I feel a little overwhelmed trying to prepare each week and come up with good ideas, as well as coming up with things on the fly because the PC's never do what you would think they would do.
Okay, I re-read your initial question. You ask about how to get good ideas for the game session and also how to manage player choices that you haven't anticipated.
Preperation and "good ideas":
I can't stress this enough, but you don't have to deliver or produce high quality adventure content just right from the beginning and on your own.
First, this is a team game and the players are not here that you give them fantastic adventures on a silver plate. The players have to put at least a small amount of effort into getting the story going by themselves. So: the responsibility for "good ideas" does not entirely lie on your shoulders.
Second, even the smallest ideas for an adventure often lead to hours of real time play. Let's say that the local lord sends the adventurers out to stop some bandids in the woods. The PCs have to find their camp first, so either they go rummaging around the woods or maybe they ask around the shady parts of town. There they stumble upon some ruffians who don't like people sniffing around, find a small hideout at the docks, where a map to the bandid camp is stashed. In the woods the PCs stumble upon a old shrine of a forest god which is corrupted and host to dire owlbears. Lastly the PCs find the camp and discuss how to investigate further or to sneak in or even to go berserk and tear down everything by sword and magic. And voila: now you have content for at least two or even three 4-hour sessions. And although this is only a variation of one of the easiest and oldest adventure plots, your group can have fun with it.
Mind you: Even if your group is experienced, has playes for years and knows every story and every trick in the book, they are not entitled to "the greatest story of all time" (tm). You are a beginning DM and they should not expect highest drama and unforseeable plot twists.
Problems with player choices:
If you have trouble to steer your players into doing what you thought would be best: just don't. I know that improvisation sounds hard to do at first. But it solves so many problems. You don't have to prepare every eventuality beforehand and you don't have to think every situation through. That does cost a lot of time. You can spend it better in preparing some small lists that help you improvise. Take a piece of paper and write down ten NPC names, ten names for taverns, ten places to buy goods, ten small sites in the region where the PCs are right now (smth like "Mungo's abandoned mill and illicit casino", "The pool of dryads that is besieged by bullywugs", "Fort Crag, infested by a handful of lazy orcs", "The tower of Dashao, a helpful but slightly confused wizard", "Smallfeet company, halfling trading outpost where the two rivers meet"). For the ten NPC names write down ten quirks or visible features that you can use in describing the improvised NPCs. Or if you want more dungeon-themed lists, write down ten features of possible dungeon rooms: "very high ceiling with a small crack through which rainwater drops", "room with broken mirrors", "room with the scent of mint and ash", "room with a statue that seems absolutely life-like" etc.
Use those basic information to improvise scenes on the fly.
Let's use it in the adventure-example from above: the PCs stumble upon the corrupted shrine, suddenly loose interest in the bandid camp and want to investigate the corruption first. You have not prepared anything for it so you start to improvise. Maybe you can throw in a small three-"room" dungeon under the shrine and use the dungeon-room table? Maybe it leads to the pool of dryds that is besieged by bullywugs (see list list above) (or is infested with an old and vengeful spirit? Maybe it was awoken through the malavolence of the bandids?) Maybe the PCs meet a half-mad druid at the shrine (and whose name and features you can pick from your ten-names and ten-features list)?
