mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-20 14:22:08 +00:00
abd58ff82c
# What does this PR do? Reworked the loading logic. Idea is to use cleaner loading code: - Remove need for `no_init_weights` - Remove all weird `bnb_linear` and `load_weights` and `post_load_weights`. New code layout: - New class `Weights` in charge of handling loading the weights from multiple files into appropiate tensors (potentially sharded) - TP layers now are "shells", they contain the code to know what kind of sharding we need + eventual `all_reduce`. They do not inherit from linear, but they contain some kind of Linear instead - the contained linear can be either FastLinear, BnbLinear or GPTq Linear next. - All modeling code is explictly made for sharding, process group is just no-ops for non sharded code (removes a lot of test cases)  --------- Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.taildb5d.ts.net> Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.ec2.internal> Co-authored-by: OlivierDehaene <olivier@huggingface.co> Co-authored-by: OlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com>
45 lines
1.2 KiB
Python
45 lines
1.2 KiB
Python
import pytest
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def neox_sharded_handle(launcher):
|
|
with launcher(
|
|
"OpenAssistant/oasst-sft-1-pythia-12b", num_shard=2, use_flash_attention=False
|
|
) as handle:
|
|
yield handle
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
async def neox_sharded(neox_sharded_handle):
|
|
await neox_sharded_handle.health(300)
|
|
return neox_sharded_handle.client
|
|
|
|
|
|
@pytest.mark.skip
|
|
@pytest.mark.asyncio
|
|
async def test_neox(neox_sharded, response_snapshot):
|
|
response = await neox_sharded.generate(
|
|
"<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>",
|
|
max_new_tokens=10,
|
|
decoder_input_details=True,
|
|
)
|
|
|
|
assert response.details.generated_tokens == 10
|
|
assert response == response_snapshot
|
|
|
|
|
|
@pytest.mark.skip
|
|
@pytest.mark.asyncio
|
|
async def test_neox_load(neox_sharded, generate_load, response_snapshot):
|
|
responses = await generate_load(
|
|
neox_sharded,
|
|
"<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>",
|
|
max_new_tokens=10,
|
|
n=4,
|
|
)
|
|
|
|
assert len(responses) == 4
|
|
assert all([r.generated_text == responses[0].generated_text for r in responses])
|
|
|
|
assert responses == response_snapshot
|