add train script

ctcvr_train.sh

+if ! ps aux | grep src/train_diary.py | grep -v grep
+then
+  nohup /home/gmuser/.virtualenvs/tf1/bin/python3 /srv/apps/gm_strategy_cvr/src/train_diary.py > ~/ctcvr_diary.log &
+fi
+
+if ! ps aux | grep src/train_tractate.py | grep -v grep
+then
+  nohup /home/gmuser/.virtualenvs/tf1/bin/python3 /srv/apps/gm_strategy_cvr/src/train_tractate.py > ~/ctcvr_tractate.log &
+fi

src/main_portrait.py

@@ -8,7 +8,7 @@ import tensorflow as tf
 from models.esmm.diary_model import model_predict_diary
 from models.esmm.fe import device_fe, diary_fe, tractate_fe
 from models.esmm.tractate_model import model_predict_tractate
-from utils.cache import redis_client2
+from utils.cache import get_essm_model_save_path, redis_client2
 from utils.grey import recommed_service_category_device_id_by_tail
 from utils.portrait import (get_user_portrait_tag3_read_v2, user_portrait_tag3_get_candidate_dict,
                             user_portrait_tag3_get_candidate_unread_list, user_portrait_tag3_write_ctcvr_data)
@@ -77,10 +77,16 @@ def main():
     tractate_dict = tractate_fe.get_tractate_dict_from_redis()
     print("redis data: " + str(len(device_dict)) + " " + str(len(diary_dict)) + " " + str(len(tractate_dict)))
 
-    diary_save_path = "/home/gmuser/data/models/diary/1596509008"
+    diary_save_path = get_essm_model_save_path("diary")
+    if not diary_save_path:
+        diary_save_path = "/home/gmuser/data/models/diary/1596509008"
+        print(diary_save_path + "!!!!!!!!!!!!!!!!!!!!!!!!!!!")
     diary_predict_fn = tf.contrib.predictor.from_saved_model(diary_save_path)
 
-    tractate_save_path = "/home/gmuser/data/models/tractate/1596509299"
+    tractate_save_path = get_essm_model_save_path("tractate")
+    if not tractate_save_path:
+        tractate_save_path = "/home/gmuser/data/models/tractate/1596509299"
+        print(tractate_save_path + "!!!!!!!!!!!!!!!!!!!!!!!!!!!")
     tractate_predict_fn = tf.contrib.predictor.from_saved_model(tractate_save_path)
 
     device_id = "androidid_a25a1129c0b38f7b"

src/train_diary.py

@@ -13,6 +13,7 @@ from models.esmm.diary_model import PREDICTION_ALL_COLUMNS, model_predict_diary
 from models.esmm.fe import click_fe, device_fe, diary_fe, fe
 from models.esmm.input_fn import esmm_input_fn
 from models.esmm.model import esmm_model_fn, model_export
+from utils.cache import set_essm_model_save_path
 
 
 def main():
@@ -44,8 +45,8 @@ def main():
     all_features = fe.build_features(df, diary_fe.INT_COLUMNS, diary_fe.FLOAT_COLUMNS, diary_fe.CATEGORICAL_COLUMNS)
     params = {"feature_columns": all_features, "hidden_units": [64, 32], "learning_rate": 0.1}
     model_path = str(Path("~/data/model_tmp/diary/").expanduser())
-    # if os.path.exists(model_path):
-    #     shutil.rmtree(model_path)
+    if os.path.exists(model_path):
+        shutil.rmtree(model_path)
 
     session_config = tf.compat.v1.ConfigProto()
     session_config.gpu_options.allow_growth = True
@@ -60,6 +61,7 @@ def main():
     model_export_path = str(Path("~/data/models/diary").expanduser())
     save_path = model_export(model, all_features, model_export_path)
     print("save to: " + save_path)
+    set_essm_model_save_path("diary", save_path)
 
     diary_train_columns = set(diary_fe.INT_COLUMNS + diary_fe.FLOAT_COLUMNS + diary_fe.CATEGORICAL_COLUMNS)
     diary_predict_columns = set(PREDICTION_ALL_COLUMNS)

src/train_tractate.py

@@ -13,6 +13,7 @@ from models.esmm.fe import click_fe, device_fe, fe, tractate_fe
 from models.esmm.input_fn import esmm_input_fn
 from models.esmm.model import esmm_model_fn, model_export
 from models.esmm.tractate_model import (PREDICTION_ALL_COLUMNS, model_predict_tractate)
+from utils.cache import set_essm_model_save_path
 
 
 def main():
@@ -41,8 +42,8 @@ def main():
     all_features = fe.build_features(df, tractate_fe.INT_COLUMNS, tractate_fe.FLOAT_COLUMNS, tractate_fe.CATEGORICAL_COLUMNS)
     params = {"feature_columns": all_features, "hidden_units": [64, 32], "learning_rate": 0.1}
     model_path = str(Path("~/data/model_tmp/tractate/").expanduser())
-    # if os.path.exists(model_path):
-    #     shutil.rmtree(model_path)
+    if os.path.exists(model_path):
+        shutil.rmtree(model_path)
 
     session_config = tf.compat.v1.ConfigProto()
     session_config.gpu_options.allow_growth = True
@@ -57,6 +58,7 @@ def main():
     model_export_path = str(Path("~/data/models/tractate/").expanduser())
     save_path = model_export(model, all_features, model_export_path)
     print("save to: " + save_path)
+    set_essm_model_save_path("tractate", save_path)
 
     tractate_train_columns = set(tractate_fe.INT_COLUMNS + tractate_fe.FLOAT_COLUMNS + tractate_fe.CATEGORICAL_COLUMNS)
     tractate_predict_columns = set(PREDICTION_ALL_COLUMNS)

src/utils/cache.py

@@ -5,3 +5,21 @@ redis_client2 = redis.StrictRedis.from_url("redis://:ReDis!GmTx*0aN9@172.16.40.1
 redis_client3 = redis.StrictRedis.from_url("redis://:ReDis!GmTx*0aN12@172.16.40.164:6379")
 redis_client4 = redis.StrictRedis.from_url("redis://:XfkMCCdWDIU%ls$h@172.16.50.145:6379")
 redis_db_client = redis.StrictRedis.from_url("redis://:ReDis!GmTx*0aN14@172.16.40.146:6379")
+
+
+def _essm_model_save_path(content_type):
+    return "doris:essm:{}:model_path".format(content_type)
+
+
+def get_essm_model_save_path(content_type):
+    key = _essm_model_save_path(content_type)
+    path = redis_client3.get(key)
+    if path:
+        path = str(path, "utf-8")
+        return path
+    return None
+
+
+def set_essm_model_save_path(content_type, path):
+    key = _essm_model_save_path(content_type)
+    redis_client3.set(key, path)