Upload folder using huggingface_hub

.gitattributes

@@ -57,3 +57,12 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+data/Cjson/C_generate_multi.jsonl filter=lfs diff=lfs merge=lfs -text
+data/Cjson/C_generate_multi_100.jsonl filter=lfs diff=lfs merge=lfs -text
+data/Cjson/C_generate_multi_all_item.jsonl filter=lfs diff=lfs merge=lfs -text
+data/Cjson/Cjson_emo_off_meta_cot_with_zh_relabel.jsonl filter=lfs diff=lfs merge=lfs -text
+generate/omit/Cjson/C_generate_multi_all_item.jsonl filter=lfs diff=lfs merge=lfs -text
+generate/quickmeme/merged_output.csv filter=lfs diff=lfs merge=lfs -text
+generate/quickmeme/user_input.jsonl filter=lfs diff=lfs merge=lfs -text
+pause_data/all_item/Cjson/C_generate_pause.jsonl filter=lfs diff=lfs merge=lfs -text
+pause_data/three_item/Cjson/C_generate_pause.jsonl filter=lfs diff=lfs merge=lfs -text

TwoStage_inference/twostage_inference.py

@@ -0,0 +1,251 @@
+import os
+import json
+import ast
+from collections import defaultdict
+import torch
+import torchvision.transforms as T
+from PIL import Image
+from torchvision.transforms.functional import InterpolationMode
+from transformers import AutoModel, AutoTokenizer
+
+# 定义常量
+IMAGENET_MEAN = (0.485, 0.456, 0.406)
+IMAGENET_STD = (0.229, 0.224, 0.225)
+SENTIMENT_CATEGORY = ['happiness', 'love', 'anger', 'sorrow', 'fear', 'hate', 'surprise']
+INTENTION_DETECTION = ['interactive', 'expressive', 'entertaining', 'offensive']
+OFFENSIVENESS_DETECTION = ['non-offensive', 'slightly', 'moderately', 'very']
+METAPHOR_CATEGORY = ['image dominant', 'text dominant', 'complementary']
+TARGET_MODALITY = ['image', 'text', 'complementary']
+SOURCE_MODALITY = ['image', 'text', 'complementary']
+
+def build_transform(input_size):
+    MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
+    transform = T.Compose([
+        T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
+        T.Resize((input_size, input_size), interpolation=InterpolationMode.BICUBIC),
+        T.ToTensor(),
+        T.Normalize(mean=MEAN, std=STD)
+    ])
+    return transform
+
+def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
+    best_ratio_diff = float('inf')
+    best_ratio = (1, 1)
+    area = width * height
+    for ratio in target_ratios:
+        target_aspect_ratio = ratio[0] / ratio[1]
+        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
+        if ratio_diff < best_ratio_diff:
+            best_ratio_diff = ratio_diff
+            best_ratio = ratio
+        elif ratio_diff == best_ratio_diff:
+            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
+                best_ratio = ratio
+    return best_ratio
+
+def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
+    orig_width, orig_height = image.size
+    aspect_ratio = orig_width / orig_height
+    target_ratios = set(
+        (i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
+        i * j <= max_num and i * j >= min_num)
+    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])
+    target_aspect_ratio = find_closest_aspect_ratio(
+        aspect_ratio, target_ratios, orig_width, orig_height, image_size)
+    target_width = image_size * target_aspect_ratio[0]
+    target_height = image_size * target_aspect_ratio[1]
+    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]
+    resized_img = image.resize((target_width, target_height))
+    processed_images = []
+    for i in range(blocks):
+        box = (
+            (i % (target_width // image_size)) * image_size,
+            (i // (target_width // image_size)) * image_size,
+            ((i % (target_width // image_size)) + 1) * image_size,
+            ((i // (target_width // image_size)) + 1) * image_size
+        )
+        split_img = resized_img.crop(box)
+        processed_images.append(split_img)
+    assert len(processed_images) == blocks
+    if use_thumbnail and len(processed_images) != 1:
+        thumbnail_img = image.resize((image_size, image_size))
+        processed_images.append(thumbnail_img)
+    return processed_images
+
+def load_image(image_file, input_size=448, max_num=12):
+    image = Image.open(image_file).convert('RGB')
+    transform = build_transform(input_size=input_size)
+    images = dynamic_preprocess(image, image_size=input_size, use_thumbnail=True, max_num=max_num)
+    pixel_values = [transform(image) for image in images]
+    pixel_values = torch.stack(pixel_values)
+    return pixel_values
+
+def calculate_accuracy(pred_list, true_list):
+    """计算准确率"""
+    correct = sum(p == t for p, t in zip(pred_list, true_list))
+    return correct / len(pred_list) if len(pred_list) > 0 else 0
+
+def main():
+    # Step 1: 从 JSONL 文件中读取图像路径
+    jsonl_path = '/mnt/afs/xueyingyi/meme/data/Cjson/Cjson_eval_emo_off_meta_relabel.jsonl'
+    image_paths = []
+    with open(jsonl_path, 'r', encoding='utf-8') as f:
+        for line in f:
+            data = json.loads(line)
+            image_paths.append(data['image'])
+
+    # Step 2: 加载模型和分词器
+    path = '/mnt/afs/xueyingyi/model/ch_meme_v6'
+    model = AutoModel.from_pretrained(
+        path,
+        torch_dtype=torch.bfloat16,
+        low_cpu_mem_usage=True,
+        trust_remote_code=True).eval().cuda()
+    tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True, use_fast=False)
+
+    # Step 3: 第一阶段推理
+    quickmeme = []
+    for image_path in image_paths:
+        try:
+            # 加载图像并预处理
+            pixel_values = load_image(image_path, max_num=12).to(torch.bfloat16).cuda()
+            generation_config = dict(max_new_tokens=1024, do_sample=False)
+
+            # 读取系统提示
+            with open("/mnt/afs/xueyingyi/meme/prompt_classification_emo_off_meta.txt", 'r', encoding='utf-8') as f:
+                system_prompt = f.read()
+            prompt = f"<image>\n{system_prompt}"
+
+            # 调用模型生成元数据
+            meme_emotion = model.chat(tokenizer, pixel_values, prompt, generation_config)
+            meme = ast.literal_eval(meme_emotion)
+            meme["uid"] = image_path
+            print(meme)
+            quickmeme.append(meme)
+
+            # 保存第一阶段结果
+            with open('/mnt/afs/xueyingyi/meme/TwoStage_inference/ch_meme_label.json', 'w') as json_file:
+                json.dump(quickmeme, json_file, ensure_ascii=False, indent=4)
+        except Exception as e:
+            print(f"Error processing {image_path}: {e}")
+            continue
+
+    #Step 4: 第二阶段推理
+    updated_quickmeme = []
+    for meme in quickmeme:
+        try:
+            image_path = meme["uid"]
+            mataphor_dict = {
+                'metaphor_occurrence': f"{meme['metaphor_occurrence']}",
+                'metaphor_category': f"{meme['metaphor_category']}",
+                'target_domain': f"{meme['target_domain']}",
+                'source_domain': f"{meme['source_domain']}",
+                'target_modality': f"{meme['target_modality']}",
+                'source_modality': f"{meme['source_modality']}"
+            }
+
+            # 加载图像并预处理
+            pixel_values = load_image(image_path, max_num=12).to(torch.bfloat16).cuda()
+            generation_config = dict(max_new_tokens=1024, do_sample=False)
+
+            # 读取系统提示
+            with open("/mnt/afs/niuyazhe/data/meme/prompt_classification_emo_off_meta_cot.txt", 'r', encoding='utf-8') as f:
+                system_prompt = f.read()
+            prompt = f"<image>{system_prompt}\nThe metophor in the sequence is{mataphor_dict}"
+
+            # 调用模型生成更新后的元数据
+            updated_meme_emotion = model.chat(tokenizer, pixel_values, prompt, generation_config)
+            emotion = ast.literal_eval(updated_meme_emotion)
+
+            # 更新元数据
+            for key in ['sentiment_category', 'sentiment_degree', 'intention_detection', 'offensiveness_detection']:
+                meme[key] = emotion[key]
+            print(meme)
+            updated_quickmeme.append(meme)
+
+            # 保存第二阶段结果
+            with open('/mnt/afs/xueyingyi/meme/TwoStage_inference/ch_meme_update_label.json', 'w') as json_file:
+                json.dump(updated_quickmeme, json_file, ensure_ascii=False, indent=4)
+        except Exception as e:
+            print(f"Error processing {meme}: {e}")
+            continue
+
+    # Step 5: 评估
+    with open('/mnt/afs/xueyingyi/meme/TwoStage_inference/ch_meme_label.json', 'r', encoding='utf-8') as json_file:
+        new_quickmeme = json.load(json_file)
+    print(len(new_quickmeme))
+    true_labels = {}
+    with open(jsonl_path, 'r', encoding='utf-8') as f:
+        for line in f:
+            data = json.loads(line)
+            image_path = data['image']
+            true_label = ast.literal_eval(data['conversations'][1]['value'])
+            true_labels[image_path] = true_label
+
+    acc_dict = defaultdict(int)
+    sentiment_true_list = []
+    sentiment_pred_list = []
+    intention_true_list = []
+    intention_pred_list = []
+    offensiveness_true_list = []
+    offensiveness_pred_list = []
+    sentiment_dict = defaultdict(int)
+    intention_dict = defaultdict(int)
+    offensiveness_dict = defaultdict(int)
+    sentiment_length_dict = defaultdict(int)
+    intention_length_dict = defaultdict(int)
+    offensiveness_length_dict = defaultdict(int)
+
+    count = 0
+    for pred in new_quickmeme:
+        image_path = pred['uid']
+        if image_path in true_labels:
+            true_label = true_labels[image_path]
+            for key in true_label:
+                if key in pred:
+                    if key == 'sentiment_category':
+                        sentiment_true_list.append(SENTIMENT_CATEGORY.index(true_label[key]))
+                        sentiment_pred_list.append(SENTIMENT_CATEGORY.index(pred[key]))
+                        sentiment_length_dict[true_label[key]] += 1
+                        if pred[key] == true_label[key]:
+                            sentiment_dict[true_label[key]] += 1
+
+                    elif key == 'intention_detection' and pred[key] in INTENTION_DETECTION:
+                        intention_true_list.append(INTENTION_DETECTION.index(true_label[key]))
+                        intention_pred_list.append(INTENTION_DETECTION.index(pred[key]))
+                        intention_length_dict[true_label[key]] += 1
+                        if pred[key] == true_label[key]:
+                            intention_dict[true_label[key]] += 1
+
+                    elif key == 'offensiveness_detection' and pred[key] in OFFENSIVENESS_DETECTION:
+                        offensiveness_true_list.append(OFFENSIVENESS_DETECTION.index(true_label[key]))
+                        offensiveness_pred_list.append(OFFENSIVENESS_DETECTION.index(pred[key]))
+                        offensiveness_length_dict[true_label[key]] += 1
+                        if pred[key] == true_label[key]:
+                            offensiveness_dict[true_label[key]] += 1
+
+                    if pred[key] == true_label[key]:
+                        acc_dict[key] += 1
+            count += 1
+
+    for key in acc_dict:
+        acc_dict[key] /= count
+
+    for key in SENTIMENT_CATEGORY:
+        if sentiment_length_dict[key] > 0:
+            acc_dict[f'sentiment_{key}'] = sentiment_dict[key] / sentiment_length_dict[key]
+
+    for key in INTENTION_DETECTION:
+        if intention_length_dict[key] > 0:
+            acc_dict[f'intention_{key}'] = intention_dict[key] / intention_length_dict[key]
+
+    for key in OFFENSIVENESS_DETECTION:
+        if offensiveness_length_dict[key] > 0:
+            acc_dict[f'offensiveness_{key}'] = offensiveness_dict[key] / offensiveness_length_dict[key]
+
+    print("Accuracy Results:")
+    for key, value in acc_dict.items():
+        print(f"{key}: {value:.4f}")
+
+if __name__ == "__main__":
+    main()

