def run(): # Crop out facial images and retrieve features. Ideally, one only need to call this function once. prepare_data.run() # Make prediction by using different features for facial_image_extension, feature_extension in itertools.product(\ prepare_data.FACIAL_IMAGE_EXTENSION_LIST, prepare_data.FEATURE_EXTENSION_LIST): make_prediction(facial_image_extension, feature_extension) print("All done!")
def run_all(userid_, songids_, K): """ runs the prepare_data functions to create datasets and then runs the knn C++ program to get results written to file and then reads the file written to by running the C++ program Return: Returns a dict with the songids and the corresponding predictions """ filename_ = prepare_data.run(userid_, songids_) num_queries = len(songids_) outfile = DATADIR_ + userid_ + "_output" command = ( "/home/tathagata/SmartPlayerModules/Audio/audiomodules/KNN/KNN/KNN_c/k-classifier\ " + filename_ + " " + str(K) + "\ " + str(num_queries) + " > " + outfile ) # subprocess.call blocks until command has completed subprocess.call(command, 0, None, None, None, None, None, False, True) return read_knn_out(outfile)
def run_all(userid_, songids_, K): """ runs the prepare_data functions to create datasets and then runs the knn C++ program to get results written to file and then reads the file written to by running the C++ program Return: Returns a dict with the songids and the corresponding predictions """ trainfile_, testfile_ = prepare_data.run(userid_, songids_) print(trainfile_) print(testfile_) outfile = DATADIR_ + userid_ + '_output' command = '/home/tathagata/SmartPlayerModules/Audio/audiomodules/KNN/KNN/KNN_c/knn ' + trainfile_ + ' ' + testfile_ + ' ' + str(K) + '\ > ' + outfile # subprocess.call blocks until command has completed subprocess.call(command, 0, None, None, None, None, None, False, True) return read_knn_out(outfile)
def run_model_LR(Base_Dir_40p208, Base_Dir_share, date): # sys.path.append(Base_Dir_40p208 + '/model_LR') import prepare_data import offline_exp # conf = get_conf.run(Base_Dir_40p208, date, cid, bid, "1") # if conf.has_key("model_LR"): # conf = conf["model_LR"] # else: # print "run_model.py - run_model_LR() : cid-{0}, bid-{1} has no model_LR section.".format(cid, bid) # exit() conf = conf_dict conf_format = {} conf_format["General"] = {} for k in conf: if k == "select_features": conf_format["General"][k] = conf[k] continue if k == "sample_proportion": conf_format["General"][k] = conf[k] continue if k == "sample": conf_format["General"][k] = conf[k] continue if k == "train_days": conf_format["General"][k] = conf[k] continue if k == "test_days": conf_format["General"][k] = conf[k] continue conf_format[k] = conf[k] train_list, test_list = prepare_data.run(Base_Dir_40p208, Base_Dir_share, date, cid, bid, conf_format) print train_list print test_list offline_exp.run(Base_Dir_40p208, date, cid, bid, conf_format, train_list, test_list)
def run_model_LR(Base_Dir_40p208, Base_Dir_share, cid, bid, date): sys.path.append(Base_Dir_40p208 + '/model_LR') import prepare_data import offline_exp conf = get_conf.run(Base_Dir_40p208, date, cid, bid, "1") if conf.has_key("model_LR"): conf = conf["model_LR"] else: print "run_model.py - run_model_LR() : cid-{0}, bid-{1} has no model_LR section.".format(cid, bid) exit() conf_format = {} conf_format["General"] = {} for k in conf: if k == "select_features": conf_format["General"][k] = conf[k] continue if k == "sample_proportion": conf_format["General"][k] = conf[k] continue if k == "sample": conf_format["General"][k] = conf[k] continue if k == "train_days": conf_format["General"][k] = conf[k] continue if k == "test_days": conf_format["General"][k] = conf[k] continue conf_format[k] = conf[k] train_list, test_list = prepare_data.run(Base_Dir_40p208, Base_Dir_share, date, cid, bid, conf_format) print train_list print test_list offline_exp.run(Base_Dir_40p208, date, cid, bid, conf_format, train_list, test_list)