SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

GPT-J-6B (#13022)

Browse Source 
* Test GPTJ implementation

* Fixed conflicts

* Update __init__.py

* Update __init__.py

* change GPT_J to GPTJ

* fix missing imports and typos

* use einops for now
(need to change to torch ops later)

* Use torch ops instead of einsum

* remove einops deps

* Update configuration_auto.py

* Added GPT J

* Update gptj.rst

* Update __init__.py

* Update test_modeling_gptj.py

* Added GPT J

* Changed configs to match GPT2 instead of GPT Neo

* Removed non-existent sequence model

* Update configuration_auto.py

* Update configuration_auto.py

* Update configuration_auto.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Progress on updating configs to agree with GPT2

* Update modeling_gptj.py

* num_layers -> n_layer

* layer_norm_eps -> layer_norm_epsilon

* attention_layers -> num_hidden_layers

* Update modeling_gptj.py

* attention_pdrop -> attn_pdrop

* hidden_act -> activation_function

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* fix layernorm and lm_head size
delete attn_type

* Update docs/source/model_doc/gptj.rst

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* removed claim that GPT J uses local attention

* Removed GPTJForSequenceClassification

* Update src/transformers/models/gptj/configuration_gptj.py

Co-authored-by: Lysandre Debut <lysandre@huggingface.co>

* Removed unsupported boilerplate

* Update tests/test_modeling_gptj.py

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Lysandre Debut <lysandre@huggingface.co>

* Update tests/test_modeling_gptj.py

Co-authored-by: Eric Hallahan <eric@hallahans.name>

* Update tests/test_modeling_gptj.py

Co-authored-by: Eric Hallahan <eric@hallahans.name>

* Update tests/test_modeling_gptj.py

Co-authored-by: Eric Hallahan <eric@hallahans.name>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update __init__.py

* Update configuration_gptj.py

* Update modeling_gptj.py

* Corrected indentation

* Remove stray backslash

* Delete .DS_Store

* Delete .DS_Store

* Delete .DS_Store

* Delete .DS_Store

* Delete .DS_Store

* Update docs to match

* Remove tf loading

* Remove config.jax

* Remove stray `else:` statement

* Remove references to `load_tf_weights_in_gptj`

* Adapt tests to match output from GPT-J 6B

* Apply suggestions from code review

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Default `activation_function` to `gelu_new`

- Specify the approximate formulation of GELU to ensure parity with the default setting of `jax.nn.gelu()`

* Fix part of the config documentation

* Revert "Update configuration_auto.py"

This reverts commit e9860e9c043b6ebf57a0e705044e9ec9ba2263bb.

* Revert "Update configuration_auto.py"

This reverts commit cfaaae4c4dc70f1fbe9abd60fc8bd0b863b8c011.

* Revert "Update configuration_auto.py"

This reverts commit 687788954fd0cfbc567fa1202d56a4ff9271944f.

* Revert "Update configuration_auto.py"

This reverts commit 194d024ea87d4fcef0dcb08e57f52c47511a9fc6.

* Hyphenate GPT-J

* Undid sorting of the models alphabetically

* Reverting previous commit

* fix style and quality issues

* Update docs/source/model_doc/gptj.rst

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/__init__.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update tests/test_modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/__init__.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/configuration_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/configuration_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/configuration_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Replaced GPTJ-specific code with generic code

* Update src/transformers/models/gptj/modeling_gptj.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Made the code always use rotary positional encodings

* Update index.rst

* Fix documentation

* Combine attention classes

- Condense all attention operations into `GPTJAttention`
- Replicate GPT-2 and improve code clarity by renaming `GPTJAttention.attn_pdrop` and `GPTJAttention.resid_pdrop` to `GPTJAttention.attn_dropout` and `GPTJAttention.resid_dropout`

* Removed `config.rotary_dim` from tests

* Update test_modeling_gptj.py

* Update test_modeling_gptj.py

* Fix formatting

* Removed depreciated argument `layer_id` to `GPTJAttention`

* Update modeling_gptj.py

* Update modeling_gptj.py

* Fix code quality

* Restore model functionality

* Save `lm_head.weight` in checkpoints

* Fix crashes when loading with reduced precision

* refactor self._attn(...)` and rename layer weights"