augment.py

@@ -0,0 +1,203 @@
+import math
+import os
+from typing import Optional, Tuple, Union
+
+import torch
+from torchvision import transforms
+
+from PIL import Image
+
+from timm.data.constants import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, DEFAULT_CROP_PCT
+from timm.data.auto_augment import rand_augment_transform, augment_and_mix_transform, auto_augment_transform
+from timm.data.transforms import str_to_interp_mode, str_to_pil_interp, RandomResizedCropAndInterpolation, \
+    ResizeKeepRatio, CenterCropOrPad, RandomCropOrPad, TrimBorder, ToNumpy, MaybeToTensor, MaybePILToTensor
+from timm.data.random_erasing import RandomErasing
+
+def augment_meme(img_size: Union[int, Tuple[int, int]] = 224,
+        scale: Optional[Tuple[float, float]] = None,
+        ratio: Optional[Tuple[float, float]] = None,
+        train_crop_mode: Optional[str] = None,
+        color_jitter: Union[float, Tuple[float, ...]] = 0.4,
+        color_jitter_prob: Optional[float] = None,
+        force_color_jitter: bool = False,
+        grayscale_prob: float = 0.,
+        gaussian_blur_prob: float = 0.,
+        auto_augment: Optional[str] = None,
+        interpolation: str = 'random',
+        mean: Tuple[float, ...] = IMAGENET_DEFAULT_MEAN,
+        std: Tuple[float, ...] = IMAGENET_DEFAULT_STD,
+        re_prob: float = 0.,
+        re_mode: str = 'const',
+        re_count: int = 1,
+        re_num_splits: int = 0,
+        use_prefetcher: bool = False,
+        normalize: bool = True,
+        separate: bool = False,
+        hflip = 0,
+        vflip = 0
+        ):
+
+    train_crop_mode = train_crop_mode or 'rrc'
+    assert train_crop_mode in {'rrc', 'rkrc', 'rkrr'}
+    if train_crop_mode in ('rkrc', 'rkrr'):
+        # FIXME integration of RKR is a WIP
+        scale = tuple(scale or (0.8, 1.00))
+        ratio = tuple(ratio or (0.9, 1/.9))
+        primary_tfl = [
+            ResizeKeepRatio(
+                img_size,
+                interpolation=interpolation,
+                random_scale_prob=0.5,
+                random_scale_range=scale,
+                random_scale_area=True,  # scale compatible with RRC
+                random_aspect_prob=0.5,
+                random_aspect_range=ratio,
+            ),
+            CenterCropOrPad(img_size, padding_mode='reflect')
+            if train_crop_mode == 'rkrc' else
+            RandomCropOrPad(img_size, padding_mode='reflect')
+        ]
+    else:
+        scale = tuple(scale or (0.08, 1.0))  # default imagenet scale range
+        ratio = tuple(ratio or (3. / 4., 4. / 3.))  # default imagenet ratio range
+        primary_tfl = []
+    if hflip > 0.:
+        primary_tfl += [transforms.RandomHorizontalFlip(p=hflip)]
+    if vflip > 0.:
+        primary_tfl += [transforms.RandomVerticalFlip(p=vflip)]
+
+    secondary_tfl = []
+    disable_color_jitter = False
+    if auto_augment:
+        assert isinstance(auto_augment, str)
+        # color jitter is typically disabled if AA/RA on,
+        # this allows override without breaking old hparm cfgs
+        disable_color_jitter = not (force_color_jitter or '3a' in auto_augment)
+        if isinstance(img_size, (tuple, list)):
+            img_size_min = min(img_size)
+        else:
+            img_size_min = img_size
+        aa_params = dict(
+            translate_const=int(img_size_min * 0.45),
+            img_mean=tuple([min(255, round(255 * x)) for x in mean]),
+        )
+        if interpolation and interpolation != 'random':
+            aa_params['interpolation'] = str_to_pil_interp(interpolation)
+        if auto_augment.startswith('rand'):
+            secondary_tfl += [rand_augment_transform(auto_augment, aa_params)]
+        elif auto_augment.startswith('augmix'):
+            aa_params['translate_pct'] = 0.3
+            secondary_tfl += [augment_and_mix_transform(auto_augment, aa_params)]
+        else:
+            secondary_tfl += [auto_augment_transform(auto_augment, aa_params)]
+
+    if color_jitter is not None and not disable_color_jitter:
+        # color jitter is enabled when not using AA or when forced
+        if isinstance(color_jitter, (list, tuple)):
+            # color jitter should be a 3-tuple/list if spec brightness/contrast/saturation
+            # or 4 if also augmenting hue
+            assert len(color_jitter) in (3, 4)
+        else:
+            # if it's a scalar, duplicate for brightness, contrast, and saturation, no hue
+            color_jitter = (float(color_jitter),) * 3
+        if color_jitter_prob is not None:
+            secondary_tfl += [
+                transforms.RandomApply([
+                        transforms.ColorJitter(*color_jitter),
+                    ],
+                    p=color_jitter_prob
+                )
+            ]
+        else:
+            secondary_tfl += [transforms.ColorJitter(*color_jitter)]
+
+    if grayscale_prob:
+        secondary_tfl += [transforms.RandomGrayscale(p=grayscale_prob)]
+
+    if gaussian_blur_prob:
+        secondary_tfl += [
+            transforms.RandomApply([
+                    transforms.GaussianBlur(kernel_size=23),  # hardcoded for now
+                ],
+                p=gaussian_blur_prob,
+            )
+        ]
+
+    final_tfl = []
+    if use_prefetcher:
+        # prefetcher and collate will handle tensor conversion and norm
+        final_tfl += [ToNumpy()]
+    elif not normalize:
+        # when normalize disable, converted to tensor without scaling, keeps original dtype
+        final_tfl += [MaybePILToTensor()]
+    else:
+        final_tfl += [
+            MaybeToTensor(),
+            transforms.Normalize(
+                mean=torch.tensor(mean),
+                std=torch.tensor(std),
+            ),
+        ]
+        if re_prob > 0.:
+            final_tfl += [
+                RandomErasing(
+                    re_prob,
+                    mode=re_mode,
+                    max_count=re_count,
+                    num_splits=re_num_splits,
+                    device='cpu',
+                )
+            ]
+
+    if separate:
+        return transforms.Compose(secondary_tfl), transforms.Compose(final_tfl)
+    else:
+        return transforms.Compose(primary_tfl + secondary_tfl)
+
+
+transform_meme = augment_meme(
+    img_size=224,
+    color_jitter=0.4,
+    grayscale_prob=0.2,
+    gaussian_blur_prob=0.5,
+    auto_augment='rand-m9-mstd0.5',
+    normalize=True
+)
+
+
+# 遍历输入目录下的所有文件
+input_dir = '/mnt/afs/niuyazhe/data/meme/data/Cimages/Cimages/Cimages/'
+N = 3
+for filename in os.listdir(input_dir):
+    # 检查文件是否为图片文件
+    if filename.lower().endswith(('.png', '.jpg', '.jpeg', '.bmp', '.gif')) and filename.lower()[-5]==')':
+        # 构建输入文件的完整路径
+        input_path = os.path.join(input_dir, filename)
+
+        # 打开图片
+        try:
+            image = Image.open(input_path).convert('RGB')
+        except Exception as e:
+            print(e)
+            continue
+
+        # 对图片进行 N 次增强
+        for i in range(N):
+            # 应用变换
+            try:
+                transformed_image = transform_meme(image)
+                # 将张量转换回 PIL 图像
+                # transformed_image = transforms.ToPILImage()(transformed_image)
+
+                # 构建输出文件的完整路径
+                base_name, ext = os.path.splitext(filename)
+                output_filename = f"{base_name}_{i}{ext}"
+                output_path = os.path.join(input_dir, output_filename)
+
+                # 保存变换后的图片
+                transformed_image.save(output_path)
+                print(f"Saved {output_path}")
+
+            except Exception as e:
+                print(e)
+                continue

compare/select_eval_data.py

@@ -0,0 +1,29 @@
+import json
+
+# 定义文件路径
+file1_path = '/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval_multi_100.jsonl'
+file2_path = '/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval_multi.jsonl'
+output_path = '/mnt/afs/xueyingyi/meme/compare/eval.jsonl'
+
+# 读取第一个文件并存储第二个image的值
+image_set = set()
+with open(file1_path, 'r') as file1:
+    for line in file1:
+        data = json.loads(line)
+        if len(data['image']) > 1:
+            image_set.add(data['image'][1])
+
+# 读取第二个文件并比较第二个image的值
+matching_data = []
+with open(file2_path, 'r') as file2:
+    for line in file2:
+        data = json.loads(line)
+        if len(data['image']) > 1 and data['image'][1] in image_set:
+            matching_data.append(data)
+
+# 将匹配的数据写入新的JSONL文件
+with open(output_path, 'w') as output_file:
+    for data in matching_data:
+        output_file.write(json.dumps(data) + '\n')
+
+print(f"匹配的数据已保存到 {output_path}")

data/C_generate_eval.jsonl

@@ -0,0 +1 @@
+{"classification_C": {"root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo", "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval.jsonl", "data_augment": false, "repeat_time": 1, "length": 400}}

data/C_generate_eval_multi.jsonl

@@ -0,0 +1,9 @@
+{
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval_multi.jsonl",
+        "data_augment": false,
+        "repeat_time": 1,
+        "length": 400
+    }
+}

data/C_generate_eval_multi_100.jsonl

@@ -0,0 +1,9 @@
+{
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval_multi_100.jsonl",
+        "data_augment": false,
+        "repeat_time": 1,
+        "length": 400
+    }
+}

data/C_generate_eval_multi_all_item.jsonl

@@ -0,0 +1,9 @@
+{
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval_multi_all_item.jsonl",
+        "data_augment": false,
+        "repeat_time": 1,
+        "length": 400
+    }
+}

data/C_generate_train.jsonl

@@ -0,0 +1 @@
+{"classification_C": {"root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo", "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_train.jsonl", "data_augment": false, "repeat_time": 1, "length": 3592}}

data/C_generate_train_multi.jsonl

@@ -0,0 +1,9 @@
+{
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_train_multi.jsonl",
+        "data_augment": false,
+        "repeat_time": 1,
+        "length": 3592
+    }
+}

data/C_generate_train_multi_100.jsonl

@@ -0,0 +1,9 @@
+{
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_train_multi_100.jsonl",
+        "data_augment": false,
+        "repeat_time": 1,
+        "length": 3592
+    }
+}

