Do you want to learn python from me with a lot of interactive quizzes, and exercises? Here is my project-based python learning course: codebasics.io/courses/python-for-beginner-and-intermediate-learners

Just started on the DSA journey combining practice with your guidance and taking notes. I started my Data science journey around 3n half years back and you were the first tutorial that I clicked for ML. I remember being so naive at that point but later on just got a flow out and got a chance to be a mature professional in data science field. Now feeling the urge for Targeting the Data Structure and Algorithms next!! Thank you so much! Enjoying it!

I have seen many videos but your videos have simplest explanation among all. Respect.

Amazing video! This was by far the clearest explanation of this I've ever seen!

this is the best practical exercise and explanation for hashtables for python in entire web, big thank you

Amazing video! Best explantion I have found and the exercises are very useful I just finnished the nyc_weather one. Used the built in hash function and handles collisions by adding (key, value) tuples to a list that is on the index. Thank you for the great video!

sooo clear and concise!!

You're an excellent teacher, Dhaval. Thanks and hope your passion does a lot of good for you!

Best explanation and implementation of hash table ever, understand it immediately

Even to this date your videos are the best on the entire youtube. Hats off to you!! Really enjoying learning DSA :)

You are a very good teacher. Thank you Mr Dhaval

This is actually a pretty good revision of the concepts too, thanks

Thanks so much! Keep making these vids!

Took me a while to really get each of the steps but good video :)

@Codebasics, I could image why HashTables are your favorite data structures...:) Exercises done and moving onto Stack and Queues ! Just brilliant tutorials.:)

Sir your way of teaching is just excellent, god bless you!

This was really helpful. Thank you.

Thank you Sir, the exercises were very helpful!

Wow thanks so much. Your explanation was very clear

Great video. Thank you!

Great videos! Thank you!

excellent explanation! thanks

Great videos, thank you so much! Do you have any projects that use these algorithms and data structures to help us imbed this information into our brains?

G.O.A.T you're my hero!!

Thankyou sir i really apreciate your way of teaching

And here I am paying college tuition and students loans to "learn" what this awesome dude gives out for free. You are the best man, love your content.

Awesome videos. Thank so much for making such precise vidoes.

Would be nice if you could touch on double hashing and also explain what happens when hash table itself needs to grow or the individual slot data structure has to grow. What happens to the time complexities in those cases?

I commend your channel and explanations. Very well put together thank you!

thanks nice session

Sir you're awesome. I subscribed and liked your videos. In the future please make videos on how to prepare for coding test and interviews and tech interviews for companies

Thanks ! I did 2 exercises by myself :) Soon going to Stack and Queue !

Thank you so much

Great explanation! You can use for else in 8:45 min.

Sir Sunday aur morning dono good ho gaye

Can anybody explain since we found the length of tuple is 2 and key is present at 0th index in the list then why do we inserting second key, value pair on that particular index.... isn't it that going to replace the first one

Hi, In the below snippet of your Linear Probing code, "if element is None: return" line is not required as it may return None if previous element of the desired element is None while traversing. However it is not running in any case when prob_range (even though it returns you all indexes) is used. def __getitem__(self, key): h = self.get_hash(key) if self.arr[h] is None: return prob_range = self.get_prob_range(h) for prob_index in prob_range: element = self.arr[prob_index] if element is None: return if element[0] == key: return element[1] For better understanding, if I change the code as shown below , then it will return None if previous element is None before the desired element is executed. Eg: [('march 17', 30), ('march 163', 355), ('march 193', 455), ('march 443', 455), ('march 553', 455), ('march 123', 455), ('march 133', 355), None, None, ('march 6', 20)] Output: It returns None for ''march 6' def __getitem__(self, key): h = self.get_hash(key) if self.arr[h] is None: return ##prob_range = self.get_prob_range(h) for prob_index in range(len(self.arr)): #Change here element = self.arr[prob_index] if element is None: return None #Change here if element[0] == key: return element[1] Thanks

very informative

Thanks

Thanks a lot Dhaval, I have been learning a lot from you and using this skill to search for a career as a developer or data scientist after 10 years of administration experience, hope I will find a career soon! I did all the exercises by myself with your encouragement but I check your every solution as well. I found the solution for Linear Probing is not working as expected when I am running the following case t = HashTable() t["march 6"] = 20 t["march 17"] = 88 del t["march 6"] t['march 17'] However, both of the following is working fine def __getitem__(self, key): h = self.get_hash(key) prob_range = self.get_prob_range(h) for prob_index in prob_range: element = self.arr[prob_index] if element is None: continue if element[0] == key: return element[1] def __getitem__(self, key): h = self.get_hash(key) for i in range(self.MAX): element = self.arr[h] h = (h+1)%self.MAX if element == None: continue if element[0] == key: return element[1]

