Add files using upload-large-folder tool

.gitattributes

@@ -1576,3 +1576,4 @@ evalkit_tf446/lib/python3.10/site-packages/nvidia/cufft/lib/libcufft.so.10 filte
 evalkit_tf449/lib/python3.10/site-packages/nvidia/cublas/lib/libcublasLt.so.12 filter=lfs diff=lfs merge=lfs -text
 infer_4_47_1/lib/python3.10/site-packages/gradio/_frontend_code/lite/dist/assets/gradio_client-1.5.3-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
 evalkit_cambrian/lib/python3.10/site-packages/transformers/models/perceiver/__pycache__/modeling_perceiver.cpython-310.pyc filter=lfs diff=lfs merge=lfs -text
+infer_4_47_1/lib/python3.10/site-packages/fontTools/pens/momentsPen.cpython-310-x86_64-linux-gnu.so filter=lfs diff=lfs merge=lfs -text

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/chinese_clip/convert_chinese_clip_original_pytorch_to_hf.py

@@ -0,0 +1,134 @@
+# coding=utf-8
+# Copyright 2022 The OFA-Sys Team Authors and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+
+import torch
+
+from transformers import ChineseCLIPConfig, ChineseCLIPModel
+
+
+def copy_attn_layer(hf_attn_layer, pt_weights, prefix):
+    q_proj, k_proj, v_proj = pt_weights[f"{prefix}.in_proj_weight"].chunk(3, dim=0)
+    q_proj_bias, k_proj_bias, v_proj_bias = pt_weights[f"{prefix}.in_proj_bias"].chunk(3, dim=0)
+
+    out_proj_weights = pt_weights[f"{prefix}.out_proj.weight"]
+    out_proj_bias = pt_weights[f"{prefix}.out_proj.bias"]
+
+    hf_attn_layer.q_proj.weight.data = q_proj
+    hf_attn_layer.q_proj.bias.data = q_proj_bias
+
+    hf_attn_layer.k_proj.weight.data = k_proj
+    hf_attn_layer.k_proj.bias.data = k_proj_bias
+
+    hf_attn_layer.v_proj.weight.data = v_proj
+    hf_attn_layer.v_proj.bias.data = v_proj_bias
+
+    hf_attn_layer.out_proj.weight.data = out_proj_weights
+    hf_attn_layer.out_proj.bias.data = out_proj_bias
+
+
+def copy_mlp(hf_mlp, pt_weights, prefix):
+    copy_linear(hf_mlp.fc1, pt_weights, f"{prefix}.c_fc")
+    copy_linear(hf_mlp.fc2, pt_weights, f"{prefix}.c_proj")
+
+
+def copy_linear(hf_linear, pt_weights, prefix):
+    hf_linear.weight.data = pt_weights[f"{prefix}.weight"].data
+    hf_linear.bias.data = pt_weights[f"{prefix}.bias"].data
+
+
+def copy_layer(hf_layer, pt_weights, prefix):
+    # copy layer norms
+    copy_linear(hf_layer.layer_norm1, pt_weights, f"{prefix}.ln_1")
+    copy_linear(hf_layer.layer_norm2, pt_weights, f"{prefix}.ln_2")
+
+    # copy MLP
+    copy_mlp(hf_layer.mlp, pt_weights, f"{prefix}.mlp")
+
+    # copy attn
+    copy_attn_layer(hf_layer.self_attn, pt_weights, f"{prefix}.attn")
+
+
+def copy_layers(hf_layers, pt_weights, prefix):
+    for layer_id, hf_layer in enumerate(hf_layers):
+        copy_layer(hf_layer, pt_weights, f"{prefix}.{layer_id}")
+
+
+def copy_text_model_and_projection(hf_model, pt_weights):
+    # copy projection
+    hf_model.text_projection.weight.data = pt_weights["text_projection"].data.T
+
+    # copy text encoder
+    for name, param in hf_model.text_model.named_parameters():
+        param.data = pt_weights[f"bert.{name}"].data
+
+
+def copy_vision_model_and_projection(hf_model, pt_weights):
+    # copy projection
+    hf_model.visual_projection.weight.data = pt_weights["visual.proj"].data.T
+
+    # copy layer norms
+    copy_linear(hf_model.vision_model.pre_layrnorm, pt_weights, "visual.ln_pre")
+    copy_linear(hf_model.vision_model.post_layernorm, pt_weights, "visual.ln_post")
+
+    # copy embeddings
+    hf_model.vision_model.embeddings.patch_embedding.weight.data = pt_weights["visual.conv1.weight"].data
+    hf_model.vision_model.embeddings.class_embedding.data = pt_weights["visual.class_embedding"].data
+    hf_model.vision_model.embeddings.position_embedding.weight.data = pt_weights["visual.positional_embedding"].data
+
+    # copy encoder
+    copy_layers(hf_model.vision_model.encoder.layers, pt_weights, "visual.transformer.resblocks")
+
+
+@torch.no_grad()
+def convert_chinese_clip_checkpoint(checkpoint_path, pytorch_dump_folder_path, config_path=None):
+    """
+    Copy/paste/tweak model's weights to transformers design.
+    """
+
+    assert config_path is not None, "Please specify the ChineseCLIP model config of the corresponding model size."
+    config = ChineseCLIPConfig.from_pretrained(config_path)
+
+    hf_model = ChineseCLIPModel(config).eval()
+
+    pt_weights = torch.load(checkpoint_path, map_location="cpu")["state_dict"]
+    pt_weights = {(name[7:] if name.startswith("module.") else name): value for name, value in pt_weights.items()}
+
+    copy_text_model_and_projection(hf_model, pt_weights)
+    copy_vision_model_and_projection(hf_model, pt_weights)
+    hf_model.logit_scale.data = pt_weights["logit_scale"].data
+
+    hf_model.save_pretrained(pytorch_dump_folder_path)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--pytorch_dump_folder_path",
+        default=None,
+        type=str,
+        help="Path to the output folder storing converted hf PyTorch model.",
+    )
+    parser.add_argument(
+        "--checkpoint_path", default=None, type=str, help="Path to original github format ChineseCLIP checkpoint."
+    )
+    parser.add_argument(
+        "--config_path", default=None, required=True, type=str, help="Path to hf config.json of model to convert."
+    )
+    args = parser.parse_args()
+
+    convert_chinese_clip_checkpoint(args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path)
+    print("The conversion is finished!")

