Skip to content

VLM Pipeline for Model Onboarding through QEff #261

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 4 commits into from
Closed

VLM Pipeline for Model Onboarding through QEff #261

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
da68b09
VLM Pipeline for onboarding of VLMs
qcdipankar Feb 5, 2025
e2199e5
VLM Pipeline for onboarding of VLMs
qcdipankar Feb 5, 2025
9efb97e
Change of conversation import over files
qcdipankar Feb 6, 2025
f55503a
Fixing of intern full model output
qcdipankar Feb 10, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions QEfficient/base/modeling_qeff.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,6 +223,7 @@ def _compile(
if onnx_path is None and self.onnx_path is None:
self.export()

# import ipdb; ipdb.set_trace()
onnx_path = Path(onnx_path or self.onnx_path)
compile_dir = Path(compile_dir or onnx_path.parent)
qpc_path = compile_dir / "qpc"
Expand Down Expand Up @@ -251,6 +252,7 @@ def _compile(
if num_speculative_tokens:
compile_hash.update(to_hashable({"num_speculative_tokens": num_speculative_tokens}))

# import ipdb; ipdb.set_trace()
# Check if already compiled
compile_hash = compile_hash.hexdigest()[:16]
compile_dir = qpc_path.with_name(qpc_path.name + "-" + compile_hash)
Expand Down
9 changes: 9 additions & 0 deletions QEfficient/base/pytorch_transforms.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,12 +40,21 @@ class ModuleMappingTransform(PytorchTransform):
def apply(cls, model: nn.Module) -> Tuple[nn.Module, bool]:
transformed = False
for module in model.modules():
# if repl_module := cls._module_mapping.get(type(module)):
if repl_module := cls._module_mapping.get(module.__class__.__name__):
module.__class__ = repl_module
# Handling the __init__ calls in the models
if hasattr(module, "__qeff_init__"):
module.__qeff_init__()
transformed = True

Comment on lines +43 to +50
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

create a new transform named like:

class ModuleMappingViaStringAndClassMatchTransform:
    _module_mapping_via_class: Dict[Type[nn.Module], Type[nn.Module]]
    _module_mapping_via_string: Dict[string, Type[nn.Module]]
    
    @classmethod
    def apply(cls, model):
        transformed=False
        for module in model.modules():
            if  repl_module := cls._module_mapping_via_class.get(type(module)):
                 # replace the class here
            elif repl_module := cls._module_mapping_via_string.get(type(module)):
                 # replace the class here

Create two different dicts basically.

And write a test that makes sure the keys on the two dicts don't match.

if repl_module := cls._module_mapping.get(type(module)):
module.__class__ = repl_module
# Handling the __init__ calls in the models
if hasattr(module, "__qeff_init__"):
module.__qeff_init__()
transformed = True

return model, transformed

@classmethod
Expand Down
9 changes: 9 additions & 0 deletions QEfficient/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,6 +77,11 @@
)

from QEfficient.customop import CustomRMSNormAIC
from QEfficient.transformers.models.InternVL.modeling_internvl_chat_vision import (
InternVLChatModel,
QEffInternVisionEmbeddings,
QEffInternVLChatModel,
)

from .models.codegen.modeling_codegen import (
QEffCodeGenAttention,
Expand Down Expand Up @@ -157,6 +162,7 @@
Starcoder2ForCausalLM.__name__,
GPTBigCodeForCausalLM.__name__,
MllamaForCausalLM.__name__,
InternVLChatModel.__name__,
]
)