Room looks clean

I'm actually just kind of confused after watching these last two videos. In linked lists you clearly helped us understand the structure of a linked list then asked us to modify what you wrote with our own functions. That was helpful. Here you showed us 50 minutes of hash functions and how to mess with lists(arrays) to store values and avoid collisions. Yet none of the excercises have anything to do with the videos themselves. > videos: I'm going to teach you how to cook a steak on your stove top exercise: now pull out some chicken so we can get started

Sir can you please make videos on dynamic programming...or can you please suggest some good channels from where i can follow.

great channel sir. Can you please make vedios on priority queue and heap data structure?

tq so much sir..................................... do more videos....

why did you check the length of element in setitem method ?

Thank you so much for your video! How would you implement a rehash to resize the current hash table? Any suggestions you could provide would be greatly appreciated!

hey if anyone having a problem understanding for loop in this program try printing idx , element, h in after if statement and append statement

Crisp and clear tutorial for hash maps sir.Thank you. Could you please explain why if len(element) ==2 is checked in setitem function

hello sir instead of using for loop directly if we check number of elements present in each list then going to for loop will it decrease the time complexity?

Which hash function does dictionary implements behind is it linear probing or chaining using list

Great video! However I do have a confusion as you call a list of tuples as linked list in the video. Shouldn't it be just an array of tuples? Since these tuples are not really linked in anyway. I mean if I want to insert another tuple or delete one tuple, the time complexity here is O(n), where as it should be O(1) for linked list, correct?

Hey man Excellent explanation and am studying from your videos right now I just found one small error in the hash table with linear probing solution We know according to our hash function 'march 6' and 'march 17' give the same value ie. 9 Now when using probing once we insert the value of march 6 followed by march 17 and if we delete the value for march 6 then calling for the value of march 17 returns none as while checking for the hash index it finds the index empty and returns none. Would love it if you could correct it as to how to get around this problem. Thanks a lot for the videos and seeing this in advance

def __delitem__(self, key): h = self.get_hash(key) for idx, k in enumerate(self.arr[h]): if k[0] == key: self.arr[h].remove(self.arr[h][idx])

Can't we use hash function which is readily available and gives the unique hash value for every input so we won't be needing chaining and there won't be any case where the index of two or more different keys will be matching with each other.

Let me take the chance to click the 1000th like!! Do we have to manually deal with csv files or can the third party/standard lib modules be used?

Why is the hashtable in python implement as nested list however , other languages use a array that is linked with linked list to store data that has collisions?

hello sir, what does the idx represents while iterating through the llist

for adding elements to the linked list just do append method, you dont need to do the enumerate loop and the found check

Is that linked Liston separate chaining or usual 2d array

i have doubt, for this setitem, getitem, delitem, still we are dependent on O(n). but how we are telling this o(1) ?

Can you explain linearprobing imlementation same way??

Hi i have a question here, to avoid collision we can resize the hashtable or increment it by one. So that we can get different hash values if there are any collisions. Is it possible to change the size of hashtable dynamically?

This one was tough. Lot of moving parts for me. bout to get started on these problems

for the setitem function, is the 'len(element) == 2 ' necessary? the every element in the linklist is the key-value pair, therefore it must be 2?

I have one question, lets say we have three columns in an excel and i want to print the third column value on the basis of two columns as key. Ex: ice and cream keys give value “ice-cream trucks” another ex: ice and water gives “cold water” as value another ex: water and cream gives “yuck taste” as value , assume this in a excel and i want to display ice cream truck or yuck water or cold water on the basis of two key inputs how can i do that!!

Hello Sir, can you please create a video developing of project using only DSA ?

what would be different if a dictionary was used in place of a simple list for chaining

I am using following function: def add(key, val): hash_value = get_hash(key) d[hash_value].append((key, val)) instead of your __setitem__() function and still my code is working fine, and I don't understand why? and when I am using your __setitem__(), I am getting an error on the following line: self.arr[h][idx] = (key,val)

I understood the video well but the exercises are like made me feel am I really understanding. Please someone help me to choose the right path, do I need to go for learning advanced python side by side?

is this implementation is just for explaning the concept of hash table or it's practical cause why make thing when we already have dictionary and other data structures working like a hash table.

How come I don't see a Linked List? I only see the array was used in the collision place?