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/chinese_clip/feature_extraction_chinese_clip.py

@@ -0,0 +1,33 @@
+# coding=utf-8
+# Copyright 2021 The OFA-Sys Team Authors and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Feature extractor class for Chinese-CLIP."""
+
+import warnings
+
+from ...utils import logging
+from .image_processing_chinese_clip import ChineseCLIPImageProcessor
+
+
+logger = logging.get_logger(__name__)
+
+
+class ChineseCLIPFeatureExtractor(ChineseCLIPImageProcessor):
+    def __init__(self, *args, **kwargs) -> None:
+        warnings.warn(
+            "The class ChineseCLIPFeatureExtractor is deprecated and will be removed in version 5 of Transformers."
+            " Please use ChineseCLIPImageProcessor instead.",
+            FutureWarning,
+        )
+        super().__init__(*args, **kwargs)

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/efficientformer/__init__.py

@@ -0,0 +1,109 @@
+# Copyright 2022 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    is_tf_available,
+    is_torch_available,
+    is_vision_available,
+)
+
+
+_import_structure = {
+    "configuration_efficientformer": [
+        "EFFICIENTFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "EfficientFormerConfig",
+    ]
+}
+
+try:
+    if not is_vision_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["image_processing_efficientformer"] = ["EfficientFormerImageProcessor"]
+
+try:
+    if not is_torch_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_efficientformer"] = [
+        "EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+        "EfficientFormerForImageClassification",
+        "EfficientFormerForImageClassificationWithTeacher",
+        "EfficientFormerModel",
+        "EfficientFormerPreTrainedModel",
+    ]
+
+try:
+    if not is_tf_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_tf_efficientformer"] = [
+        "TF_EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+        "TFEfficientFormerForImageClassification",
+        "TFEfficientFormerForImageClassificationWithTeacher",
+        "TFEfficientFormerModel",
+        "TFEfficientFormerPreTrainedModel",
+    ]
+
+if TYPE_CHECKING:
+    from .configuration_efficientformer import EFFICIENTFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, EfficientFormerConfig
+
+    try:
+        if not is_vision_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .image_processing_efficientformer import EfficientFormerImageProcessor
+
+    try:
+        if not is_torch_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_efficientformer import (
+            EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+            EfficientFormerForImageClassification,
+            EfficientFormerForImageClassificationWithTeacher,
+            EfficientFormerModel,
+            EfficientFormerPreTrainedModel,
+        )
+    try:
+        if not is_tf_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_tf_efficientformer import (
+            TF_EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+            TFEfficientFormerForImageClassification,
+            TFEfficientFormerForImageClassificationWithTeacher,
+            TFEfficientFormerModel,
+            TFEfficientFormerPreTrainedModel,
+        )
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/univnet/__init__.py