* make sure logits are in fp32 for sampling

* improve docs

* Add `GPTJForCausalLM` to `TextGenerationPipeline` whitelist

* Added GPT-J to the README

* Fix doc/readme consistency

* Add rough parallelization support

- Remove unused imports and variables
- Clean up docstrings
- Port experimental parallelization code from GPT-2 into GPT-J

* Clean up loose ends

* Fix index.rst

Co-authored-by: kurumuz <kurumuz1@gmail.com>
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
Co-authored-by: Eric Hallahan <eric@hallahans.name>
Co-authored-by: Leo Gao <54557097+leogao2@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: your_github_username <your_github_email>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
This commit is contained in:
Stella Biderman
2021-08-31 11:53:02 -04:00
committed by GitHub
parent e53af030c0
commit c02cd95c56
13 changed files with 1938 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

560
tests/test_modeling_gptj.py Normal file
View File

@@ -0,0 +1,560 @@
# coding=utf-8
# Copyright 2021 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import datetime
import unittest
from transformers import GPTJConfig, is_torch_available
from transformers.testing_utils import require_torch, slow, torch_device
from .test_configuration_common import ConfigTester
from .test_modeling_common import floats_tensor, ids_tensor, random_attention_mask
if is_torch_available():
import torch
from transformers import (
GPTJ_PRETRAINED_MODEL_ARCHIVE_LIST,
AutoTokenizer,
GPTJForCausalLM,
GPTJForSequenceClassification,
GPTJModel,
)
class GPTJModelTester:
def __init__(
self,
parent,
batch_size=14,
seq_length=7,
is_training=True,
use_token_type_ids=True,
use_input_mask=True,
use_labels=True,
use_mc_token_ids=True,
vocab_size=99,
hidden_size=32,
num_hidden_layers=5,
num_attention_heads=4,
intermediate_size=37,
hidden_act="gelu",
hidden_dropout_prob=0.0,
attention_probs_dropout_prob=0.0,
max_position_embeddings=512,
type_vocab_size=16,
type_sequence_label_size=2,
initializer_range=0.02,
num_labels=3,
num_choices=4,
):
self.parent = parent
self.batch_size = batch_size
self.seq_length = seq_length
self.is_training = is_training
self.use_token_type_ids = use_token_type_ids
self.use_input_mask = use_input_mask
self.use_labels = use_labels
self.use_mc_token_ids = use_mc_token_ids
self.vocab_size = vocab_size
self.hidden_size = hidden_size
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.intermediate_size = intermediate_size
self.hidden_act = hidden_act
self.hidden_dropout_prob = hidden_dropout_prob
self.attention_probs_dropout_prob = attention_probs_dropout_prob
self.max_position_embeddings = max_position_embeddings
self.type_vocab_size = type_vocab_size
self.type_sequence_label_size = type_sequence_label_size
self.initializer_range = initializer_range
self.num_labels = num_labels
self.num_choices = num_choices
self.scope = None
self.bos_token_id = vocab_size - 1
self.eos_token_id = vocab_size - 1
self.pad_token_id = vocab_size - 1
def get_large_model_config(self):
return GPTJConfig.from_pretrained("EleutherAI/gpt-j-6B")
def prepare_config_and_inputs(self, gradient_checkpointing=False):
input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
input_mask = None
if self.use_input_mask:
input_mask = random_attention_mask([self.batch_size, self.seq_length])
token_type_ids = None
if self.use_token_type_ids:
token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
mc_token_ids = None
if self.use_mc_token_ids:
mc_token_ids = ids_tensor([self.batch_size, self.num_choices], self.seq_length)
sequence_labels = None
token_labels = None
choice_labels = None
if self.use_labels:
sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
choice_labels = ids_tensor([self.batch_size], self.num_choices)
config = self.get_config(gradient_checkpointing=gradient_checkpointing)
head_mask = ids_tensor([self.num_hidden_layers, self.num_attention_heads], 2)