Instead of appending (key, value) tuple, can we append a list [key,value]? What changes will it make?

love u

Sir, your video was of great help. Thank you. However, I have a doubt it would be helpful if you please clarify it. In chaining as said in video linked list is used. As I know every node in LL has a head and a next. So, if LL is used here then don't we have to use head and next in set and del methods? or simple list has been used here?

Could someone please tell me why are we not appending the key value pair whenever we find key to be already present. We should append the key value pair to that element, isn't it? def __setitem__(self, key, value): h = self.get_hash(key) found = False # if key exists already then append the value to the key for idx, k in enumerate(self.arr[h]): if len(k) == 2 and k[0] == key: self.arr[h].append((key, value)) found = True break # if the key value pair doesn't exist, add the key and value if not found: self.arr[h].append((key, value))

You said "chaining" is using linked list; but I only see adding tuples to a list. Where is the linked list? Am I missing something? Thanks!

Dhaval aap apni har video mai bol diya kijiye ki apka hindi channel bhi hai jo hindi mai comfortable hai woh bhi same benefits le sakta hai . Yeh ek request hai na ki salah. You are doing great job.

Hello Sir. Thank you so much for your videos, it has really been of help. I have just worked on the linear probing exercise and my solution is a little different from yours but it does what its suppose to do. I was hoping you could look into mine and comment on it, if it's right and why you didn't work it this way. class HashTable: def __init__(self): self.MAX = 10 self.arr = [None for i in range(self.MAX)] def get_hash(self, key): h = 0 for char in key: h += ord(char) return h % self.MAX def __setitem__(self, key, value): h = self.get_hash(key) if self.arr[h] is None: self.arr[h] = (key, value) else: empty_space = self.arr.index(None) # index() finds the first occurrence of an empty space self.arr[empty_space] = (key, value) def __getitem__(self, key): h = self.get_hash(key) return self.arr[h] h = HashTable() h['march 6'] = 130 h['march 7'] = 78 h['march 8'] = 210 h['march 9'] = 530 h['march 17'] = 28 print(h.arr) print(h['march 7']) Thank you very much sir. I look forward to your reply.

Keys 'march 9' and 'march 10' has the same h, but are different keys, how do we deal with that. Thanks :-)

linear probing: class HashTable: def __init__(self): self.MAX = 10 self.arr = [(None,None) for i in range(self.MAX)] def get_hash(self, key): hash = 0 for char in key: hash += ord(char) return hash % self.MAX def __getitem__(self, index): h = self.get_hash(index) found=False if self.arr[h][0]==index: return self.arr[h][1] else: for idx,element in enumerate(self.arr[h+1:]+self.arr[:h-1]): if element[0] == index: return element[1] return self.arr[h] def __setitem__(self, key, val): h = self.get_hash(key) set = False for idx,element in enumerate(self.arr): if idx==h and element[0]!=None: for i,j in enumerate(self.arr[idx+1:]+self.arr[:idx-1]): if j[0]==None: self.arr[i]=(key, val) set=True break if not set: self.arr[h] = (key, val)

Thanks for the great explanation......but I'm getting h value for 'march 6' and 'march 17' is different that is 9 and 59 ...so how to resolve this issue?

I can not find your exercises anywhere on-site . No Datastructure folder . Where are .cvs files .

can anyone explain whats happening in 7:51

The case where all the locations are filled in #LinearProbing i.e., 10 elements are filled according to the example what we need to do?

sir, I have a doubt !!! why you saying that list as link-list at index h, as it is working as a list, not link-list.

__setitem__ has a bug on python 3.11.0, the value cannot be set

Just a quick question: If Collision Handling methods make the search runtime go up to O(n), then is it really worth it to use hashmaps in the first place? I believe the lookup time also won't be constant time, it should be a linear time in the worst case.... Correct me if I'm wrong!

Hello Sir, Since we already we have dict handy in python, then why we exactly use hashmap? Please help me to understand. Is it just to get to know how dict works

Hi how do i download the files for exercises

Isn't the following incorrect?it always replaces the tuple in index 0 and it doesn't append

def get_prob_range(self, index): return [*range(index, len(self.arr))] + [*range(0,index)] please Explain it

Small correction: You meant to say (single) list instead of linked list.

My IDLE compiler is asking me to define f

why the len(element) == 2 condition in the __setitem__ function

Sir i am getting march 17 hash as 59 is it wrong or its related ASCII code according to any other configuration of the computer system

Sir,I have a doubt.Suppose all the elements got collided .Won't it append all the elements in a single array and therefore Won't the searching time be o(n)

Nobel prize

please can anyone explain,what is the line means def get_prob_range(self, index): return [*range(index, len(self.arr))] + [*range(0,index)]

why not using dict element in the self.arr instead of linked lists?