data/C_generate_train_multi_all_item.jsonl

@@ -0,0 +1,9 @@
+{
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_train_multi_all_item.jsonl",
+        "data_augment": false,
+        "repeat_time": 1,
+        "length": 3592
+    }
+}

data/Cjson/C_generate_multi.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41c02372e8467032b410df6d8647b882cb64506db72bb88661601218da1fb110
+size 10904214

data/Cjson/C_generate_multi_100.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee04f04226c1b3d7478110b655e3c00713dad5a7b913c6125d8291d28e4e7da2
+size 10984054

data/Cjson/C_generate_multi_all_item.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49ed1eb86d150f34cdd991c01332a65edac56dc8bd1e50de84733b41a49b691d
+size 11357727

data/Cjson/Cjson_emo_off_meta_cot_with_zh_relabel.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57e597208314bb3afdeb9d8e7b6dc891d05111ad9cfeb79142114d89ecf903e9
+size 24658330

data/data_C_json_cot_with_zh_relabel.jsonl

@@ -0,0 +1 @@
+{"classification_C": {"root": "/mnt/afs/niuyazhe/data/meme/data/Cimages/Cimages/Cimages/", "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/Cjson_emo_off_meta_cot_with_zh_relabel.jsonl", "data_augment": false, "repeat_time": 1, "length": 11744}}

data/data_C_json_eval_cot_with_zh_relabel.jsonl

@@ -0,0 +1 @@
+{"classification_C": {"root": "/mnt/afs/niuyazhe/data/meme/data/Cimages/Cimages/Cimages/", "annotation": "//mnt/afs/xueyingyi/meme/data/Cjson/Cjson_eval_emo_off_meta_cot_with_zh_relabel.jsonl", "data_augment": false, "repeat_time": 1, "length": 1272}}

data/data_C_json_eval_relabel.jsonl

@@ -0,0 +1 @@
+{"classification_C": {"root": "/mnt/afs/niuyazhe/data/meme/data/Cimages/Cimages/Cimages/", "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/Cjson_eval_emo_off_meta_relabel.jsonl", "data_augment": false, "repeat_time": 1, "length": 318}}

data/data_C_json_relabel.jsonl

@@ -0,0 +1 @@
+{"classification_C": {"root": "/mnt/afs/niuyazhe/data/meme/data/Cimages/Cimages/Cimages/", "annotation": "/mnt/afs/xueyingyi/meme/data/Cjson/Cjson_emo_off_meta_relabel.jsonl", "data_augment": false, "repeat_time": 1, "length": 2936}}