return (
config,
input_ids,
input_mask,
head_mask,
token_type_ids,
mc_token_ids,
sequence_labels,
token_labels,
choice_labels,
)
def get_config(self, gradient_checkpointing=False):
return GPTJConfig(
vocab_size=self.vocab_size,
n_embd=self.hidden_size,
n_layer=self.num_hidden_layers,
n_head=self.num_attention_heads,
intermediate_size=self.intermediate_size,
hidden_act=self.hidden_act,
hidden_dropout_prob=self.hidden_dropout_prob,
attention_probs_dropout_prob=self.attention_probs_dropout_prob,
n_positions=self.max_position_embeddings,
n_ctx=self.max_position_embeddings,
type_vocab_size=self.type_vocab_size,
initializer_range=self.initializer_range,
use_cache=not gradient_checkpointing,
bos_token_id=self.bos_token_id,
eos_token_id=self.eos_token_id,
pad_token_id=self.pad_token_id,
gradient_checkpointing=gradient_checkpointing,
)
def prepare_config_and_inputs_for_decoder(self):
(
config,
input_ids,
input_mask,
head_mask,
token_type_ids,
mc_token_ids,
sequence_labels,
token_labels,
choice_labels,
) = self.prepare_config_and_inputs()
encoder_hidden_states = floats_tensor([self.batch_size, self.seq_length, self.hidden_size])
encoder_attention_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
return (
config,
input_ids,
input_mask,
head_mask,
token_type_ids,
sequence_labels,
token_labels,
choice_labels,
encoder_hidden_states,
encoder_attention_mask,
)
def create_and_check_gptj_model(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
model = GPTJModel(config=config)
model.to(torch_device)
model.eval()
result = model(input_ids, token_type_ids=token_type_ids, head_mask=head_mask)
result = model(input_ids, token_type_ids=token_type_ids)
result = model(input_ids)
self.parent.assertEqual(result.last_hidden_state.shape, (self.batch_size, self.seq_length, self.hidden_size))
self.parent.assertEqual(len(result.past_key_values), config.n_layer)
def create_and_check_gptj_model_past(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
model = GPTJModel(config=config)
model.to(torch_device)
model.eval()
# first forward pass
outputs = model(input_ids, token_type_ids=token_type_ids, use_cache=True)
outputs_use_cache_conf = model(input_ids, token_type_ids=token_type_ids)
outputs_no_past = model(input_ids, token_type_ids=token_type_ids, use_cache=False)
self.parent.assertTrue(len(outputs) == len(outputs_use_cache_conf))
self.parent.assertTrue(len(outputs) == len(outputs_no_past) + 1)
output, past = outputs.to_tuple()
# create hypothetical next token and extent to next_input_ids
next_tokens = ids_tensor((self.batch_size, 1), config.vocab_size)
next_token_types = ids_tensor([self.batch_size, 1], self.type_vocab_size)
# append to next input_ids and token_type_ids
next_input_ids = torch.cat([input_ids, next_tokens], dim=-1)
next_token_type_ids = torch.cat([token_type_ids, next_token_types], dim=-1)
output_from_no_past = model(next_input_ids, token_type_ids=next_token_type_ids)["last_hidden_state"]
output_from_past = model(next_tokens, token_type_ids=next_token_types, past_key_values=past)[
"last_hidden_state"
]
# select random slice
random_slice_idx = ids_tensor((1,), output_from_past.shape[-1]).item()
output_from_no_past_slice = output_from_no_past[:, -1, random_slice_idx].detach()
output_from_past_slice = output_from_past[:, 0, random_slice_idx].detach()
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(output_from_past_slice, output_from_no_past_slice, atol=1e-3))
def create_and_check_gptj_model_attention_mask_past(
self, config, input_ids, input_mask, head_mask, token_type_ids, *args
):
model = GPTJModel(config=config)
model.to(torch_device)
model.eval()
# create attention mask
attn_mask = torch.ones(input_ids.shape, dtype=torch.long, device=torch_device)
half_seq_length = self.seq_length // 2
attn_mask[:, half_seq_length:] = 0
# first forward pass
output, past = model(input_ids, attention_mask=attn_mask).to_tuple()
