github/autoMate
1
0
Fork 0
mirror of https://github.com/yuruotong1/autoMate.git synced 2025-12-25 21:06:47 +08:00
Code Issues Packages Projects Releases Wiki Activity

继续尝试绕开HF

Browse Source
This commit is contained in:
Dan Li 2025-03-18 02:00:54 +03:00
parent 9c5ede79b1
commit 8af1a66ffc
11 changed files with 274 additions and 327106 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
.DS_Store vendored
View File

Binary file not shown.

13
gradio_ui/agent/vision_agent.py
View File

@ -18,13 +18,13 @@ class UIElement(BaseModel):
text: Optional[str] = None
class VisionAgent:
def __init__(self, yolo_model_path: str, caption_model_path: str = 'microsoft/Florence-2-base-ft'):
def __init__(self, yolo_model_path: str, caption_model_path: str):
"""
Initialize the vision agent
Parameters:
yolo_model_path: Path to YOLO model
caption_model_path: Path to image caption model, default is Florence-2
caption_model_path: Path to image caption model
"""
# determine the available device and the best dtype
self.device, self.dtype = self._get_optimal_device_and_dtype()
@ -33,14 +33,15 @@ class VisionAgent:
# load the image caption model and processor
self.caption_processor = AutoProcessor.from_pretrained(
"processor",
trust_remote_code=True
"weights/AI-ModelScope/Florence-2-base",
trust_remote_code=True,
local_files_only=True
)
try:
self.caption_model = AutoModelForCausalLM.from_pretrained(
caption_model_path,
torch_dtype=torch.float32,
torch_dtype=self.dtype,
trust_remote_code=True
).to(self.device)
@ -53,7 +54,7 @@ class VisionAgent:
if self.device.type == 'cuda':
self.batch_size = 128
elif self.device.type == 'mps':
self.batch_size = 32
self.batch_size = 128
else:
self.batch_size = 16

1026
processor/added_tokens.json
View File

File diff suppressed because it is too large Load Diff

50001
processor/merges.txt
View File

File diff suppressed because it is too large Load Diff

33
processor/preprocessor_config.json
View File

@ -1,33 +0,0 @@
{
"auto_map": {
"AutoProcessor": "microsoft/Florence-2-base--processing_florence2.Florence2Processor"
},
"crop_size": {
"height": 768,
"width": 768
},
"do_center_crop": false,
"do_convert_rgb": null,
"do_normalize": true,
"do_rescale": true,
"do_resize": true,
"image_mean": [
0.485,
0.456,
0.406
],
"image_processor_type": "CLIPImageProcessor",
"image_seq_length": 577,
"image_std": [
0.229,
0.224,
0.225
],
"processor_class": "Florence2Processor",
"resample": 3,
"rescale_factor": 0.00392156862745098,
"size": {
"height": 768,
"width": 768
}
}