data/label_C_evaluate.csv

@@ -0,0 +1,605 @@
+Image_(3482).jpg,3(anger),3(very),4(offensive),3(very),0,,,,,
+Image_(4321).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2473).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1208).jpg,3(anger),1(slightly),3(entertaining),0(non-offensive),1,text dominant,胡扯,壶扯,text,complementary
+Image_(5462).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(500).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),1,text dominant,掌上明珠,掌上明猪,text,image
+Image_(1093).jpg,1(happiness),2(moderately),2(expressive),1(slightly),1,text dominant,傻屌,沙雕,text,text
+Image_(1227).jpg,3(anger),2(moderately),4(offensive),3(very),1,text dominant,是,柿,text,complementary
+Image_(181).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,text dominant,谈恋爱联系我,cpdd,text,text
+Image_(2884).jpg,6(hate),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(4297).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3759).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(1799).jpg,3(anger),3(very),4(offensive),2(moderately),1,text dominant,操你妈,草泥马,text,text
+Image_(3299).jpg,1(happiness),1(slightly),4(offensive),0(non-offensive),0,,,,,
+Image_(2053).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,肛门,菊花,text,text
+Image_(4269).jpg,3(anger),2(moderately),4(offensive),1(slightly),0,,,,,
+Image_(5162).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5860).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),1,text dominant,慈爱欣慰的笑,姨母笑,text,text
+Image_(5786).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(859).jpg,6(hate),3(very),4(offensive),3(very),1,text dominant,特么；神经病,他妈；蛇精病,text,text
+Image_(5593).jpg,2(love),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(1452).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,脚,jio,complementary,text
+Image_(5719).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(3127).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(2247).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),1,complementary,我,猫,text,image
+Image_(857).jpg,6(hate),1(slightly),4(offensive),2(moderately),1,text dominant,神经病,蛇精病,text,text
+Image_(5625).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(512).jpg,4(sorrow),1(slightly),3(entertaining),0(non-offensive),1,complementary,废物,猫,text,image
+Image_(5670).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3572).jpg,3(anger),1(slightly),4(offensive),3(very),0,,,,,
+Image_(1809).jpg,5(fear),1(slightly),1(interactive),0(non-offensive),1,text dominant,逼逼,bb,text,text
+Image_(1211).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,沦陷,轮陷,text,complementary
+Image_(3531).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2495).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(540).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,屁,句子,text,text
+Image_(751).jpg,5(fear),2(moderately),1(interactive),0(non-offensive),1,text dominant,服,佛,text,text
+Image_(674).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,image dominant,生活,人,text,image
+Image_(2744).jpg,6(hate),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(2313).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(606).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,打死,打洗,complementary,text
+Image_(5170).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1157).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),1,text dominant,知道,吱道,text,complementary
+Image_(2398).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4776).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2738).jpg,5(fear),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3240).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),1,text dominant,慌,方,text,text
+Image_(5830).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(808).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,text dominant,好,耐思,complementary,text
+Image_(1375).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,text dominant,无原则讨好他人的人,舔狗,text,text
+Image_(5159).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1471).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,complementary,终,钟,text,complementary
+Image_(5605).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(240).jpg,3(anger),1(slightly),1(interactive),1(slightly),1,complementary,你,鸟,text,image
+Image_(2975).jpg,2(love),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(3427).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(3392).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(666).jpg,3(anger),2(moderately),4(offensive),1(slightly),1,text dominant,多损,夺笋,text,text
+Image_(4331).jpg,4(sorrow),2(moderately),3(entertaining),0(non-offensive),0,,,,,
+Image_(4627).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5645).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1393).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,工作或上学,搬砖,text,complementary
+Image_(3214).jpg,7(surprise),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(2642).jpg,7(surprise),2(moderately),4(offensive),1(slightly),0,,,,,
+Image_(2700).jpg,3(anger),2(moderately),4(offensive),3(very),0,,,,,
+Image_(5976).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2299).jpg,7(surprise),2(moderately),1(interactive),1(slightly),1,text dominant,单身的人,单身狗,text,text
+Image_(1715).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,text dominant,我,劳资,text,text
+Image_(5494).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(719).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),1,complementary,呀,鸭,text,complementary
+Image_(180).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,操他妈的,CTMD,text,text
+Image_(4017).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(166).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,不玩了,玩你妈,text,text
+Image_(2582).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1170).jpg,1(happiness),1(slightly),4(offensive),1(slightly),1,complementary,垃圾,辣鸡,text,complementary
+Image_(5877).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(4815).jpg,4(sorrow),1(slightly),5(other),0(non-offensive),0,,,,,
+Image_(5393).jpg,2(love),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(2756).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2654).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(365).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,complementary,猪,你,image,text
+Image_(1618).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,text dominant,就是;牛逼,94;牛啤,text,image
+Image_(232).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),1,complementary,蔡徐坤；吃,葵葵；次,complementary,text
+Image_(827).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,text dominant,薪狠手辣,心狠手辣,text,text
+Image_(2783).jpg,6(hate),3(very),4(offensive),3(very),0,,,,,
+Image_(5154).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1015).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,我操你妈,握草泥马,text,complementary
+Image_(2801).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(210).jpg,6(hate),2(moderately),2(expressive),1(slightly),1,complementary,猪,你,image,text
+Image_(3927).jpg,1(happiness),2(moderately),3(entertaining),0(non-offensive),0,,,,,
+Image_(3940).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1438).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),1,text dominant,谈恋爱联系我,cpdd,text,text
+Image_(1541).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,complementary,算,蒜,text,complementary
+Image_(1516).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),1,text dominant,呼呼,fufu,text,text
+Image_(5349).jpg,6(hate),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(4238).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2412).jpg,2(love),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(312).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),1,complementary,慌,方,text,complementary
+Image_(1016).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,操,草,text,complementary
+Image_(1955).jpg,5(fear),3(very),2(expressive),1(slightly),1,text dominant,滚,爬,text,text
+Image_(1012).jpg,5(fear),1(slightly),2(expressive),0(non-offensive),1,text dominant,投降,投翔,text,complementary
+Image_(137).jpg,7(surprise),1(slightly),3(entertaining),0(non-offensive),1,complementary,小黄鸭,周冬雨,image,text
+Image_(4050).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5458).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1140).jpg,1(happiness),1(slightly),4(offensive),1(slightly),1,text dominant,损,笋,text,text
+Image_(6029).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3119).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(5251).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1243).jpg,3(anger),1(slightly),2(expressive),1(slightly),1,text dominant,精神上剧变;特么,黑化;他妈,text,text
+Image_(2049).jpg,1(happiness),1(slightly),4(offensive),3(very),1,text dominant,肛门,菊花,text,text
+Image_(5084).jpg,3(anger),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(2355).jpg,2(love),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(1861).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,出轨,绿帽子,text,image
+Image_(3810).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5274).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3473).jpg,1(happiness),1(slightly),3(entertaining),1(slightly),0,,,,,
+Image_(248).jpg,6(hate),1(slightly),1(interactive),1(slightly),1,text dominant,谈恋爱时的备选项,备,text,text
+Image_(4504).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5406).jpg,4(sorrow),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(3081).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(505).jpg,3(anger),2(moderately),4(offensive),3(very),1,complementary,杀,鲨,complementary,text
+Image_(5207).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4300).jpg,3(anger),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(2578).jpg,6(hate),2(moderately),2(expressive),1(slightly),0,,,,,
+Image_(5769).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5103).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(813).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,text dominant,多损,夺笋,text,text
+Image_(756).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),1,text dominant,不可思议,book思议,text,text
+Image_(84).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,text dominant,呀,鸭,text,complementary
+Image_(7).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,垃圾,你,text,text
+Image_(2498).jpg,7(surprise),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(3378).jpg,3(anger),3(very),4(offensive),1(slightly),0,,,,,
+Image_(5841).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1670).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,text dominant,傻屌,沙雕,text,text
+Image_(626).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,text dominant,铁子,铁汁,text,complementary
+Image_(1688).jpg,3(anger),2(moderately),4(offensive),3(very),1,text dominant,逼,B,text,text
+Image_(3274).jpg,6(hate),1(slightly),4(offensive),2(moderately),0,,,,,
+Image_(1313).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),1,text dominant,去死,狗带,text,text
+Image_(1059).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,老子,劳资,text,text
+Image_(1800).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,妈,马,text,image
+Image_(4205).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4618).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5911).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(4943).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3327).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4852).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5716).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1609).jpg,1(happiness),2(moderately),4(offensive),1(slightly),1,complementary,猪,楼上,image,text
+Image_(3040).jpg,3(anger),3(very),2(expressive),3(very),0,,,,,
+Image_(5595).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5467).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2560).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1333).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),1,text dominant,获取东西却不付出代价,白嫖,text,text
+Image_(2401).jpg,3(anger),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(6008).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1912).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,complementary,网抑云,网易云,text,complementary
+Image_(4402).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(806).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,text dominant,好,奶思,text,text
+Image_(5420).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5384).jpg,3(anger),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1226).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),1,complementary,牛奶,我,text,complementary
+Image_(1825).jpg,2(love),1(slightly),1(interactive),0(non-offensive),1,complementary,茶；日食,java；java图标,complementary,complementary
+Image_(238).jpg,3(anger),2(moderately),4(offensive),3(very),1,text dominant,傻瓜,傻叉,text,complementary
+Image_(1935).jpg,4(sorrow),1(slightly),3(entertaining),1(slightly),1,text dominant,男性生殖器官,牛子,text,complementary
+Image_(5997).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5532).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1717).jpg,7(surprise),3(very),2(expressive),0(non-offensive),1,text dominant,老子,劳资,text,text
+Image_(2901).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2303).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(805).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,难受,蓝瘦,text,text
+Image_(757).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),1,text dominant,深藏不露,深藏blue,text,text
+Image_(5944).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(3018).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2023).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,complementary,厉害,大拇指,text,image
+Image_(5695).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3244).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(2739).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(6036).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2692).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3163).jpg,3(anger),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(807).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,text dominant,好,奶思,complementary,text
+Image_(3025).jpg,2(love),2(moderately),1(interactive),1(slightly),0,,,,,
+Image_(1930).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),1,text dominant,互联网上浏览信息,冲浪,text,complementary
+Image_(682).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,image dominant,生活；我,杠铃；人,text,image
+Image_(3038).jpg,3(anger),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(1731).jpg,3(anger),3(very),4(offensive),2(moderately),1,text dominant,特么,TM,text,text
+Image_(2545).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(1259).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),1,text dominant,无原则讨好他人的人,舔狗,text,text
+Image_(2779).jpg,6(hate),3(very),4(offensive),2(moderately),0,,,,,
+Image_(1209).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),1,text dominant,凭,苹,text,complementary
+Image_(2604).jpg,2(love),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(2275).jpg,2(love),3(very),1(interactive),0(non-offensive),1,complementary,爱情,玫瑰花,text,image
+Image_(2620).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2152).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,complementary,我闺蜜,老鼠,text,image
+Image_(1664).jpg,3(anger),2(moderately),2(expressive),0(non-offensive),1,complementary,不;你,布;泥,text,complementary
+Image_(1915).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),1,complementary,股票涨,红色,text,complementary
+Image_(2181).jpg,6(hate),3(very),4(offensive),2(moderately),1,complementary,人；畜生；祁醉,直线方向;右转;车,text,image
+Image_(4102).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2175).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,complementary,力量,团结,complementary,complementary
+Image_(4022).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(817).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,难,南,text,text
+Image_(1347).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),1,text dominant,获取东西却不付出代价的人,白嫖怪,text,text
+Image_(100).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),1,text dominant,呀,鸭,text,complementary
+Image_(4229).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4394).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2321).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4816).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(341).jpg,7(surprise),3(very),4(offensive),3(very),1,text dominant,被出轨,戴绿帽子,text,text
+Image_(4708).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(3494).jpg,3(anger),3(very),4(offensive),3(very),0,,,,,
+Image_(3540).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(619).jpg,5(fear),1(slightly),1(interactive),0(non-offensive),1,text dominant,饶；吧,摇；8,text,text
+Image_(5449).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(712).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),1,complementary,呀,鸭,text,complementary
+Image_(1449).jpg,7(surprise),1(slightly),4(offensive),2(moderately),1,text dominant,你妈,nm,text,text
+Image_(5741).jpg,6(hate),1(slightly),1(interactive),1(slightly),0,,,,,
+Image_(5192).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4806).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3206).jpg,2(love),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(4686).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(2854).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(3283).jpg,6(hate),1(slightly),4(offensive),2(moderately),0,,,,,
+Image_(5133).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2721).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5881).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2774).jpg,2(love),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(1106).jpg,3(anger),1(slightly),4(offensive),2(moderately),1,text dominant,傻屌,鲨雕,text,complementary
+Image_(3623).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5580).jpg,7(surprise),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(3295).jpg,1(happiness),2(moderately),4(offensive),3(very),0,,,,,
+Image_(2793).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1775).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,text dominant,牛逼,留批,text,text
+Image_(1555).jpg,1(happiness),3(very),2(expressive),0(non-offensive),1,text dominant,超,敲,text,text
+Image_(4812).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(4648).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4541).jpg,6(hate),2(moderately),2(expressive),1(slightly),0,,,,,
+Image_(3344).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2988).jpg,3(anger),3(very),4(offensive),2(moderately),0,,,,,
+Image_(1804).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,text dominant,壳,阔,text,text
+Image_(2635).jpg,2(love),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(729).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,complementary,注,猪,text,complementary
+Image_(2810).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2891).jpg,6(hate),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(3717).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5669).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3433).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1013).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,操你妈,敲里吗,text,complementary
+Image_(3342).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4803).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4428).jpg,6(hate),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(3480).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2996).jpg,7(surprise),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4087).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5827).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4108).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2348).jpg,2(love),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(3014).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3266).jpg,3(anger),3(very),4(offensive),3(very),0,,,,,
+Image_(5128).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(646).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),1,text dominant,好自为之,耗子尾汁,text,complementary
+Image_(4678).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1730).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),1,text dominant,可乐,阔落,text,text
+Image_(1406).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,complementary,生活;我,手;点心,text,image
+Image_(4880).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(695).jpg,6(hate),2(moderately),4(offensive),2(moderately),1,text dominant,垃圾,辣鸡,text,text
+Image_(2541).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1643).jpg,3(anger),2(moderately),4(offensive),3(very),1,text dominant,操你妈,草泥马,text,text
+Image_(2081).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,text dominant,笨蛋,2+248,text,text
+Image_(5918).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(968).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,做危险的事,玩火,text,text
+Image_(5782).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(1263).jpg,3(anger),1(slightly),3(entertaining),0(non-offensive),1,text dominant,无原则讨好他人的人,舔狗,text,text
+Image_(50).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),1,complementary,学习；我,狗；狗,text,image
+Image_(495).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),1,text dominant,虚假的友谊,塑料花友谊,text,text
+Image_(5820).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5705).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(615).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,image dominant,人,鸡,image,image
+Image_(644).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),1,text dominant,好自为之,耗子煨汁,text,text
+Image_(4420).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4912).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1615).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,text dominant,牛逼,牛批,text,text
+Image_(4749).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(126).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,text dominant,呀,鸭,text,text
+Image_(1868).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),1,complementary,出轨,绿帽子,text,image
+Image_(1535).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,complementary,焦头烂额,蕉头烂额,text,complementary
+Image_(3109).jpg,7(surprise),1(slightly),1(interactive),1(slightly),0,,,,,
+Image_(4869).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3919).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3931).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(952).jpg,1(happiness),2(moderately),3(entertaining),0(non-offensive),1,text dominant,造化弄人,皂滑弄人,text,text
+Image_(3567).jpg,6(hate),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1250).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,text dominant,自抱自泣,自暴自弃,text,text
+Image_(2895).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(119).jpg,1(happiness),3(very),2(expressive),0(non-offensive),1,text dominant,垃圾,辣鸡,text,text
+Image_(522).jpg,2(love),3(very),1(interactive),1(slightly),1,text dominant,特么,他妈,text,text
+Image_(452).jpg,1(happiness),1(slightly),4(offensive),1(slightly),1,complementary,猪,你,complementary,text
+Image_(1251).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,精神上剧变,黑化,text,text
+Image_(804).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,难过,蓝过,text,text
+Image_(889).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,强人所难,强人锁男,text,complementary
+Image_(1468).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),1,text dominant,对不起,骚凹瑞,text,text
+Image_(1952).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,text dominant,花大量精力做事,肝,text,text