Expand Down Expand Up @@ -241,4 +247,7 @@
GPTBigCodeAttention: QEffGPTBigCodeAttention,
GPTBigCodeBlock: QEffGPTBigCodeBlock,
GPTBigCodeModel: QEffGPTBigCodeModel,
# InternVL
"InternVLChatModel": QEffInternVLChatModel,
"InternVisionEmbeddings": QEffInternVisionEmbeddings,
}
6 changes: 6 additions & 0 deletions QEfficient/transformers/models/InternVL/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# -----------------------------------------------------------------------------
#
# Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved.
# SPDX-License-Identifier: BSD-3-Clause
#
# -----------------------------------------------------------------------------
137 changes: 137 additions & 0 deletions QEfficient/transformers/models/InternVL/config.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,137 @@
{
"_commit_hash": null,
"architectures": [
"InternVLChatModel"
],
"auto_map": {
"AutoConfig": "configuration_internvl_chat.InternVLChatConfig",
"AutoModel": "modeling_internvl_chat.InternVLChatModel",
"AutoModelForCausalLM": "modeling_internvl_chat.InternVLChatModel"
},
"downsample_ratio": 0.5,
"dynamic_image_size": true,
"force_image_size": 448,
"llm_config": {
"_name_or_path": "Qwen/Qwen2.5-0.5B-Instruct",
"add_cross_attention": false,
"architectures": [
"Qwen2ForCausalLM"
],
"_attn_implementation": "eager",
"attention_dropout": 0.0,
"bad_words_ids": null,
"begin_suppress_tokens": null,
"bos_token_id": 151643,
"chunk_size_feed_forward": 0,
"cross_attention_hidden_size": null,
"decoder_start_token_id": null,
"diversity_penalty": 0.0,
"do_sample": false,
"early_stopping": false,
"encoder_no_repeat_ngram_size": 0,
"eos_token_id": 151645,
"exponential_decay_length_penalty": null,
"finetuning_task": null,
"forced_bos_token_id": null,
"forced_eos_token_id": null,
"hidden_act": "silu",
"hidden_size": 896,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"initializer_range": 0.02,
"intermediate_size": 4864,
"is_decoder": false,
"is_encoder_decoder": false,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"length_penalty": 1.0,
"max_length": 20,
"max_position_embeddings": 32768,
"max_window_layers": 21,
"min_length": 0,
"model_type": "qwen2",
"no_repeat_ngram_size": 0,
"num_attention_heads": 14,
"num_beam_groups": 1,
"num_beams": 1,
"num_hidden_layers": 24,
"num_key_value_heads": 2,
"num_return_sequences": 1,
"output_attentions": false,
"output_hidden_states": false,
"output_scores": false,
"pad_token_id": null,
"prefix": null,
"problem_type": null,
"pruned_heads": {},
"remove_invalid_values": false,
"repetition_penalty": 1.0,
"return_dict": true,
"return_dict_in_generate": false,
"rms_norm_eps": 1e-06,
"rope_theta": 1000000.0,
"sep_token_id": null,
"sliding_window": 32768,
"suppress_tokens": null,
"task_specific_params": null,
"temperature": 1.0,
"tf_legacy_loss": false,
"tie_encoder_decoder": false,
"tie_word_embeddings": false,
"tokenizer_class": null,
"top_k": 50,
"top_p": 1.0,
"torch_dtype": "float32",
"torchscript": false,
"transformers_version": "4.37.2",
"typical_p": 1.0,
"use_bfloat16": false,
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 151674
},
"max_dynamic_patch": 12,
"min_dynamic_patch": 1,
"model_type": "internvl_chat",
"ps_version": "v2",
"select_layer": -1,
"template": "internvl2_5",
"torch_dtype": "float32",
"use_backbone_lora": 0,
"use_llm_lora": 0,
"use_thumbnail": true,
"vision_config": {
"architectures": [
"InternVisionModel"
],
"attention_dropout": 0.0,
"drop_path_rate": 0.0,
"dropout": 0.0,
"hidden_act": "gelu",
"hidden_size": 1024,
"image_size": 448,
"initializer_factor": 1.0,
"initializer_range": 0.02,
"intermediate_size": 4096,
"layer_norm_eps": 1e-06,
"model_type": "intern_vit_6b",
"norm_type": "layer_norm",
"num_attention_heads": 16,
"num_channels": 3,
"num_hidden_layers": 24,
"output_attentions": false,
"output_hidden_states": false,
"patch_size": 14,
"qk_normalization": false,
"qkv_bias": true,
"return_dict": true,
"torch_dtype": "float32",
"transformers_version": "4.37.2",
"use_bfloat16": false,
"use_flash_attn": false
}
}
Loading
Loading