File size: 5,645 Bytes
c3f3b0b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
# ------------------------------------------------------------------------
#    Copyright 2023 Haotian Liu
#
#    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.
# ------------------------------------------------------------------------
# Modified from LLaVA (https://github.com/haotian-liu/LLaVA) and S^2(https://github.com/bfshi/scaling_on_scales)
# Copyright 2024 Jiachen Li
# ------------------------------------------------------------------------

import argparse
import torch
import os
import json
from tqdm import tqdm
import shortuuid

from cumo.constants import IMAGE_TOKEN_INDEX, DEFAULT_IMAGE_TOKEN, DEFAULT_IM_START_TOKEN, DEFAULT_IM_END_TOKEN
from cumo.conversation import conv_templates, SeparatorStyle
from cumo.model.builder import load_pretrained_model
from cumo.utils import disable_torch_init
from cumo.mm_utils import tokenizer_image_token, process_images, get_model_name_from_path

from PIL import Image
import math


def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)  # integer division
    return [lst[i:i+chunk_size] for i in range(0, len(lst), chunk_size)]


def get_chunk(lst, n, k):
    chunks = split_list(lst, n)
    return chunks[k]


def eval_model(args):
    # Model
    disable_torch_init()
    model_path = os.path.expanduser(args.model_path)
    model_name = get_model_name_from_path(model_path)
    tokenizer, model, image_processor, context_len = load_pretrained_model(model_path, args.model_base, model_name)
    model.config.training = False
    questions = json.load(open(os.path.expanduser(args.question_file), "r"))
    questions = get_chunk(questions, args.num_chunks, args.chunk_idx)
    answers_file = os.path.expanduser(args.answers_file)
    os.makedirs(os.path.dirname(answers_file), exist_ok=True)
    ans_file = open(answers_file, "w")
    for i, line in enumerate(tqdm(questions)):
        idx = line["id"]
        question = line['conversations'][0]
        qs = question['value'].replace('<image>', '').strip()
        cur_prompt = qs

        if 'image' in line:
            image_file = line["image"]
            image = Image.open(os.path.join(args.image_folder, image_file))
            image_tensor = process_images([image], image_processor, model.config)[0]
            images = image_tensor.unsqueeze(0).half().cuda()
            image_sizes = [image.size]
            if getattr(model.config, 'mm_use_im_start_end', False):
                qs = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN + DEFAULT_IM_END_TOKEN + '\n' + qs
            else:
                qs = DEFAULT_IMAGE_TOKEN + '\n' + qs
            cur_prompt = '<image>' + '\n' + cur_prompt
        else:
            images = None
            image_sizes = None

        if args.single_pred_prompt:
            qs = qs + '\n' + "Answer with the option's letter from the given choices directly."
            cur_prompt = cur_prompt + '\n' + "Answer with the option's letter from the given choices directly."

        conv = conv_templates[args.conv_mode].copy()
        conv.append_message(conv.roles[0], qs)
        conv.append_message(conv.roles[1], None)
        prompt = conv.get_prompt()
        input_ids = tokenizer_image_token(prompt, tokenizer, IMAGE_TOKEN_INDEX, return_tensors='pt').unsqueeze(0).cuda()

        with torch.inference_mode():
            output_ids = model.generate(
                input_ids,
                images=images,
                image_sizes=image_sizes,
                do_sample=True if args.temperature > 0 else False,
                #temperature=args.temperature,
                max_new_tokens=1024,
                pad_token_id=tokenizer.eos_token_id,
                use_cache=True,
            )

        outputs = tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0].strip()

        ans_id = shortuuid.uuid()
        ans_file.write(json.dumps({"question_id": idx,
                                   "prompt": cur_prompt,
                                   "text": outputs,
                                   "answer_id": ans_id,
                                   "model_id": model_name,
                                   "metadata": {}}) + "\n")
        ans_file.flush()
    ans_file.close()

if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--model-path", type=str, default="facebook/opt-350m")
    parser.add_argument("--model-base", type=str, default=None)
    parser.add_argument("--image-folder", type=str, default="")
    parser.add_argument("--question-file", type=str, default="tables/question.json")
    parser.add_argument("--answers-file", type=str, default="answer.jsonl")
    parser.add_argument("--conv-mode", type=str, default="llava_v0")
    parser.add_argument("--num-chunks", type=int, default=1)
    parser.add_argument("--chunk-idx", type=int, default=0)
    parser.add_argument("--temperature", type=float, default=0.2)
    parser.add_argument("--answer-prompter", action="store_true")
    parser.add_argument("--single-pred-prompt", action="store_true")
    args = parser.parse_args()

    eval_model(args)