+Image_(2815).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(598).jpg,3(anger),1(slightly),1(interactive),1(slightly),1,text dominant,打死,打洗,text,text
+Image_(4994).jpg,1(happiness),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(3126).jpg,3(anger),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(2170).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),1,image dominant,猪,云,image,image
+Image_(5444).jpg,1(happiness),2(moderately),1(interactive),1(slightly),0,,,,,
+Image_(4848).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2103).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,complementary,坟墓,婚姻,complementary,complementary
+Image_(891).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),1,text dominant,与我无关,雨我无瓜,text,text
+Image_(1876).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,complementary,哭,枯,text,complementary
+Image_(3894).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2390).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5676).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5361).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3459).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1701).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,逼逼,bb,text,text
+Image_(5249).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(278).jpg,6(hate),3(very),4(offensive),3(very),1,text dominant,男性生殖器官；男性生殖器官,鸡鸡；蛋,text,text
+Image_(3361).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4015).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2337).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(366).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),1,text dominant,厉害,牛啤,text,image
+Image_(5551).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3060).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5570).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4324).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4178).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5009).jpg,1(happiness),2(moderately),3(entertaining),0(non-offensive),0,,,,,
+Image_(1169).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,complementary,谢,蟹,text,complementary
+Image_(2579).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5120).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1761).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,text dominant,比心,笔芯,complementary,text
+Image_(3516).jpg,6(hate),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(4730).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(2553).jpg,6(hate),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(989).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,好自为之,耗子尾汁,text,text
+Image_(2969).jpg,6(hate),2(moderately),4(offensive),1(slightly),0,,,,,
+Image_(3552).jpg,5(fear),2(moderately),2(expressive),1(slightly),0,,,,,
+Image_(1595).jpg,1(happiness),1(slightly),4(offensive),2(moderately),1,text dominant,狗,成员,text,text
+Image_(2871).jpg,5(fear),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1420).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),1,complementary,哈,蛤,text,complementary
+Image_(6038).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3205).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4680).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3158).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4426).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4455).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5794).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2045).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),1,complementary,肥宅快乐水,可乐,text,complementary
+Image_(4281).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(790).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,难受；想哭,蓝瘦；香菇,text,text
+Image_(3953).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(3349).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(3228).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(2146).jpg,2(love),1(slightly),3(entertaining),0(non-offensive),1,image dominant,金钱,时间,image,image
+Image_(3772).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(177).jpg,5(fear),1(slightly),2(expressive),0(non-offensive),1,complementary,不行,达咩,complementary,text
+Image_(5317).jpg,6(hate),3(very),4(offensive),2(moderately),0,,,,,
+Image_(208).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,complementary,后悔药,药片,text,image
+Image_(4781).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2308).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1926).jpg,1(happiness),3(very),2(expressive),0(non-offensive),1,text dominant,互联网上浏览信息,冲浪,text,complementary
+Image_(3695).jpg,6(hate),3(very),4(offensive),3(very),0,,,,,
+Image_(5511).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3264).jpg,3(anger),1(slightly),4(offensive),3(very),0,,,,,
+Image_(3648).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2870).jpg,1(happiness),1(slightly),1(interactive),1(slightly),0,,,,,
+Image_(3797).jpg,3(anger),2(moderately),1(interactive),1(slightly),0,,,,,
+Image_(3369).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2128).jpg,6(hate),2(moderately),4(offensive),1(slightly),1,complementary,珍珠奶茶,羊粪,complementary,image
+Image_(5704).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(518).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,image dominant,诱惑,放电,text,image
+Image_(5755).jpg,6(hate),2(moderately),4(offensive),1(slightly),0,,,,,
+Image_(1751).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,逼;妈逼,比;麻痹,text,text
+Image_(410).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,complementary,痛经,痛京,text,image
+Image_(5441).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3243).jpg,5(fear),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(3915).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(4931).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5517).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(226).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),1,text dominant,无言以对,无盐以对,text,complementary
+Image_(2757).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(2811).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2046).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),1,complementary,肥宅快乐水,可乐,text,complementary
+Image_(2288).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,complementary,工作,搬砖,text,image
+Image_(1965).jpg,5(fear),2(moderately),4(offensive),2(moderately),1,text dominant,死亡,上西天,text,text
+Image_(1875).jpg,2(love),1(slightly),2(expressive),0(non-offensive),1,complementary,强思考能力;弱思考能力,Intel i9;pentium,text,image
+Image_(5038).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1830).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,text dominant,妈的智障,MDZZ,text,text
+Image_(5355).jpg,2(love),1(slightly),1(interactive),0(non-offensive),1,text dominant,责任；单身的人,锅；单身狗,text,text
+Image_(4176).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1967).jpg,2(love),1(slightly),3(entertaining),0(non-offensive),1,complementary,三级甲,防晒霜,text,complementary
+Image_(3599).jpg,3(anger),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(347).jpg,2(love),3(very),2(expressive),0(non-offensive),1,text dominant,崇拜,葱白,text,text
+Image_(3733).jpg,4(sorrow),1(slightly),1(interactive),1(slightly),0,,,,,
+Image_(5656).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2906).jpg,5(fear),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1203).jpg,4(sorrow),1(slightly),3(entertaining),0(non-offensive),1,complementary,活在裆下,活在当下,complementary,text
+Image_(3613).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4845).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(265).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,发脾气,野鸡情绪失控综合症,text,text
+Image_(2171).jpg,2(love),1(slightly),1(interactive),0(non-offensive),1,text dominant,你,被指到的人,text,text
+Image_(4844).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2915).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5291).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1848).jpg,7(surprise),1(slightly),4(offensive),1(slightly),1,complementary,出轨,绿帽子,text,image
+Image_(3058).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(4092).jpg,6(hate),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(2954).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3416).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1445).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,complementary,合理;鸡巴,河里;鸡吧,text,complementary
+Image_(4705).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1245).jpg,4(sorrow),1(slightly),3(entertaining),0(non-offensive),1,text dominant,精神上剧变,黑化,text,text
+Image_(1734).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,垃圾,辣鸡,text,text
+Image_(600).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,image dominant,卑,杯,complementary,image
+Image_(3710).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(291).jpg,6(hate),2(moderately),4(offensive),0(non-offensive),1,text dominant,厉害,屌,text,text
+Image_(146).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,text dominant,呀,鸭,text,complementary
+Image_(2281).jpg,2(love),3(very),2(expressive),0(non-offensive),1,text dominant,花,心,text,text
+Image_(4561).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5732).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4797).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(5506).jpg,5(fear),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(3624).jpg,6(hate),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(256).jpg,4(sorrow),2(moderately),4(offensive),2(moderately),1,text dominant,出轨,绿,text,text
+Image_(4599).jpg,3(anger),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(4560).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2092).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),1,complementary,花朵,孩子,text,image
+Image_(5182).jpg,1(happiness),2(moderately),2(expressive),1(slightly),0,,,,,
+Image_(1790).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),1,text dominant,呼,肤,text,text
+Image_(4838).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1762).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),1,text dominant,脚,jio,complementary,text
+Image_(2599).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(6012).jpg,7(surprise),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(5246).jpg,6(hate),1(slightly),4(offensive),2(moderately),0,,,,,
+Image_(5345).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4037).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(2522).jpg,6(hate),2(moderately),1(interactive),1(slightly),0,,,,,
+Image_(583).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),1,image dominant,生活；我,狗；狗,text,image
+Image_(2943).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4761).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(197).jpg,6(hate),3(very),2(expressive),1(slightly),1,complementary,猪,你,image,text
+Image_(87).jpg,3(anger),2(moderately),2(expressive),0(non-offensive),1,complementary,呼,fu,image,text
+Image_(3036).jpg,3(anger),3(very),2(expressive),1(slightly),0,,,,,
+Image_(3653).jpg,5(fear),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5077).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1097).jpg,1(happiness),1(slightly),4(offensive),2(moderately),1,text dominant,傻屌,沙雕,text,text
+Image_(3496).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1150).jpg,3(anger),2(moderately),4(offensive),3(very),1,text dominant,妈,马,text,complementary
+Image_(2724).jpg,3(anger),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(2850).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4458).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5398).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2840).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(4465).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2615).jpg,2(love),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(3601).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3906).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3021).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2060).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,没有梦想的人,咸鱼,text,complementary
+Image_(3805).jpg,2(love),3(very),2(expressive),0(non-offensive),1,complementary,木马,亲亲,text,complementary
+Image_(1275).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,complementary,网抑云,网易云,text,complementary
+Image_(4071).jpg,3(anger),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(5094).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3755).jpg,3(anger),3(very),2(expressive),2(moderately),0,,,,,
+Image_(703).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,回事,肥四,text,text
+Image_(1939).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),1,text dominant,死个,skr,text,complementary
+Image_(5158).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5390).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2742).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(476).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,complementary,草莓,思卓博瑞,image,text
+Image_(4654).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(474).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,complementary,橘子,欧润菊,image,text
+Image_(2869).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1566).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),1,text dominant,你;管,泥;寡,text,text
+Image_(1112).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,text dominant,迷之,蜜汁,text,text
+Image_(3019).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(5034).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(4213).jpg,5(fear),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(2009).jpg,1(happiness),3(very),1(interactive),0(non-offensive),1,image dominant,佛祖,汤普森,complementary,image
+Image_(3893).jpg,3(anger),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4188).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5143).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5201).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(3640).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2844).jpg,1(happiness),1(slightly),4(offensive),2(moderately),0,,,,,
+Image_(131).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),1,text dominant,别,憋,text,text
+Image_(2880).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(1979).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,complementary,人,暖壶,text,image
+Image_(2113).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,text dominant,老干妈,你,text,text
+Image_(1973).jpg,3(anger),2(moderately),2(expressive),0(non-offensive),1,text dominant,嫉妒,吃醋,text,text
+Image_(4484).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1457).jpg,6(hate),1(slightly),4(offensive),2(moderately),1,text dominant,贱,鉴,text,text
+Image_(2026).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,complementary,厉害,大拇指,text,image
+Image_(1291).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),1,text dominant,青春结束了,爷青结,text,text
+Image_(3713).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1572).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),1,text dominant,呀,鸭,text,text
+Image_(4288).jpg,3(anger),2(moderately),3(entertaining),0(non-offensive),0,,,,,
+Image_(3279).jpg,4(sorrow),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4422).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(946).jpg,2(love),3(very),1(interactive),0(non-offensive),1,text dominant,看,康,complementary,text
+Image_(2717).jpg,6(hate),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3681).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4090).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(5787).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(264).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,弄,neng,text,text
+Image_(5984).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1396).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),1,complementary,看八卦,吃瓜,text,complementary
+Image_(5376).jpg,4(sorrow),2(moderately),3(entertaining),0(non-offensive),1,complementary,弃婴,弃音,complementary,text
+Image_(2340).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1411).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,逼逼,bb,text,text
+Image_(4206).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2246).jpg,3(anger),2(moderately),1(interactive),0(non-offensive),1,text dominant,工作或上学,搬砖,text,text
+Image_(4997).jpg,6(hate),2(moderately),1(interactive),1(slightly),0,,,,,
+Image_(1837).jpg,2(love),1(slightly),1(interactive),0(non-offensive),1,text dominant,爽约,鸽子,text,image
+Image_(3673).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2125).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),1,complementary,快乐,奶茶,text,image
+Image_(5859).jpg,4(sorrow),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(1515).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,text dominant,傻屌,沙雕,text,text
+Image_(2295).jpg,2(love),1(slightly),1(interactive),0(non-offensive),1,text dominant,厉害,666,text,text
+Image_(1480).jpg,6(hate),1(slightly),1(interactive),0(non-offensive),1,text dominant,关,瓜,text,text
+Image_(5218).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(1264).jpg,1(happiness),3(very),4(offensive),1(slightly),1,text dominant,无原则讨好他人的人；house,舔狗;好死,text,text
+Image_(5198).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4789).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4525).jpg,2(love),3(very),3(entertaining),1(slightly),0,,,,,
+Image_(5565).jpg,6(hate),2(moderately),4(offensive),1(slightly),0,,,,,
+Image_(4940).jpg,7(surprise),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4101).jpg,7(surprise),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(1923).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,text dominant,厉害,666,text,text
+Image_(1120).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,你厉害你妈呢,你牛你马呢,text,complementary
+Image_(2788).jpg,6(hate),2(moderately),4(offensive),1(slightly),0,,,,,
+Image_(5904).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(2475).jpg,6(hate),2(moderately),2(expressive),1(slightly),0,,,,,
+Image_(5796).jpg,1(happiness),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3558).jpg,3(anger),3(very),2(expressive),1(slightly),0,,,,,
+Image_(621).jpg,6(hate),1(slightly),2(expressive),0(non-offensive),1,text dominant,多损,夺笋,text,text
+Image_(2461).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3114).jpg,5(fear),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(2889).jpg,2(love),3(very),1(interactive),1(slightly),0,,,,,
+Image_(1056).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,老子,劳资,text,text
+Image_(4194).jpg,7(surprise),3(very),3(entertaining),0(non-offensive),0,,,,,
+Image_(1090).jpg,3(anger),2(moderately),4(offensive),2(moderately),1,text dominant,傻屌,沙雕,text,text
+Image_(1589).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,complementary,不行,布星,text,complementary
+Image_(4301).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4891).jpg,1(happiness),2(moderately),3(entertaining),0(non-offensive),0,,,,,
+Image_(5910).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2985).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(3678).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(4252).jpg,3(anger),1(slightly),3(entertaining),0(non-offensive),0,,,,,
+Image_(779).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,text dominant,焦虑,焦绿,text,complementary
+Image_(1759).jpg,1(happiness),1(slightly),4(offensive),1(slightly),1,complementary,猪,你,image,text
+Image_(4162).jpg,3(anger),2(moderately),4(offensive),2(moderately),0,,,,,
+Image_(1829).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,complementary,丑,7+1,complementary,complementary
+Image_(5773).jpg,2(love),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(4631).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5795).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(2916).jpg,5(fear),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2780).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5814).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(770).jpg,6(hate),1(slightly),4(offensive),1(slightly),1,text dominant,多损,夺笋,text,complementary
+Image_(4744).jpg,1(happiness),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(5639).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5187).jpg,2(love),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(4255).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4840).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),0,,,,,
+Image_(4025).jpg,2(love),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(852).jpg,6(hate),1(slightly),4(offensive),2(moderately),1,text dominant,神经病,蛇精病,text,text
+Image_(3936).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3831).jpg,2(love),3(very),1(interactive),0(non-offensive),0,,,,,
+Image_(3055).jpg,3(anger),1(slightly),4(offensive),2(moderately),0,,,,,
+Image_(4790).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(4436).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2123).jpg,2(love),2(moderately),1(interactive),0(non-offensive),1,complementary,减肥;奶茶;我,直线方向;右转;车,text,image
+Image_(4173).jpg,3(anger),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2066).jpg,1(happiness),1(slightly),3(entertaining),0(non-offensive),1,complementary,800000公斤,肚子,text,image
+Image_(3574).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2238).jpg,7(surprise),1(slightly),3(entertaining),1(slightly),1,complementary,尿,？,image,text
+Image_(2108).jpg,1(happiness),2(moderately),4(offensive),1(slightly),1,complementary,猪,你,image,text
+Image_(5692).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(3836).jpg,2(love),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(3345).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(2790).jpg,6(hate),3(very),4(offensive),3(very),0,,,,,
+Image_(4882).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(991).jpg,3(anger),2(moderately),4(offensive),3(very),1,text dominant,垃圾,辣鸡,text,text
+Image_(117).jpg,3(anger),3(very),4(offensive),3(very),1,text dominant,妈逼,麻痹,text,text
+Image_(1559).jpg,4(sorrow),2(moderately),4(offensive),2(moderately),1,text dominant,傻屌,沙雕,text,text
+Image_(2759).jpg,1(happiness),1(slightly),1(interactive),2(moderately),0,,,,,
+Image_(67).jpg,4(sorrow),1(slightly),2(expressive),0(non-offensive),1,text dominant,情人节,44422,text,text
+Image_(142).jpg,3(anger),1(slightly),4(offensive),1(slightly),1,text dominant,傻狗,你,text,text
+Image_(4698).jpg,1(happiness),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(459).jpg,2(love),1(slightly),3(entertaining),0(non-offensive),1,complementary,好运连连,好运莲莲,text,image
+Image_(4119).jpg,1(happiness),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(5129).jpg,4(sorrow),2(moderately),2(expressive),0(non-offensive),0,,,,,
+Image_(2236).jpg,4(sorrow),3(very),2(expressive),0(non-offensive),1,text dominant,嫉妒,吃醋,text,text
+Image_(5042).jpg,1(happiness),2(moderately),1(interactive),0(non-offensive),0,,,,,
+Image_(5382).jpg,1(happiness),1(slightly),4(offensive),1(slightly),0,,,,,
+Image_(5247).jpg,6(hate),2(moderately),4(offensive),2(moderately),1,text dominant,特么,他妈,text,text
+Image_(1427).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,complementary,我;手机,兔子;兔子,text,image
+Image_(2196).jpg,6(hate),1(slightly),4(offensive),2(moderately),1,text dominant,狗,男人,text,text
+Image_(1743).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),1,text dominant,特么,TM,text,text
+Image_(2149).jpg,2(love),2(moderately),2(expressive),0(non-offensive),1,text dominant,力量,知识,text,text
+Image_(5718).jpg,3(anger),1(slightly),1(interactive),0(non-offensive),0,,,,,
+Image_(5096).jpg,2(love),1(slightly),2(expressive),0(non-offensive),0,,,,,
+Image_(4913).jpg,3(anger),3(very),1(interactive),0(non-offensive),0,,,,,