@@ -0,0 +1,65 @@
+# Copyright 2023 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    is_torch_available,
+)
+
+
+_import_structure = {
+    "configuration_univnet": [
+        "UNIVNET_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "UnivNetConfig",
+    ],
+    "feature_extraction_univnet": ["UnivNetFeatureExtractor"],
+}
+
+try:
+    if not is_torch_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_univnet"] = [
+        "UNIVNET_PRETRAINED_MODEL_ARCHIVE_LIST",
+        "UnivNetModel",
+    ]
+
+
+if TYPE_CHECKING:
+    from .configuration_univnet import (
+        UNIVNET_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        UnivNetConfig,
+    )
+    from .feature_extraction_univnet import UnivNetFeatureExtractor
+
+    try:
+        if not is_torch_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_univnet import (
+            UNIVNET_PRETRAINED_MODEL_ARCHIVE_LIST,
+            UnivNetModel,
+        )
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/univnet/convert_univnet.py

@@ -0,0 +1,162 @@
+# Copyright 2023 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+
+import torch
+
+from transformers import UnivNetConfig, UnivNetModel, logging
+
+
+logging.set_verbosity_info()
+logger = logging.get_logger("transformers.models.univnet")
+
+
+def get_kernel_predictor_key_mapping(config: UnivNetConfig, old_prefix: str = "", new_prefix: str = ""):
+    mapping = {}
+    # Initial conv layer
+    mapping[f"{old_prefix}.input_conv.0.weight_g"] = f"{new_prefix}.input_conv.weight_g"
+    mapping[f"{old_prefix}.input_conv.0.weight_v"] = f"{new_prefix}.input_conv.weight_v"
+    mapping[f"{old_prefix}.input_conv.0.bias"] = f"{new_prefix}.input_conv.bias"
+
+    # Kernel predictor resnet blocks
+    for i in range(config.kernel_predictor_num_blocks):
+        mapping[f"{old_prefix}.residual_convs.{i}.1.weight_g"] = f"{new_prefix}.resblocks.{i}.conv1.weight_g"
+        mapping[f"{old_prefix}.residual_convs.{i}.1.weight_v"] = f"{new_prefix}.resblocks.{i}.conv1.weight_v"
+        mapping[f"{old_prefix}.residual_convs.{i}.1.bias"] = f"{new_prefix}.resblocks.{i}.conv1.bias"
+
+        mapping[f"{old_prefix}.residual_convs.{i}.3.weight_g"] = f"{new_prefix}.resblocks.{i}.conv2.weight_g"
+        mapping[f"{old_prefix}.residual_convs.{i}.3.weight_v"] = f"{new_prefix}.resblocks.{i}.conv2.weight_v"
+        mapping[f"{old_prefix}.residual_convs.{i}.3.bias"] = f"{new_prefix}.resblocks.{i}.conv2.bias"
+
+    # Kernel output conv
+    mapping[f"{old_prefix}.kernel_conv.weight_g"] = f"{new_prefix}.kernel_conv.weight_g"
+    mapping[f"{old_prefix}.kernel_conv.weight_v"] = f"{new_prefix}.kernel_conv.weight_v"
+    mapping[f"{old_prefix}.kernel_conv.bias"] = f"{new_prefix}.kernel_conv.bias"
+
+    # Bias output conv
+    mapping[f"{old_prefix}.bias_conv.weight_g"] = f"{new_prefix}.bias_conv.weight_g"
+    mapping[f"{old_prefix}.bias_conv.weight_v"] = f"{new_prefix}.bias_conv.weight_v"
+    mapping[f"{old_prefix}.bias_conv.bias"] = f"{new_prefix}.bias_conv.bias"
+
+    return mapping
+
+
+def get_key_mapping(config: UnivNetConfig):
+    mapping = {}
+
+    # NOTE: inital conv layer keys are the same
+
+    # LVC Residual blocks
+    for i in range(len(config.resblock_stride_sizes)):
+        # LVCBlock initial convt layer
+        mapping[f"res_stack.{i}.convt_pre.1.weight_g"] = f"resblocks.{i}.convt_pre.weight_g"
+        mapping[f"res_stack.{i}.convt_pre.1.weight_v"] = f"resblocks.{i}.convt_pre.weight_v"
