Using LendingClub with custom filters

[phaedrus75]

Hi
FIrst of all thanks for creating this great script - and it works as designed

What I'm trying to do is to run my own set of criteria - to buy notes on the LC website. My criteria resides on its own python code that returns the id of the notes I am interested in

When I pass the ids to add to the order and execute it - the script is throwing an error. I suspect this has to do with the fact that I cant figure out how to stage the loans once I know the ids

Can you pls advise how I would use the script to add to the order and execute it - if I have a list of ids. Apologies but my knowledge of python is very basic

regards

[jgillick]

Hi. I'm glad that you're using the module.

How are you trying to add the IDs to the order? Are you following the first example in the README?

When creating an order, it should handle staging the loans for you. Currently, however, the batch_add method requires a list of dictionaries -- not a list of IDs (I should fix this) Here's an example of how it should work:

from lendingclub import LendingClub

lc = LendingClub(email='test@test.com', password='secret123')

# Create list of loan IDs
ids = [
    {'loan_id': 1234}, 
    {'loan_id': 2345}, 
    {'loan_id': 3456}
]

# Invest $25 in all loans
order = lc.start_order()
order.add_batch(ids, 25)
order_id = order.execute() 

print 'Order placed: {0}'.format(order_id)

If this doesn't work, send me the error message you're seeing and we'll go from there. Cheers.

[jgillick]

I think I might have found a bug in my order staging code. Please send me the error you're getting so I can verify. Thank you.

[jgillick]

I found and fixed some bugs with the loan note staging code. I'm not sure if these were your bugs, but update to version 0.1.1 and let me know.

You can now add a list of IDs to the add_batch method, instead of a list of dictionaries. The example I showed above still works, but here is a simpler version, using version 0.1.1:

from lendingclub import LendingClub

lc = LendingClub(email='test@test.com', password='secret123')

# Create list of loan IDs
ids = [1234, 2345, 3456]

# Invest $25 in all loans
order = lc.start_order()
order.add_batch(ids, 25)
order_id = order.execute() 

print 'Order placed: {0}'.format(order_id)

[phaedrus75]

Thank you so much

I'll try running this and let you know if it fixes the issue

Regards

Sent from my iPhone

On 30 Jul, 2013, at 12:03 PM, Jeremy Gillick notifications@github.com wrote:

I found and fixed some bugs with the loan note staging code. I'm not sure if these were your bugs, but update to version 0.1.1 and let me know.

You can now add a list of IDs to the add_batch method, instead of a list of dictionaries. The example I showed above still works, but here is a simpler version, using version 0.1.1:

from lendingclub import LendingClub

lc = LendingClub(email='test@test.com', password='secret123')

Create list of loan IDs

ids = [1234, 2345, 3456]

Invest $25 in all loans

order = lc.start_order()
order.add_batch(ids, 25)
order_id = order.execute()

print 'Order placed: {0}'.format(order_id)
—
Reply to this email directly or view it on GitHub.

[phaedrus75]

I think it fixed the previous issue but it is still not working..

Here's what I am trying to do.

My program downloads the InFunding2Statsnew csv file from lending club - the file that is periodically refreshed and has new loans available for lending.

I am running criteria off the file - reason is there are far more attributes in the file than are available in web UI and my criteria uses some of these variables

Now since the file is updated in batches - there are notes in the file that are filled up by the time the program runs and thus is no longer available on the web. In such a case I just want to be able to pick whatever is available out of the pool that I find through parsing the file

Here is the error it is picking up. I found 45 matches on the file ... but only 22 were available when I run the code to add batch order as you explain above

Found 45 selected ones
Traceback (most recent call last):
File "csv_lc.py", line 109, in
order.execute()
File "/Users/xxx/Documents/LendingClub-master/lendingclub/init.py", line 714, in execute
self.stage_order()
File "/Users/xxx/Documents/LendingClub-master/lendingclub/__init.py", line 771, in __stage_order
raise LendingClubError('Could not stage the loans. The number of loans in your batch does not match totalRecords. {0} != {1}'.format(len(self.loans), results['totalRecords']), results)
lendingclub.LendingClubError: 'Could not stage the loans. The number of loans in your batch does not match totalRecords. 45 != 22'

[jgillick]

Ah, this makes sense. Currently, when you create an order it treats it as an atomic transaction, all loans are invested in or none. I might change this in the future, but there is a good workaround.

First I've updated the module to make this process smoother, so update your version of lendingclub to version 0.1.2 or greater:

sudo pip install lendingclub --upgrade

Now, get the list of loan IDs from your CSV and perform a search by ID. Only the loans that are still available will return. You can then add these to the order and invest. In the following example, it searches for 4 loans by ID and then invests $25 in whichever ones are still available:

from lendingclub.filters import FilterByLoanID

# Search for loans by ID
loan_ids = [54321, 65432, 76543, 876543]
filter = FilterByLoanID(loan_ids)
results = lc.search(filter)
print len(results['loans'])

# Invest $25 in all the loans that were found
order = lc.start_order()
order.add_batch(results['loans'], 25)
order.execute()

I haven't tried this myself, but it should work. The search results contain quite a bit of up-to-date information about each loan. So you could also check the funding progress on each one and adjust amount you invest per loan note -- you don't want to invest $100 in a loan that only has $25 left to fund.

I also finally have the full API documentation hosted, which might help:
https://python-lendingclub.readthedocs.org/en/latest/index.html

Feel free to message me with any other questions.