过滤元组内的元组

Question

I am trying to filter a big number of tuples that look like this one:

tuple = ([], [('print', -108.0), ('denim', -144.0), ('floral', -158.0), ('summer', -172.0), ('chiffon', -217.0), ('crochet', -272.0), ('cotton', -275.0), ('graphic', -279.0), ('pattern', -281.0), ('sheer', -294.0)])

What I am trying to do is to create some lines of code that I can use to say to Python that I want to retrieve a list that contains just the strings with a number that is higher than -200,in this case I would like to obtain:

list = ["print", "denim", "floral", "summer"]

NEW PART:

Now thanks to @Andrej Kesely I have the solution to this problem, but if I want to do the same things but with a list of tuples like this one:

predictions = [([],
  [('print', -72.0),
   ('summer', -141.0),
   ('chiffon', -157.0),
   ('floral', -266.0),
   ('graphic', -279.0),
   ('cotton', -279.0),
   ('denim', -282.0),
   ('cute', -321.0),
   ('striped', -331.0),
   ('pattern', -337.0)]),
 ([],
  [('chiffon', -89.0),
   ('summer', -214.0),
   ('pleated', -250.0),
   ('sheer', -280.0),
   ('woven', -286.0),
   ('crochet', -293.0),
   ('solid', -295.0),
   ('cotton', -300.0),
   ('mesh', -316.0),
   ('party', -332.0)]),
 ([],
  [('crochet', -188.0),
   ('chiffon', -204.0),
   ('summer', -209.0),
   ('floral', -209.0),
   ('pattern', -214.0),
   ('print', -233.0),
   ('cotton', -252.0),
   ('sheer', -282.0),
   ('elegant', -294.0),
   ('striped', -300.0)]),
 (['striped'],
  [('striped', 73.0),
   ('summer', -216.0),
   ('cotton', -252.0),
   ('chiffon', -287.0),
   ('denim', -299.0),
   ('print', -336.0),
   ('cute', -357.0),
   ('linen', -365.0),
   ('sheer', -372.0),
   ('chic', -385.0)]),
 ([],
  [('summer', -146.0),
   ('crochet', -227.0),
   ('party', -233.0),
   ('cute', -251.0),
   ('mesh', -281.0),
   ('chiffon', -289.0),
   ('solid', -294.0),
   ('cotton', -301.0),
   ('floral', -322.0),
   ('beach', -323.0)]),
 ([],
  [('cotton', -168.0),
   ('summer', -197.0),
   ('print', -211.0),
   ('striped', -227.0),
   ('graphic', -248.0),
   ('denim', -260.0),
   ('pattern', -302.0),
   ('floral', -336.0),
   ('solid', -341.0),
   ('cute', -341.0)]),
 ([],
  [('print', -165.0),
   ('denim', -216.0),
   ('cotton', -260.0),
   ('striped', -262.0),
   ('graphic', -301.0),
   ('leather', -327.0),
   ('mesh', -339.0),
   ('stretch', -341.0),
   ('pattern', -346.0),
   ('summer', -347.0)]),
 ([],
  [('crochet', -124.0),
   ('chiffon', -165.0),
   ('floral', -169.0),
   ('summer', -170.0),
   ('sheer', -227.0),
   ('party', -245.0),
   ('cute', -260.0),
   ('mesh', -286.0),
   ('elegant', -302.0),
   ('print', -308.0)])]

How can I create a list of lists that contain a list for each tuple in which there are shown just the strings with a number that is higher than -200? The problem comes from the output of an image recognition model and I tried to embed the code of @Andrej Kesely in my for loop on the output of the model in this way:

predictions=[]
for x in df_images["Media"]:
    p= predict(x, labels, model)
    prediction = [i for lst in p for i, v in lst if v > -200]
    predictions.append(prediction)

But it gives me this error: "too many values to unpack (expected 2)"

Answer 1

You can use list comprehension:

tpl = (
    [],
    [
        ("print", -108.0),
        ("denim", -144.0),
        ("floral", -158.0),
        ("summer", -172.0),
        ("chiffon", -217.0),
        ("crochet", -272.0),
        ("cotton", -275.0),
        ("graphic", -279.0),
        ("pattern", -281.0),
        ("sheer", -294.0),
    ],
)

out = [i for lst in tpl for i, v in lst if v > -200]
print(out)

Prints:

['print', 'denim', 'floral', 'summer']

---

EDIT: With the new list predictions from the question:

out = [[i for i, v in lst if v > -200] for _, lst in predictions]
print(out)

Prints:

[['print', 'summer', 'chiffon'], ['chiffon'], ['crochet'], ['striped'], ['summer'], ['cotton', 'summer'], ['print'], ['crochet', 'chiffon', 'floral', 'summer']]