+        mapping[f"res_stack.{i}.convt_pre.1.bias"] = f"resblocks.{i}.convt_pre.bias"
+
+        # Kernel predictor
+        kernel_predictor_mapping = get_kernel_predictor_key_mapping(
+            config, old_prefix=f"res_stack.{i}.kernel_predictor", new_prefix=f"resblocks.{i}.kernel_predictor"
+        )
+        mapping.update(kernel_predictor_mapping)
+
+        # LVC Residual blocks
+        for j in range(len(config.resblock_dilation_sizes[i])):
+            mapping[f"res_stack.{i}.conv_blocks.{j}.1.weight_g"] = f"resblocks.{i}.resblocks.{j}.conv.weight_g"
+            mapping[f"res_stack.{i}.conv_blocks.{j}.1.weight_v"] = f"resblocks.{i}.resblocks.{j}.conv.weight_v"
+            mapping[f"res_stack.{i}.conv_blocks.{j}.1.bias"] = f"resblocks.{i}.resblocks.{j}.conv.bias"
+
+    # Output conv layer
+    mapping["conv_post.1.weight_g"] = "conv_post.weight_g"
+    mapping["conv_post.1.weight_v"] = "conv_post.weight_v"
+    mapping["conv_post.1.bias"] = "conv_post.bias"
+
+    return mapping
+
+
+def rename_state_dict(state_dict, keys_to_modify, keys_to_remove):
+    model_state_dict = {}
+    for key, value in state_dict.items():
+        if key in keys_to_remove:
+            continue
+
+        if key in keys_to_modify:
+            new_key = keys_to_modify[key]
+            model_state_dict[new_key] = value
+        else:
+            model_state_dict[key] = value
+    return model_state_dict
+
+
+def convert_univnet_checkpoint(
+    checkpoint_path,
+    pytorch_dump_folder_path,
+    config_path=None,
+    repo_id=None,
+    safe_serialization=False,
+):
+    model_state_dict_base = torch.load(checkpoint_path, map_location="cpu")
+    # Get the generator's state dict
+    state_dict = model_state_dict_base["model_g"]
+
+    if config_path is not None:
+        config = UnivNetConfig.from_pretrained(config_path)
+    else:
+        config = UnivNetConfig()
+
+    keys_to_modify = get_key_mapping(config)
+    keys_to_remove = set()
+    hf_state_dict = rename_state_dict(state_dict, keys_to_modify, keys_to_remove)
+
+    model = UnivNetModel(config)
+    # Apply weight norm since the original checkpoint has weight norm applied
+    model.apply_weight_norm()
+    model.load_state_dict(hf_state_dict)
+    # Remove weight norm in preparation for inference
+    model.remove_weight_norm()
+
+    model.save_pretrained(pytorch_dump_folder_path, safe_serialization=safe_serialization)
+
+    if repo_id:
+        print("Pushing to the hub...")
+        model.push_to_hub(repo_id)
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--checkpoint_path", required=True, default=None, type=str, help="Path to original checkpoint")
+    parser.add_argument("--config_path", default=None, type=str, help="Path to hf config.json of model to convert")
+    parser.add_argument(
+        "--pytorch_dump_folder_path", required=True, default=None, type=str, help="Path to the output PyTorch model."
+    )
+    parser.add_argument(
+        "--push_to_hub", default=None, type=str, help="Where to upload the converted model on the 🤗 hub."
+    )
+    parser.add_argument(
+        "--safe_serialization", action="store_true", help="Whether to save the model using `safetensors`."
+    )
+
+    args = parser.parse_args()
+
+    convert_univnet_checkpoint(
+        args.checkpoint_path,
+        args.pytorch_dump_folder_path,
+        args.config_path,
+        args.push_to_hub,
+        args.safe_serialization,
+    )
+
+
+if __name__ == "__main__":
+    main()