data/label_C_evaluate_relabel.csv

@@ -0,0 +1,605 @@
+Image_(3482).jpg,reader,reader
+Image_(4321).jpg,character,reader
+Image_(2473).jpg,character,reader
+Image_(1208).jpg,character,character
+Image_(5462).jpg,character,reader
+Image_(500).jpg,character,reader
+Image_(1093).jpg,character,character
+Image_(1227).jpg,reader,reader
+Image_(181).jpg,character,character
+Image_(2884).jpg,character,reader
+Image_(4297).jpg,character,author
+Image_(3759).jpg,character,reader
+Image_(1799).jpg,character,reader
+Image_(3299).jpg,author,author/reader
+Image_(2053).jpg,character,reader
+Image_(4269).jpg,character,reader
+Image_(5162).jpg,character,reader
+Image_(5860).jpg,character,reader
+Image_(5786).jpg,character,reader
+Image_(859).jpg,reader,author
+Image_(5593).jpg,character,character/author
+Image_(1452).jpg,reader,author
+Image_(5719).jpg,character,reader
+Image_(3127).jpg,character,reader
+Image_(2247).jpg,character,reader
+Image_(857).jpg,character,reader
+Image_(5625).jpg,reader,reader
+Image_(512).jpg,reader,reader
+Image_(5670).jpg,character,author
+Image_(3572).jpg,character,reader
+Image_(1809).jpg,character,reader
+Image_(1211).jpg,character,reader
+Image_(3531).jpg,character,reader
+Image_(2495).jpg,,reader
+Image_(540).jpg,character,author/reader
+Image_(751).jpg,reader,reader
+Image_(674).jpg,reader,author
+Image_(2744).jpg,reader,reader
+Image_(2313).jpg,character,reader
+Image_(606).jpg,character,reader
+Image_(5170).jpg,character,reader
+Image_(1157).jpg,reader,reader
+Image_(2398).jpg,reader,reader
+Image_(4776).jpg,character,character/reader
+Image_(2738).jpg,reader,reader
+Image_(3240).jpg,character,author
+Image_(5830).jpg,reader,author
+Image_(808).jpg,author,author
+Image_(1375).jpg,character,reader
+Image_(5159).jpg,reader,reader
+Image_(1471).jpg,reader,reader
+Image_(5605).jpg,reader,reader
+Image_(240).jpg,reader,reader/author
+Image_(2975).jpg,character,character
+Image_(3427).jpg,character,reader
+Image_(3392).jpg,character,reader
+Image_(666).jpg,reader,reader
+Image_(4331).jpg,character,reader
+Image_(4627).jpg,reader,reader
+Image_(5645).jpg,character,character
+Image_(1393).jpg,reader,character/author
+Image_(3214).jpg,character,reader
+Image_(2642).jpg,character,reader
+Image_(2700).jpg,character,reader
+Image_(5976).jpg,reader,reader
+Image_(2299).jpg,reader,author
+Image_(1715).jpg,character,reader
+Image_(5494).jpg,character,reader
+Image_(719).jpg,character,reader
+Image_(180).jpg,character,reader
+Image_(4017).jpg,character,reader
+Image_(166).jpg,reader,reader
+Image_(2582).jpg,character,author
+Image_(1170).jpg,character,reader
+Image_(5877).jpg,character,reader
+Image_(4815).jpg,reader,reader
+Image_(5393).jpg,character,reader
+Image_(2756).jpg,character,author
+Image_(2654).jpg,reader,character
+Image_(365).jpg,character,reader
+Image_(1618).jpg,character,author
+Image_(232).jpg,reader,reader
+Image_(827).jpg,character,reader
+Image_(2783).jpg,character,reader
+Image_(5154).jpg,character,reader
+Image_(1015).jpg,character,reader
+Image_(2801).jpg,character,author
+Image_(210).jpg,reader,reader
+Image_(3927).jpg,reader,reader
+Image_(3940).jpg,reader,reader
+Image_(1438).jpg,reader,reader
+Image_(1541).jpg,reader,author
+Image_(1516).jpg,character,reader
+Image_(5349).jpg,character,reader
+Image_(4238).jpg,reader,author
+Image_(2412).jpg,character,reader
+Image_(312).jpg,reader,author
+Image_(1016).jpg,reader,reader
+Image_(1955).jpg,character,reader
+Image_(1012).jpg,character,reader
+Image_(137).jpg,character,reader
+Image_(4050).jpg,character,reader
+Image_(5458).jpg,character,author/reader
+Image_(1140).jpg,character,reader
+Image_(6029).jpg,character,character
+Image_(3119).jpg,reader,reader
+Image_(5251).jpg,character,reader
+Image_(1243).jpg,character,author/reader
+Image_(2049).jpg,reader,reader
+Image_(5084).jpg,character,reader
+Image_(2355).jpg,reader,reader
+Image_(1861).jpg,characters,reader
+Image_(3810).jpg,character,reader
+Image_(5274).jpg,reader,reader
+Image_(3473).jpg,character,reader
+Image_(248).jpg,character,reader
+Image_(4504).jpg,character,character
+Image_(5406).jpg,character,reader
+Image_(3081).jpg,character,author
+Image_(505).jpg,reader,author
+Image_(5207).jpg,reader,reader
+Image_(4300).jpg,character,reader
+Image_(2578).jpg,reader,reader
+Image_(5769).jpg,character,reader
+Image_(5103).jpg,character,character
+Image_(813).jpg,character,author
+Image_(756).jpg,reader,reader
+Image_(84).jpg,reader,reader
+Image_(7).jpg,character,reader
+Image_(2498).jpg,reader,reader
+Image_(3378).jpg,reader,reader
+Image_(5841).jpg,reader,reader
+Image_(1670).jpg,character,reader
+Image_(626).jpg,character,character
+Image_(1688).jpg,reader,reader
+Image_(3274).jpg,character,reader
+Image_(1313).jpg,reader,reader
+Image_(1059).jpg,reader,reader/author
+Image_(1800).jpg,reader,author
+Image_(4205).jpg,character,character
+Image_(4618).jpg,character,reader
+Image_(5911).jpg,character,character/reader
+Image_(4943).jpg,character,reader
+Image_(3327).jpg,character/reader,author/reader
+Image_(4852).jpg,reader,reader
+Image_(5716).jpg,character,author
+Image_(1609).jpg,character,reader
+Image_(3040).jpg,character,reader
+Image_(5595).jpg,character,character
+Image_(5467).jpg,character,reader
+Image_(2560).jpg,character,reader
+Image_(1333).jpg,character,reader
+Image_(2401).jpg,reader,reader
+Image_(6008).jpg,character,character
+Image_(1912).jpg,character,character
+Image_(4402).jpg,reader,reader
+Image_(806).jpg,character,reader
+Image_(5420).jpg,reader,reader
+Image_(5384).jpg,character,author
+Image_(1226).jpg,reader,reader
+Image_(1825).jpg,reader,reader
+Image_(238).jpg,character,reader
+Image_(1935).jpg,reader,reader
+Image_(5997).jpg,character,reader
+Image_(5532).jpg,character,reader
+Image_(1717).jpg,reader,reader
+Image_(2901).jpg,character,reader
+Image_(2303).jpg,character,reader
+Image_(805).jpg,character,reader
+Image_(757).jpg,character/reader,reader
+Image_(5944).jpg,character,author
+Image_(3018).jpg,character,reader
+Image_(2023).jpg,character,reader
+Image_(5695).jpg,reader,reader
+Image_(3244).jpg,reader,reader
+Image_(2739).jpg,character,reader
+Image_(6036).jpg,reader,reader
+Image_(2692).jpg,reader,author
+Image_(3163).jpg,character,character
+Image_(807).jpg,character,reader
+Image_(3025).jpg,character,character
+Image_(1930).jpg,character,reader
+Image_(682).jpg,character,reader
+Image_(3038).jpg,reader,reader
+Image_(1731).jpg,character,character
+Image_(2545).jpg,character,reader
+Image_(1259).jpg,character,author
+Image_(2779).jpg,reader,reader
+Image_(1209).jpg,character,reader
+Image_(2604).jpg,character,character/reader
+Image_(2275).jpg,character,reader
+Image_(2620).jpg,reader,reader
+Image_(2152).jpg,character,reader
+Image_(1664).jpg,character,character
+Image_(1915).jpg,character,reader
+Image_(2181).jpg,reader,reader
+Image_(4102).jpg,character,character
+Image_(2175).jpg,reader,author
+Image_(4022).jpg,character,reader
+Image_(817).jpg,character,author
+Image_(1347).jpg,reader,reader
+Image_(100).jpg,reader,author
+Image_(4229).jpg,reader,reader
+Image_(4394).jpg,author,author
+Image_(2321).jpg,character,reader
+Image_(4816).jpg,character,reader
+Image_(341).jpg,reader,reader
+Image_(4708).jpg,reader,author
+Image_(3494).jpg,reader,reader
+Image_(3540).jpg,character,reader
+Image_(619).jpg,character,character/reader
+Image_(5449).jpg,reader,reader
+Image_(712).jpg,reader,reader
+Image_(1449).jpg,character,reader
+Image_(5741).jpg,character,character
+Image_(5192).jpg,character,character
+Image_(4806).jpg,character,reader
+Image_(3206).jpg,character,character/reader
+Image_(4686).jpg,character,character
+Image_(2854).jpg,reader,reader
+Image_(3283).jpg,character,reader
+Image_(5133).jpg,character,author
+Image_(2721).jpg,character,character/reader
+Image_(5881).jpg,reader,author
+Image_(2774).jpg,character,character
+Image_(1106).jpg,reader,reader
+Image_(3623).jpg,character,author
+Image_(5580).jpg,character,reader
+Image_(3295).jpg,character/reader,author
+Image_(2793).jpg,character/reader,character/reader
+Image_(1775).jpg,character,reader
+Image_(1555).jpg,character,reader
+Image_(4812).jpg,reader,reader
+Image_(4648).jpg,character/reader,character/reader
+Image_(4541).jpg,character,character
+Image_(3344).jpg,character,character
+Image_(2988).jpg,character,author
+Image_(1804).jpg,character,reader
+Image_(2635).jpg,character,character
+Image_(729).jpg,character,author
+Image_(2810).jpg,reader,reader
+Image_(2891).jpg,reader,reader
+Image_(3717).jpg,character,reader
+Image_(5669).jpg,character,character
+Image_(3433).jpg,character,reader
+Image_(1013).jpg,reader,reader
+Image_(3342).jpg,reader,reader/author
+Image_(4803).jpg,reader,reader
+Image_(4428).jpg,character/reader,reader
+Image_(3480).jpg,characters,character
+Image_(2996).jpg,reader,reader
+Image_(4087).jpg,character,reader
+Image_(5827).jpg,character,reader
+Image_(4108).jpg,character,reader
+Image_(2348).jpg,reader,reader
+Image_(3014).jpg,author,author/reader
+Image_(3266).jpg,reader,author
+Image_(5128).jpg,character,reader
+Image_(646).jpg,character,reader
+Image_(4678).jpg,reader,reader
+Image_(1730).jpg,reader,reader
+Image_(1406).jpg,reader,reader
+Image_(4880).jpg,character,author
+Image_(695).jpg,character,character
+Image_(2541).jpg,reader,reader
+Image_(1643).jpg,characters,reader
+Image_(2081).jpg,reader,reader
+Image_(5918).jpg,character,author
+Image_(968).jpg,character,reader
+Image_(5782).jpg,character/reader,reader
+Image_(1263).jpg,reader,reader
+Image_(50).jpg,character,reader
+Image_(495).jpg,reader,reader
+Image_(5820).jpg,reader,character/author
+Image_(5705).jpg,character,author
+Image_(615).jpg,character,character
+Image_(644).jpg,reader,reader
+Image_(4420).jpg,character,author
+Image_(4912).jpg,reader,reader
+Image_(1615).jpg,reader,reader
+Image_(4749).jpg,character,reader
+Image_(126).jpg,character,reader
+Image_(1868).jpg,reader,reader
+Image_(1535).jpg,character,reader
+Image_(3109).jpg,reader,reader
+Image_(4869).jpg,reader,author
+Image_(3919).jpg,character,reader
+Image_(3931).jpg,character,character
+Image_(952).jpg,reader,reader
+Image_(3567).jpg,reader,reader
+Image_(1250).jpg,character,reader
+Image_(2895).jpg,character,reader
+Image_(119).jpg,reader,reader
+Image_(522).jpg,character,reader
+Image_(452).jpg,reader,reader
+Image_(1251).jpg,character,reader
+Image_(804).jpg,character,reader
+Image_(889).jpg,character,reader
+Image_(1468).jpg,character,author
+Image_(1952).jpg,reader,author/reader
+Image_(2815).jpg,reader,reader
+Image_(598).jpg,character,reader
