Improve special_token_id logic in run_generation.py and add tests (#2885)

* improving generation

* finalized special token behaviour for no_beam_search generation

* solved modeling_utils merge conflict

* solve merge conflicts in modeling_utils.py

* add run_generation improvements from PR #2749

* adapted language generation to not use hardcoded -1 if no padding token is available

* remove the -1 removal as hard coded -1`s are not necessary anymore

* add lightweight language generation testing for randomely initialized models - just checking whether no errors are thrown

* add slow language generation tests for pretrained models using hardcoded output with pytorch seed

* delete ipdb

* check that all generated tokens are valid

* renaming

* renaming Generation -> Generate

* make style

* updated so that generate_beam_search has same token behavior than generate_no_beam_search

* consistent return format for run_generation.py

* deleted pretrain lm generate tests -> will be added in another PR

* cleaning of unused if statements and renaming

* run_generate will always return an iterable

* make style

* consistent renaming

* improve naming, make sure generate function always returns the same tensor, add docstring

* add slow tests for all lmhead models

* make style and improve example comments modeling_utils

* better naming and refactoring in modeling_utils

* improving generation

* finalized special token behaviour for no_beam_search generation

* solved modeling_utils merge conflict

* solve merge conflicts in modeling_utils.py

* add run_generation improvements from PR #2749

* adapted language generation to not use hardcoded -1 if no padding token is available

* remove the -1 removal as hard coded -1`s are not necessary anymore

* add lightweight language generation testing for randomely initialized models - just checking whether no errors are thrown

* add slow language generation tests for pretrained models using hardcoded output with pytorch seed

* delete ipdb

* check that all generated tokens are valid

* renaming

* renaming Generation -> Generate

* make style

* updated so that generate_beam_search has same token behavior than generate_no_beam_search

* consistent return format for run_generation.py

* deleted pretrain lm generate tests -> will be added in another PR

* cleaning of unused if statements and renaming

* run_generate will always return an iterable

* make style

* consistent renaming

* improve naming, make sure generate function always returns the same tensor, add docstring

* add slow tests for all lmhead models

* make style and improve example comments modeling_utils

* better naming and refactoring in modeling_utils

* changed fast random lm generation testing design to more general one

* delete in old testing design in gpt2

* correct old variable name

* temporary fix for encoder_decoder lm generation tests - has to be updated when t5 is fixed

* adapted all fast random generate tests to new design

* better warning description in modeling_utils

* better comment

* better comment and error message

Co-authored-by: Thomas Wolf <thomwolf@users.noreply.github.com>

tests/test_modeling_openai.py

@@ -39,6 +39,9 @@ class OpenAIGPTModelTest(ModelTesterMixin, unittest.TestCase):
     all_model_classes = (
         (OpenAIGPTModel, OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel) if is_torch_available() else ()
     )
+    all_generative_model_classes = (
+        (OpenAIGPTLMHeadModel,) if is_torch_available() else ()
+    )  # TODO (PVP): Add Double HeadsModel when generate() function is changed accordingly
 
     class OpenAIGPTModelTester(object):
         def __init__(