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/univnet/feature_extraction_univnet.py

@@ -0,0 +1,456 @@
+# Copyright 2023 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Feature extractor class for UnivNetModel."""
+
+from typing import Any, Dict, List, Optional, Union
+
+import numpy as np
+
+from ...audio_utils import mel_filter_bank, optimal_fft_length, spectrogram, window_function
+from ...feature_extraction_sequence_utils import SequenceFeatureExtractor
+from ...feature_extraction_utils import BatchFeature
+from ...utils import PaddingStrategy, TensorType, logging
+
+
+logger = logging.get_logger(__name__)
+
+
+class UnivNetFeatureExtractor(SequenceFeatureExtractor):
+    r"""
+    Constructs a UnivNet feature extractor.
+
+    This class extracts log-mel-filter bank features from raw speech using the short time Fourier Transform (STFT). The
+    STFT implementation follows that of TacoTron 2 and Hifi-GAN.
+
+    This feature extractor inherits from [`~feature_extraction_sequence_utils.SequenceFeatureExtractor`] which contains
+    most of the main methods. Users should refer to this superclass for more information regarding those methods.
+
+    Args:
+        feature_size (`int`, *optional*, defaults to 1):
+            The feature dimension of the extracted features.
+        sampling_rate (`int`, *optional*, defaults to 24000):
+            The sampling rate at which the audio files should be digitalized expressed in hertz (Hz).
+        padding_value (`float`, *optional*, defaults to 0.0):
+            The value to pad with when applying the padding strategy defined by the `padding` argument to
+            [`UnivNetFeatureExtractor.__call__`]. Should correspond to audio silence. The `pad_end` argument to
+            `__call__` will also use this padding value.
+        do_normalize (`bool`, *optional*, defaults to `False`):
+            Whether to perform Tacotron 2 normalization on the input. Normalizing can help to significantly improve the
+            performance for some models.
+        num_mel_bins (`int`, *optional*, defaults to 100):
+            The number of mel-frequency bins in the extracted spectrogram features. This should match
+            `UnivNetModel.config.num_mel_bins`.
+        hop_length (`int`, *optional*, defaults to 256):
+            The direct number of samples between sliding windows. Otherwise referred to as "shift" in many papers. Note
+            that this is different from other audio feature extractors such as [`SpeechT5FeatureExtractor`] which take
+            the `hop_length` in ms.
+        win_length (`int`, *optional*, defaults to 1024):
+            The direct number of samples for each sliding window. Note that this is different from other audio feature
+            extractors such as [`SpeechT5FeatureExtractor`] which take the `win_length` in ms.
+        win_function (`str`, *optional*, defaults to `"hann_window"`):
+            Name for the window function used for windowing, must be accessible via `torch.{win_function}`
+        filter_length (`int`, *optional*, defaults to 1024):
+            The number of FFT components to use. If `None`, this is determined using
+            `transformers.audio_utils.optimal_fft_length`.
+        max_length_s (`int`, *optional*, defaults to 10):
+            The maximum input lenght of the model in seconds. This is used to pad the audio.
+        fmin (`float`, *optional*, defaults to 0.0):
+            Minimum mel frequency in Hz.
+        fmax (`float`, *optional*):
+            Maximum mel frequency in Hz. If not set, defaults to `sampling_rate / 2`.
+        mel_floor (`float`, *optional*, defaults to 1e-09):
+            Minimum value of mel frequency banks. Note that the way [`UnivNetFeatureExtractor`] uses `mel_floor` is
+            different than in [`transformers.audio_utils.spectrogram`].
+        center (`bool`, *optional*, defaults to `False`):
+            Whether to pad the waveform so that frame `t` is centered around time `t * hop_length`. If `False`, frame
+            `t` will start at time `t * hop_length`.
+        compression_factor (`float`, *optional*, defaults to 1.0):
+            The multiplicative compression factor for dynamic range compression during spectral normalization.
+        compression_clip_val (`float`, *optional*, defaults to 1e-05):