7185
processor/special_tokens_map.json
View File

File diff suppressed because it is too large Load Diff

259573
processor/tokenizer.json
View File

File diff suppressed because it is too large Load Diff

9277
processor/tokenizer_config.json
View File

File diff suppressed because it is too large Load Diff

1
processor/vocab.json
View File

File diff suppressed because one or more lines are too long

239
util/config.json Normal file
View File

@ -0,0 +1,239 @@
{
"_name_or_path": "./Florence-2-base-ft",
"architectures": [
"Florence2ForConditionalGeneration"
],
"auto_map": {
"AutoConfig": "configuration_florence2.Florence2Config",
"AutoModelForCausalLM": "modeling_florence2.Florence2ForConditionalGeneration"
},
"bos_token_id": 2,
"eos_token_id": 1,
"ignore_index": -100,
"is_encoder_decoder": true,
"model_type": "florence2",
"pad_token_id": 0,
"projection_dim": 768,
"text_config": {
"_attn_implementation_autoset": true,
"_name_or_path": "",
"activation_dropout": 0.1,
"activation_function": "gelu",
"add_bias_logits": false,
"add_cross_attention": false,
"add_final_layer_norm": false,
"architectures": null,
"attention_dropout": 0.1,
"bad_words_ids": null,
"begin_suppress_tokens": null,
"bos_token_id": 0,
"chunk_size_feed_forward": 0,
"classif_dropout": 0.1,
"classifier_dropout": 0.0,
"cross_attention_hidden_size": null,
"d_model": 768,
"decoder_attention_heads": 12,
"decoder_ffn_dim": 3072,
"decoder_layerdrop": 0.0,
"decoder_layers": 6,
"decoder_start_token_id": 2,
"diversity_penalty": 0.0,
"do_sample": false,
"dropout": 0.1,
"early_stopping": true,
"encoder_attention_heads": 12,
"encoder_ffn_dim": 3072,
"encoder_layerdrop": 0.0,
"encoder_layers": 6,
"encoder_no_repeat_ngram_size": 0,
"eos_token_id": 2,
"exponential_decay_length_penalty": null,
"finetuning_task": null,
"forced_bos_token_id": 0,
"forced_eos_token_id": 2,
"gradient_checkpointing": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1",
"2": "LABEL_2"
},
"init_std": 0.02,
"is_decoder": false,
"is_encoder_decoder": true,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1,
"LABEL_2": 2
},
"length_penalty": 1.0,
"max_length": 20,
"max_position_embeddings": 1024,
"min_length": 0,
"model_type": "florence2_language",
"no_repeat_ngram_size": 3,
"normalize_before": false,
"num_beam_groups": 1,
"num_beams": 3,
"num_hidden_layers": 6,
"num_return_sequences": 1,
"output_attentions": false,
"output_hidden_states": false,
"output_scores": false,
"pad_token_id": 1,
"prefix": null,
"problem_type": null,
"pruned_heads": {},
"remove_invalid_values": false,
"repetition_penalty": 1.0,
"return_dict": true,
"return_dict_in_generate": false,
"scale_embedding": false,
"sep_token_id": null,
"suppress_tokens": null,
"task_specific_params": null,
"temperature": 1.0,
"tf_legacy_loss": false,
"tie_encoder_decoder": false,
"tie_word_embeddings": true,
"tokenizer_class": null,
"top_k": 50,
"top_p": 1.0,
"torch_dtype": null,
"torchscript": false,
"typical_p": 1.0,
"use_bfloat16": false,
"use_cache": true,
"vocab_size": 51289
},
"torch_dtype": "float32",
"transformers_version": "4.46.1",
"vision_config": {
"_attn_implementation_autoset": false,
"_name_or_path": "",
"add_cross_attention": false,
"architectures": null,
"bad_words_ids": null,
"begin_suppress_tokens": null,
"bos_token_id": null,
"chunk_size_feed_forward": 0,
"cross_attention_hidden_size": null,
"decoder_start_token_id": null,
"depths": [
1,
1,
9,
1
],
"dim_embed": [
128,
256,
512,
1024
],
"diversity_penalty": 0.0,
"do_sample": false,
"drop_path_rate": 0.1,
"early_stopping": false,
"enable_checkpoint": false,
"encoder_no_repeat_ngram_size": 0,
"eos_token_id": null,
"exponential_decay_length_penalty": null,
"finetuning_task": null,
"forced_bos_token_id": null,
"forced_eos_token_id": null,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"image_feature_source": [
"spatial_avg_pool",
"temporal_avg_pool"
],
"image_pos_embed": {
"max_pos_embeddings": 50,
"type": "learned_abs_2d"
},
"is_decoder": false,
"is_encoder_decoder": false,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"length_penalty": 1.0,
"max_length": 20,
"min_length": 0,
"model_type": "davit",
"no_repeat_ngram_size": 0,
"num_beam_groups": 1,
"num_beams": 1,
"num_groups": [
4,
8,
16,
32
],
"num_heads": [
4,
8,
16,
32
],
"num_return_sequences": 1,
"output_attentions": false,
"output_hidden_states": false,
"output_scores": false,
"pad_token_id": null,
"patch_padding": [
3,
1,
1,
1
],
"patch_prenorm": [
false,
true,
true,
true
],
"patch_size": [
7,
3,
3,
3
],
"patch_stride": [
4,
2,
2,
2
],
"prefix": null,
"problem_type": null,
"projection_dim": 768,
"pruned_heads": {},
"remove_invalid_values": false,
"repetition_penalty": 1.0,
"return_dict": true,
"return_dict_in_generate": false,
"sep_token_id": null,
"suppress_tokens": null,
"task_specific_params": null,
"temperature": 1.0,
"tf_legacy_loss": false,
"tie_encoder_decoder": false,
"tie_word_embeddings": true,
"tokenizer_class": null,
"top_k": 50,
"top_p": 1.0,
"torch_dtype": null,
"torchscript": false,
"typical_p": 1.0,
"use_bfloat16": false,
"visual_temporal_embedding": {
"max_temporal_embeddings": 100,
"type": "COSINE"
},
"window_size": 12
},
"vocab_size": 51289
}

32
util/download_weights.py
View File

@ -1,8 +1,11 @@
import os
from pathlib import Path
from modelscope import snapshot_download
import subprocess
import shutil
__WEIGHTS_DIR = Path("weights")
MODEL_DIR = os.path.join(__WEIGHTS_DIR, "AI-ModelScope", "OmniParser-v2___0")
MODEL_DIR = os.path.join(__WEIGHTS_DIR, "AI-ModelScope", "OmniParser-v2___0")
PROCESSOR_DIR = os.path.join(__WEIGHTS_DIR, "AI-ModelScope", "Florence-2-base")
def download():
# Create weights directory
@ -13,9 +16,14 @@ def download():
"icon_detect/train_args.yaml",
"icon_detect/model.pt",
"icon_detect/model.yaml",
"icon_caption/config.json",
"icon_caption/generation_config.json",
"icon_caption/model.safetensors"
"icon_caption/model.safetensors",
]
# Extra config files downloaded from Florence2 repo
config_files = [
"configuration_florence2.py",
"modeling_florence2.py"
]
# Check and download missing files
@ -32,9 +40,25 @@ def download():
snapshot_download(
'AI-ModelScope/OmniParser-v2.0',
cache_dir='weights'
cache_dir='weights',
ignore_file_pattern=['config.json']
)
snapshot_download(
'AI-ModelScope/Florence-2-base',
cache_dir='weights',
allow_file_pattern=['*.py', '*.json']
)
# Move downloaded Florence config files into icon_caption
for file_path in config_files:
source_dir = os.path.join(PROCESSOR_DIR, file_path)
dest_dir = os.path.join(MODEL_DIR, "icon_caption", file_path)
shutil.copy(source_dir, dest_dir)
# Move customized config.json into icon_caption to load the model from local path
shutil.copy(os.path.join("util", "config.json"), os.path.join(MODEL_DIR, "icon_caption", "config.json"))
print("Download complete")
if __name__ == "__main__":