# create hypothetical next token and extent to next_input_ids
next_tokens = ids_tensor((self.batch_size, 1), config.vocab_size)
# change a random masked slice from input_ids
random_seq_idx_to_change = ids_tensor((1,), half_seq_length).item() + 1
random_other_next_tokens = ids_tensor((self.batch_size, 1), config.vocab_size).squeeze(-1)
input_ids[:, -random_seq_idx_to_change] = random_other_next_tokens
# append to next input_ids and attn_mask
next_input_ids = torch.cat([input_ids, next_tokens], dim=-1)
attn_mask = torch.cat(
[attn_mask, torch.ones((attn_mask.shape[0], 1), dtype=torch.long, device=torch_device)],
dim=1,
)
# get two different outputs
output_from_no_past = model(next_input_ids, attention_mask=attn_mask)["last_hidden_state"]
output_from_past = model(next_tokens, past_key_values=past, attention_mask=attn_mask)["last_hidden_state"]
# select random slice
random_slice_idx = ids_tensor((1,), output_from_past.shape[-1]).item()
output_from_no_past_slice = output_from_no_past[:, -1, random_slice_idx].detach()
output_from_past_slice = output_from_past[:, 0, random_slice_idx].detach()
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(output_from_past_slice, output_from_no_past_slice, atol=1e-3))
def create_and_check_gptj_model_past_large_inputs(
self, config, input_ids, input_mask, head_mask, token_type_ids, *args
):
model = GPTJModel(config=config)
model.to(torch_device)
model.eval()
# first forward pass
outputs = model(input_ids, token_type_ids=token_type_ids, attention_mask=input_mask, use_cache=True)
output, past = outputs.to_tuple()
# create hypothetical next token and extent to next_input_ids
next_tokens = ids_tensor((self.batch_size, 3), config.vocab_size)
next_token_types = ids_tensor([self.batch_size, 3], self.type_vocab_size)
next_mask = ids_tensor((self.batch_size, 3), vocab_size=2)
# append to next input_ids and token_type_ids
next_input_ids = torch.cat([input_ids, next_tokens], dim=-1)
next_token_type_ids = torch.cat([token_type_ids, next_token_types], dim=-1)
next_attention_mask = torch.cat([input_mask, next_mask], dim=-1)
output_from_no_past = model(
next_input_ids, token_type_ids=next_token_type_ids, attention_mask=next_attention_mask
)["last_hidden_state"]
output_from_past = model(
next_tokens, token_type_ids=next_token_types, attention_mask=next_attention_mask, past_key_values=past
)["last_hidden_state"]
self.parent.assertTrue(output_from_past.shape[1] == next_tokens.shape[1])
# select random slice
random_slice_idx = ids_tensor((1,), output_from_past.shape[-1]).item()
output_from_no_past_slice = output_from_no_past[:, -3:, random_slice_idx].detach()
output_from_past_slice = output_from_past[:, :, random_slice_idx].detach()
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(output_from_past_slice, output_from_no_past_slice, atol=1e-3))
def create_and_check_lm_head_model(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
model = GPTJForCausalLM(config)
model.to(torch_device)
model.eval()
result = model(input_ids, token_type_ids=token_type_ids, labels=input_ids)
self.parent.assertEqual(result.loss.shape, ())
self.parent.assertEqual(result.logits.shape, (self.batch_size, self.seq_length, self.vocab_size))
def create_and_check_forward_and_backwards(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
model = GPTJForCausalLM(config)
model.to(torch_device)
result = model(input_ids, token_type_ids=token_type_ids, labels=input_ids)
self.parent.assertEqual(result.loss.shape, ())
self.parent.assertEqual(result.logits.shape, (self.batch_size, self.seq_length, self.vocab_size))
result.loss.backward()
def prepare_config_and_inputs_for_common(self):
config_and_inputs = self.prepare_config_and_inputs()
(
config,
input_ids,
input_mask,
head_mask,
token_type_ids,
mc_token_ids,
sequence_labels,
token_labels,
choice_labels,
) = config_and_inputs
inputs_dict = {"input_ids": input_ids, "token_type_ids": token_type_ids, "head_mask": head_mask}