+Image_(4994).jpg,character,reader
+Image_(3126).jpg,reader,author
+Image_(2170).jpg,reader,reader
+Image_(5444).jpg,character,character
+Image_(4848).jpg,character,reader
+Image_(2103).jpg,character,reader
+Image_(891).jpg,reader,reader
+Image_(1876).jpg,reader,reader
+Image_(3894).jpg,character,author
+Image_(2390).jpg,character,reader
+Image_(5676).jpg,reader,reader
+Image_(5361).jpg,character,reader
+Image_(3459).jpg,reader,reader
+Image_(1701).jpg,character,reader/author
+Image_(5249).jpg,character,character
+Image_(278).jpg,reader,reader
+Image_(3361).jpg,character,reader
+Image_(4015).jpg,character,reader
+Image_(2337).jpg,character,character
+Image_(366).jpg,character,reader
+Image_(5551).jpg,character,reader
+Image_(3060).jpg,character,reader
+Image_(5570).jpg,reader,reader
+Image_(4324).jpg,character,reader
+Image_(4178).jpg,character,character
+Image_(5009).jpg,reader,reader
+Image_(1169).jpg,character,reader
+Image_(2579).jpg,reader,reader
+Image_(5120).jpg,character,author
+Image_(1761).jpg,character,reader
+Image_(3516).jpg,author,reader
+Image_(4730).jpg,character,author
+Image_(2553).jpg,character,author
+Image_(989).jpg,character,character
+Image_(2969).jpg,reader,reader
+Image_(3552).jpg,character,reader
+Image_(1595).jpg,reader,reader
+Image_(2871).jpg,character,reader
+Image_(1420).jpg,author,author/reader
+Image_(6038).jpg,character,reader
+Image_(3205).jpg,reader,reader
+Image_(4680).jpg,character,reader
+Image_(3158).jpg,character,reader
+Image_(4426).jpg,reader,reader
+Image_(4455).jpg,reader,reader
+Image_(5794).jpg,character,character
+Image_(2045).jpg,character,reader
+Image_(4281).jpg,character,character
+Image_(790).jpg,character,reader
+Image_(3953).jpg,reader,reader
+Image_(3349).jpg,character,reader
+Image_(3228).jpg,character,reader
+Image_(2146).jpg,character,reader
+Image_(3772).jpg,reader,reader
+Image_(177).jpg,reader,reader
+Image_(5317).jpg,reader,reader
+Image_(208).jpg,character,reader
+Image_(4781).jpg,character,reader
+Image_(2308).jpg,character,reader
+Image_(1926).jpg,character,author/reader
+Image_(3695).jpg,character,reader
+Image_(5511).jpg,character,reader
+Image_(3264).jpg,character,reader
+Image_(3648).jpg,character,reader
+Image_(2870).jpg,reader,author
+Image_(3797).jpg,character,reader
+Image_(3369).jpg,character,character
+Image_(2128).jpg,reader,reader
+Image_(5704).jpg,character,reader
+Image_(518).jpg,character,reader
+Image_(5755).jpg,reader,reader
+Image_(1751).jpg,character,reader
+Image_(410).jpg,character,reader
+Image_(5441).jpg,character,character/author
+Image_(3243).jpg,character,reader
+Image_(3915).jpg,character,reader
+Image_(4931).jpg,character,author
+Image_(5517).jpg,character,reader
+Image_(226).jpg,character,reader
+Image_(2757).jpg,character,reader
+Image_(2811).jpg,character,reader
+Image_(2046).jpg,character,reader
+Image_(2288).jpg,author,author
+Image_(1965).jpg,character,reader
+Image_(1875).jpg,reader,reader
+Image_(5038).jpg,character,reader
+Image_(1830).jpg,reader,reader
+Image_(5355).jpg,reader,author
+Image_(4176).jpg,character,character
+Image_(1967).jpg,reader,reader
+Image_(3599).jpg,character,character
+Image_(347).jpg,reader,reader
+Image_(3733).jpg,character,character
+Image_(5656).jpg,character/reader,reader
+Image_(2906).jpg,reader,reader
+Image_(1203).jpg,character,reader
+Image_(3613).jpg,character,reader
+Image_(4845).jpg,reader,reader
+Image_(265).jpg,character,reader
+Image_(2171).jpg,reader,reader
+Image_(4844).jpg,character,reader
+Image_(2915).jpg,reader,reader
+Image_(5291).jpg,character,character
+Image_(1848).jpg,characters,reader
+Image_(3058).jpg,character,character
+Image_(4092).jpg,character,reader
+Image_(2954).jpg,character,author
+Image_(3416).jpg,character,character/reader
+Image_(1445).jpg,reader,reader
+Image_(4705).jpg,character,author
+Image_(1245).jpg,reader,reader
+Image_(1734).jpg,characters,reader
+Image_(600).jpg,character,reader
+Image_(3710).jpg,character,character/reader
+Image_(291).jpg,character,reader
+Image_(146).jpg,reader,reader
+Image_(2281).jpg,character,reader
+Image_(4561).jpg,character,reader
+Image_(5732).jpg,character,author/reader
+Image_(4797).jpg,character,reader
+Image_(5506).jpg,character,character
+Image_(3624).jpg,character,reader
+Image_(256).jpg,author,reader
+Image_(4599).jpg,character,reader
+Image_(4560).jpg,reader,reader
+Image_(2092).jpg,character,reader
+Image_(5182).jpg,character,reader
+Image_(1790).jpg,character,reader
+Image_(4838).jpg,character,reader
+Image_(1762).jpg,characters,author
+Image_(2599).jpg,reader,reader
+Image_(6012).jpg,character,reader
+Image_(5246).jpg,character,reader
+Image_(5345).jpg,reader,reader
+Image_(4037).jpg,character,character
+Image_(2522).jpg,character,reader
+Image_(583).jpg,character,reader
+Image_(2943).jpg,character,author
+Image_(4761).jpg,character,reader
+Image_(197).jpg,reader,reader
+Image_(87).jpg,reader,reader
+Image_(3036).jpg,character,character
+Image_(3653).jpg,character,character
+Image_(5077).jpg,character,reader
+Image_(1097).jpg,reader,reader
+Image_(3496).jpg,character,reader
+Image_(1150).jpg,character,character
+Image_(2724).jpg,reader,reader
+Image_(2850).jpg,reader,author
+Image_(4458).jpg,character,reader
+Image_(5398).jpg,reader,reader
+Image_(2840).jpg,reader,author
+Image_(4465).jpg,character,character
+Image_(2615).jpg,character,character/reader
+Image_(3601).jpg,reader,reader
+Image_(3906).jpg,reader,reader
+Image_(3021).jpg,character,reader
+Image_(2060).jpg,reader,reader
+Image_(3805).jpg,character,character/reader
+Image_(1275).jpg,reader,reader
+Image_(4071).jpg,reader,reader
+Image_(5094).jpg,character,author
+Image_(3755).jpg,character,author
+Image_(703).jpg,character,reader
+Image_(1939).jpg,character,reader
+Image_(5158).jpg,reader,reader
+Image_(5390).jpg,character,reader
+Image_(2742).jpg,character,character
+Image_(476).jpg,reader,reader
+Image_(4654).jpg,character,character
+Image_(474).jpg,character,reader
+Image_(2869).jpg,reader,reader
+Image_(1566).jpg,reader,reader
+Image_(1112).jpg,character,reader
+Image_(3019).jpg,reader,reader
+Image_(5034).jpg,character,reader
+Image_(4213).jpg,reader,reader
+Image_(2009).jpg,reader,reader
+Image_(3893).jpg,character,author
+Image_(4188).jpg,character,reader
+Image_(5143).jpg,character,reader
+Image_(5201).jpg,reader,author
+Image_(3640).jpg,character,reader
+Image_(2844).jpg,character,reader
+Image_(131).jpg,character,reader
+Image_(2880).jpg,character,reader
+Image_(1979).jpg,reader,reader
+Image_(2113).jpg,reader,author
+Image_(1973).jpg,character,reader
+Image_(4484).jpg,character,reader
+Image_(1457).jpg,character,character/reader
+Image_(2026).jpg,character,character
+Image_(1291).jpg,character,character
+Image_(3713).jpg,reader,reader
+Image_(1572).jpg,reader,reader
+Image_(4288).jpg,character,reader
+Image_(3279).jpg,character,reader
+Image_(4422).jpg,character,character/reader
+Image_(946).jpg,reader,reader
+Image_(2717).jpg,character,character
+Image_(3681).jpg,character,character
+Image_(4090).jpg,character,reader
+Image_(5787).jpg,reader,reader
+Image_(264).jpg,character,reader
+Image_(5984).jpg,reader,reader
+Image_(1396).jpg,character,reader
+Image_(5376).jpg,reader,author
+Image_(2340).jpg,reader,author
+Image_(1411).jpg,character,character/author
+Image_(4206).jpg,character,reader
+Image_(2246).jpg,reader,reader
+Image_(4997).jpg,reader,reader
+Image_(1837).jpg,reader,author
+Image_(3673).jpg,character,character/reader
+Image_(2125).jpg,reader,reader
+Image_(5859).jpg,reader,reader
+Image_(1515).jpg,character,author
+Image_(2295).jpg,reader,reader
+Image_(1480).jpg,reader,reader
+Image_(5218).jpg,reader,reader
+Image_(1264).jpg,character,reader
+Image_(5198).jpg,character,character
+Image_(4789).jpg,character,reader
+Image_(4525).jpg,character,reader
+Image_(5565).jpg,character,reader
+Image_(4940).jpg,reader,reader
+Image_(4101).jpg,reader,reader
+Image_(1923).jpg,character,character
+Image_(1120).jpg,reader,reader
+Image_(2788).jpg,character,reader
+Image_(5904).jpg,character,author
+Image_(2475).jpg,reader,author
+Image_(5796).jpg,character,author
+Image_(3558).jpg,character,reader
+Image_(621).jpg,character,author
+Image_(2461).jpg,character,reader
+Image_(3114).jpg,character,character
+Image_(2889).jpg,character,character
+Image_(1056).jpg,character,reader
+Image_(4194).jpg,character,author
+Image_(1090).jpg,reader,reader
+Image_(1589).jpg,reader,reader
+Image_(4301).jpg,character,reader
+Image_(4891).jpg,character,character/reader
+Image_(5910).jpg,character,reader
+Image_(2985).jpg,reader,reader
+Image_(3678).jpg,reader,reader
+Image_(4252).jpg,reader,reader
+Image_(779).jpg,character,reader
+Image_(1759).jpg,character,author
+Image_(4162).jpg,reader,reader
+Image_(1829).jpg,reader,reader
+Image_(5773).jpg,reader,reader
+Image_(4631).jpg,character,character
+Image_(5795).jpg,character,reader
+Image_(2916).jpg,character,reader
+Image_(2780).jpg,reader,author
+Image_(5814).jpg,character,character/reader
+Image_(770).jpg,reader,author
+Image_(4744).jpg,character,character
+Image_(5639).jpg,character/reader,author
+Image_(5187).jpg,reader,reader
+Image_(4255).jpg,reader,reader
+Image_(4840).jpg,character,reader
+Image_(4025).jpg,character,reader
+Image_(852).jpg,reader,reader
+Image_(3936).jpg,reader,reader
+Image_(3831).jpg,reader,reader
+Image_(3055).jpg,reader,reader
+Image_(4790).jpg,character,reader
+Image_(4436).jpg,character,reader
+Image_(2123).jpg,reader,reader
+Image_(4173).jpg,author,author
+Image_(2066).jpg,reader,author
+Image_(3574).jpg,character,reader
+Image_(2238).jpg,reader,author
+Image_(2108).jpg,character,reader
+Image_(5692).jpg,character,author/reader
+Image_(3836).jpg,character,author
+Image_(3345).jpg,character,character
+Image_(2790).jpg,reader,reader
+Image_(4882).jpg,character,reader
+Image_(991).jpg,reader,author
+Image_(117).jpg,character,character/reader
+Image_(1559).jpg,character,reader
+Image_(2759).jpg,reader,reader
+Image_(67).jpg,character,author/reader
+Image_(142).jpg,character,reader
+Image_(4698).jpg,character,author
+Image_(459).jpg,characters,characters
+Image_(4119).jpg,character,reader
+Image_(5129).jpg,reader,reader
+Image_(2236).jpg,character,reader
+Image_(5042).jpg,reader,reader
+Image_(5382).jpg,character,character
+Image_(5247).jpg,reader,reader
+Image_(1427).jpg,character/reader,character
+Image_(2196).jpg,reader,reader
+Image_(1743).jpg,character/reader,character/author
+Image_(2149).jpg,character,character
+Image_(5718).jpg,character,reader
+Image_(5096).jpg,character,reader
+Image_(4913).jpg,character,author/reader

