with_input(reset=False):
+    if reset:
+        st.write("")
+        return 0
+    returned_x = trun.run(st.session_state["userInput"])
+    csv_x = returned_x.to_csv()
+    st.write("model prediction: ", returned_x)
+    return csv_x
+
+st.title("Tiger gen prediction")
+st.session_state['userInput'] = ""
+st.session_state["userInput"] = st.text_input('type gen sequence')
+cvs_data = "first run model to generate data"
+if len(st.session_state['userInput']) < 23:
+    st.write("Sorry your input length must be at least 23 bases. It is %s chars"%len(st.session_state['userInput']))
+    run_with_input(reset=True)
+elif all([True if item in "ACGTacgt" else False for item in st.session_state['userInput']]):
+    st.write('This is your sequence', st.session_state["userInput"])
+    csv_data = run_with_input()
+else:
+    st.write("only ACTG is allowed")    
+st.download_button(label="Download as CVS File", data=csv_data)

run.py

@@ -0,0 +1,61 @@
+import os
+import sys
+import tensorflow as tf
+import pandas as pd
+
+GUIDE_LEN = 23
+NUCLEOTIDE_TOKENS = dict(zip(['A', 'C', 'G', 'T'], [0, 1, 2, 3]))
+
+# load model
+if os.path.exists('model'):
+    tiger = tf.keras.models.load_model('model')
+else:
+    print('no saved model!')
+    exit()
+
+
+def process_data(x):
+    x = [item.upper() for item in x]
+    number_of_input = len(x) - GUIDE_LEN + 1
+    input_gens = []
+    for i in range(number_of_input):
+        input_gens.append("".join(x[i:i + GUIDE_LEN]))
+    merged_token = []
+    token_x = [NUCLEOTIDE_TOKENS[item] for item in x]
+    for i in range(number_of_input):
+        merged_token.extend(token_x[i:i + GUIDE_LEN])
+    one_hot_x = tf.one_hot(merged_token, depth=4)
+    model_input_x = tf.reshape(one_hot_x, [-1, GUIDE_LEN * 4])
+    return input_gens, model_input_x
+
+
+def gen_report_table(input_gens, res):
+    res = res.numpy().flatten().tolist()
+    # print("ftaltten res: ", res)
+    data = {"Gene": input_gens, "res": res}
+    tbl = pd.DataFrame.from_dict(data)
+    return tbl
+
+
+def run(x):
+    input_gens, model_input_x = process_data(x)
+    # print("input gene: ", input_gens)
+    # print("model_input: ", model_input_x)
+    res = tiger.model.predict_step(model_input_x)
+    # print("res: ", res)
+    return gen_report_table(input_gens, res)
+
+
+if __name__ == "__main__":
+    if len(sys.argv) == 1:
+        print("you need to specify 23 character gen information")
+        exit()
+    x = sys.argv[1]
+    if len(x) != 23:
+        print("you need to specify 23 character gen information. You typed %s chars" % len(x))
+        exit()
+    elif all([True if item in "ACGT" else False for item in x]):
+        print("run succesfully: ", run(x))
+    else:
+        print("your gen sequence need 23 character only from ACGT")
+        exit()