return config, inputs_dict
@require_torch
class GPTJModelTest(unittest.TestCase):
all_model_classes = (GPTJModel, GPTJForCausalLM, GPTJForSequenceClassification) if is_torch_available() else ()
all_generative_model_classes = (GPTJForCausalLM,) if is_torch_available() else ()
fx_ready_model_classes = all_model_classes
test_pruning = False
test_missing_keys = False
test_model_parallel = False
# special case for DoubleHeads model
def _prepare_for_class(self, inputs_dict, model_class, return_labels=False):
inputs_dict = super()._prepare_for_class(inputs_dict, model_class, return_labels=return_labels)
return inputs_dict
def setUp(self):
self.model_tester = GPTJModelTester(self)
self.config_tester = ConfigTester(self, config_class=GPTJConfig, n_embd=37)
def test_config(self):
self.config_tester.run_common_tests()
def test_gptj_model(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_gptj_model(*config_and_inputs)
def test_gptj_model_past(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_gptj_model_past(*config_and_inputs)
def test_gptj_model_att_mask_past(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_gptj_model_attention_mask_past(*config_and_inputs)
def test_gptj_model_past_large_inputs(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_gptj_model_past_large_inputs(*config_and_inputs)
def test_gptj_lm_head_model(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_lm_head_model(*config_and_inputs)
def test_gptj_gradient_checkpointing(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs(gradient_checkpointing=True)
self.model_tester.create_and_check_forward_and_backwards(*config_and_inputs)
@slow
def test_batch_generation(self):
model = GPTJForCausalLM.from_pretrained("EleutherAI/gpt-j-6B")
model.to(torch_device)
tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-j-6B")
tokenizer.padding_side = "left"
# Define PAD Token = EOS Token = 50256
tokenizer.pad_token = tokenizer.eos_token
model.config.pad_token_id = model.config.eos_token_id
# use different length sentences to test batching
sentences = [
"Hello, my dog is a little",
"Today, I",
]
inputs = tokenizer(sentences, return_tensors="pt", padding=True)
input_ids = inputs["input_ids"].to(torch_device)
token_type_ids = torch.cat(
[
input_ids.new_full((input_ids.shape[0], input_ids.shape[1] - 1), 0),
input_ids.new_full((input_ids.shape[0], 1), 500),
],
dim=-1,
)
outputs = model.generate(
input_ids=input_ids,
attention_mask=inputs["attention_mask"].to(torch_device),
)
outputs_tt = model.generate(
input_ids=input_ids,
attention_mask=inputs["attention_mask"].to(torch_device),
token_type_ids=token_type_ids,
)
inputs_non_padded = tokenizer(sentences[0], return_tensors="pt").input_ids.to(torch_device)
output_non_padded = model.generate(input_ids=inputs_non_padded)
num_paddings = inputs_non_padded.shape[-1] - inputs["attention_mask"][-1].long().sum().cpu().item()
inputs_padded = tokenizer(sentences[1], return_tensors="pt").input_ids.to(torch_device)
output_padded = model.generate(input_ids=inputs_padded, max_length=model.config.max_length - num_paddings)
batch_out_sentence = tokenizer.batch_decode(outputs, skip_special_tokens=True)
batch_out_sentence_tt = tokenizer.batch_decode(outputs_tt, skip_special_tokens=True)
non_padded_sentence = tokenizer.decode(output_non_padded[0], skip_special_tokens=True)
padded_sentence = tokenizer.decode(output_padded[0], skip_special_tokens=True)
expected_output_sentence = [
"Hello, my dog is a little over a year old and has been diagnosed with a heart murmur",
"Today, I’m going to talk about the most important thing in the",
]
self.assertListEqual(expected_output_sentence, batch_out_sentence)
self.assertTrue(batch_out_sentence_tt != batch_out_sentence) # token_type_ids should change output
self.assertListEqual(expected_output_sentence, [non_padded_sentence, padded_sentence])
@slow