generate/data_preprocess.py

@@ -0,0 +1,130 @@
+import csv
+import json
+import random
+
+# 文件路径
+csv_file_path = '/mnt/afs/xueyingyi/meme/generate/E_text_1.csv'  # CSV文件路径
+user_input_jsonl_path = '/mnt/afs/xueyingyi/meme/generate/user_input_all.jsonl'  # user_input.jsonl文件路径
+output_jsonl_path = '/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_multi_all_item.jsonl'  # 输出JSONL文件路径
+train_jsonl_path = '/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_train_multi_all_item.jsonl'  # 训练集路径
+eval_jsonl_path = '/mnt/afs/xueyingyi/meme/data/Cjson/C_generate_eval_multi_all_item.jsonl'  # 测试集路径
+train_config_path = '/mnt/afs/xueyingyi/meme/data/C_generate_train_multi_all_item.jsonl'  # 训练集配置文件路径
+eval_config_path = '/mnt/afs/xueyingyi/meme/data/C_generate_eval_multi_all_item.jsonl'  # 测试集配置文件路径
+
+# 读取CSV文件
+csv_data = {}
+with open(csv_file_path, 'r', encoding='utf-8') as csv_file:
+    csv_reader = csv.DictReader(csv_file)
+    for row in csv_reader:
+        file_name = row['file_name']
+        text = row['text'].strip()
+        csv_data[file_name] = text  # 存储file_name和text的映射关系
+
+# 读取user_input.jsonl文件
+user_input_data = []
+with open(user_input_jsonl_path, 'r', encoding='utf-8') as f:
+    for line in f:
+        user_input_data.append(json.loads(line.strip()))
+
+# 构建JSONL数据
+jsonl_data = []
+for idx, item in enumerate(user_input_data):
+    file_name = item['file_name']
+    user_input = item['user_input']
+    
+    # 检查file_name是否在CSV数据中
+    if file_name not in csv_data:
+        print(f"警告: {file_name} 在CSV文件中未找到，跳过此条数据")
+        continue
+    
+    # 获取对应的text
+    text = csv_data[file_name]
+    
+    # 构建提示词
+    with open('/mnt/afs/xueyingyi/vl2.5/InternVL/inference/text_new.txt', 'r') as prompt_file:
+        PROMPT = prompt_file.read()
+
+    with open('/mnt/afs/xueyingyi/vl2.5/InternVL/inference/text_example.txt', 'r') as prompt_file:
+        PROMPT_example = prompt_file.read()
+    
+    # 构建conversations
+    conversations = [
+        {
+            "from": "human",
+            "value": f"{PROMPT}<image>\n{PROMPT_example}\n<image>\n{user_input}"
+        },
+        {
+            "from": "gpt",
+            "value": text  # 使用CSV中的文字
+        }
+    ]
+    
+    # 构建JSON对象(单图像)
+    # json_obj = {
+    #     "id": idx,
+    #     "image": f"/mnt/afs/xueyingyi/image_vague/inpainting_demo/{file_name}",
+    #     "conversations": conversations
+    # }
+    # 构建JSON对象(多图像)
+    json_obj = {
+        "id": idx,
+        "image": [
+        f"/mnt/afs/xueyingyi/vl2.5/InternVL/inference/example.jpg",
+        f"/mnt/afs/xueyingyi/image_vague/inpainting_demo/{file_name}"
+        ], 
+        "conversations": conversations
+    }
+    
+    jsonl_data.append(json_obj)
+
+# 保存为JSONL文件
+with open(output_jsonl_path, 'w', encoding='utf-8') as f:
+    for item in jsonl_data:
+        f.write(json.dumps(item, ensure_ascii=False) + '\n')
+
+# 划分训练集和测试集
+random.seed(42)  # 设置随机种子以确保可重复性
+random.shuffle(jsonl_data)  # 打乱数据
+train_size = int(len(jsonl_data) * 0.9)  # 90%训练集
+train_data = jsonl_data[:train_size]
+eval_data = jsonl_data[train_size:]
+
+# 保存训练集
+with open(train_jsonl_path, 'w', encoding='utf-8') as f:
+    for item in train_data:
+        f.write(json.dumps(item, ensure_ascii=False) + '\n')
+
+# 保存测试集
+with open(eval_jsonl_path, 'w', encoding='utf-8') as f:
+    for item in eval_data:
+        f.write(json.dumps(item, ensure_ascii=False) + '\n')
+
+# 生成训练集配置文件
+train_config = {
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": train_jsonl_path,
+        "data_augment": False,
+        "repeat_time": 1,
+        "length": len(train_data)
+    }
+}
+with open(train_config_path, 'w', encoding='utf-8') as f:
+    json.dump(train_config, f, ensure_ascii=False, indent=4)
+
+# 生成测试集配置文件
+eval_config = {
+    "classification_C": {
+        "root": "/mnt/afs/xueyingyi/image_vague/inpainting_demo",
+        "annotation": eval_jsonl_path,
+        "data_augment": False,
+        "repeat_time": 1,
+        "length": len(eval_data)
+    }
+}
+with open(eval_config_path, 'w', encoding='utf-8') as f:
+    json.dump(eval_config, f, ensure_ascii=False, indent=4)
+
+print("数据处理完成！")
+print(f"训练集大小: {len(train_data)}")
+print(f"测试集大小: {len(eval_data)}")