def test_model_from_pretrained(self):
for model_name in GPTJ_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
model = GPTJModel.from_pretrained(model_name)
self.assertIsNotNone(model)
@require_torch
class GPTJModelLanguageGenerationTest(unittest.TestCase):
@slow
def test_lm_generate_gptj(self):
for checkpointing in [True, False]:
model = GPTJForCausalLM.from_pretrained("EleutherAI/gpt-j-6B", gradient_checkpointing=checkpointing)
model.to(torch_device)
input_ids = torch.tensor([[464, 3290]], dtype=torch.long, device=torch_device) # The dog
expected_output_ids = [
464,
3290,
1528,
286,
3931,
389,
2402,
514,
11,
290,
326,
1724,
340,
447,
247,
82,
640,
284,
923,
3612,
] # The dog days of summer are upon us, and that means it’s time to start thinking
output_ids = model.generate(input_ids, do_sample=False)
self.assertListEqual(output_ids[0].tolist(), expected_output_ids)
@slow
def test_gptj_sample(self):
tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-j-6B")
model = GPTJForCausalLM.from_pretrained("EleutherAI/gpt-j-6B")
model.to(torch_device)
torch.manual_seed(0)
tokenized = tokenizer("Today is a nice day and", return_tensors="pt", return_token_type_ids=True)
input_ids = tokenized.input_ids.to(torch_device)
output_ids = model.generate(input_ids, do_sample=True)
output_str = tokenizer.decode(output_ids[0], skip_special_tokens=True)
token_type_ids = tokenized.token_type_ids.to(torch_device)
output_seq = model.generate(input_ids=input_ids, do_sample=True, num_return_sequences=5)
output_seq_tt = model.generate(
input_ids=input_ids, token_type_ids=token_type_ids, do_sample=True, num_return_sequences=5
)
output_seq_strs = tokenizer.batch_decode(output_seq, skip_special_tokens=True)
output_seq_tt_strs = tokenizer.batch_decode(output_seq_tt, skip_special_tokens=True)
EXPECTED_OUTPUT_STR = "Today is a nice day and I've already been enjoying it. I walked to work with my wife"
self.assertEqual(output_str, EXPECTED_OUTPUT_STR)
self.assertTrue(
all([output_seq_strs[idx] != output_seq_tt_strs[idx] for idx in range(len(output_seq_tt_strs))])
) # token_type_ids should change output
@slow
def test_gptj_sample_max_time(self):
tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-j-6B")
model = GPTJForCausalLM.from_pretrained("EleutherAI/gpt-j-6B")
model.to(torch_device)
torch.manual_seed(0)
tokenized = tokenizer("Today is a nice day and", return_tensors="pt", return_token_type_ids=True)
input_ids = tokenized.input_ids.to(torch_device)
MAX_TIME = 0.5
start = datetime.datetime.now()
model.generate(input_ids, do_sample=True, max_time=MAX_TIME, max_length=256)
duration = datetime.datetime.now() - start
self.assertGreater(duration, datetime.timedelta(seconds=MAX_TIME))
self.assertLess(duration, datetime.timedelta(seconds=1.5 * MAX_TIME))
start = datetime.datetime.now()
model.generate(input_ids, do_sample=False, max_time=MAX_TIME, max_length=256)
duration = datetime.datetime.now() - start
self.assertGreater(duration, datetime.timedelta(seconds=MAX_TIME))
self.assertLess(duration, datetime.timedelta(seconds=1.5 * MAX_TIME))
start = datetime.datetime.now()
model.generate(input_ids, do_sample=False, num_beams=2, max_time=MAX_TIME, max_length=256)
duration = datetime.datetime.now() - start
self.assertGreater(duration, datetime.timedelta(seconds=MAX_TIME))
self.assertLess(duration, datetime.timedelta(seconds=1.5 * MAX_TIME))
start = datetime.datetime.now()
model.generate(input_ids, do_sample=True, num_beams=2, max_time=MAX_TIME, max_length=256)
duration = datetime.datetime.now() - start
self.assertGreater(duration, datetime.timedelta(seconds=MAX_TIME))
self.assertLess(duration, datetime.timedelta(seconds=1.5 * MAX_TIME))
start = datetime.datetime.now()
model.generate(input_ids, do_sample=False, max_time=None, max_length=256)
duration = datetime.datetime.now() - start
self.assertGreater(duration, datetime.timedelta(seconds=1.5 * MAX_TIME))