+            The clip value applied to the waveform before applying dynamic range compression during spectral
+            normalization.
+        normalize_min (`float`, *optional*, defaults to -11.512925148010254):
+            The min value used for Tacotron 2-style linear normalization. The default is the original value from the
+            Tacotron 2 implementation.
+        normalize_max (`float`, *optional*, defaults to 2.3143386840820312):
+            The max value used for Tacotron 2-style linear normalization. The default is the original value from the
+            Tacotron 2 implementation.
+        model_in_channels (`int`, *optional*, defaults to 64):
+            The number of input channels to the [`UnivNetModel`] model. This should match
+            `UnivNetModel.config.model_in_channels`.
+        pad_end_length (`int`, *optional*, defaults to 10):
+            If padding the end of each waveform, the number of spectrogram frames worth of samples to append. The
+            number of appended samples will be `pad_end_length * hop_length`.
+        return_attention_mask (`bool`, *optional*, defaults to `True`):
+            Whether or not [`~UnivNetFeatureExtractor.__call__`] should return `attention_mask`.
+    """
+
+    model_input_names = ["input_features", "noise_sequence", "padding_mask"]
+
+    def __init__(
+        self,
+        feature_size: int = 1,
+        sampling_rate: int = 24000,
+        padding_value: float = 0.0,
+        do_normalize: bool = False,
+        num_mel_bins: int = 100,
+        hop_length: int = 256,
+        win_length: int = 1024,
+        win_function: str = "hann_window",
+        filter_length: Optional[int] = 1024,
+        max_length_s: int = 10,
+        fmin: float = 0.0,
+        fmax: Optional[float] = None,
+        mel_floor: float = 1e-9,
+        center: bool = False,
+        compression_factor: float = 1.0,
+        compression_clip_val: float = 1e-5,
+        normalize_min: float = -11.512925148010254,
+        normalize_max: float = 2.3143386840820312,
+        model_in_channels: int = 64,
+        pad_end_length: int = 10,
+        return_attention_mask=True,
+        **kwargs,
+    ):
+        super().__init__(
+            feature_size=feature_size,
+            sampling_rate=sampling_rate,
+            padding_value=padding_value,
+            return_attention_mask=return_attention_mask,
+            **kwargs,
+        )
+
+        self.do_normalize = do_normalize
+
+        self.num_mel_bins = num_mel_bins
+        self.hop_length = hop_length
+        self.win_length = win_length
+        self.win_function = win_function
+        self.filter_length = filter_length
+        self.fmin = fmin
+        if fmax is None:
+            # Follows the librosa.filters.mel implementation
+            fmax = float(sampling_rate) / 2
+        self.fmax = fmax
+        self.mel_floor = mel_floor
+
+        self.max_length_s = max_length_s
+        self.num_max_samples = max_length_s * sampling_rate
+
+        if self.filter_length is None:
+            self.n_fft = optimal_fft_length(self.win_length)
+        else:
+            self.n_fft = self.filter_length
+        self.n_freqs = (self.n_fft // 2) + 1
+
+        self.window = window_function(window_length=self.win_length, name=self.win_function, periodic=True)
+
+        self.mel_filters = mel_filter_bank(
+            num_frequency_bins=self.n_freqs,
+            num_mel_filters=self.num_mel_bins,
+            min_frequency=self.fmin,
+            max_frequency=self.fmax,
+            sampling_rate=self.sampling_rate,
+            norm="slaney",
+            mel_scale="slaney",
+        )
+
+        self.center = center
+        self.compression_factor = compression_factor
+        self.compression_clip_val = compression_clip_val
+        self.normalize_min = normalize_min
+        self.normalize_max = normalize_max
+        self.model_in_channels = model_in_channels
+        self.pad_end_length = pad_end_length
+
+    def normalize(self, spectrogram):
+        return 2 * ((spectrogram - self.normalize_min) / (self.normalize_max - self.normalize_min)) - 1
+
+    def denormalize(self, spectrogram):
+        return self.normalize_min + (self.normalize_max - self.normalize_min) * ((spectrogram + 1) / 2)
+
+    def mel_spectrogram(self, waveform: np.ndarray) -> np.ndarray:
+        """
+        Calculates log MEL spectrograms from a batch of waveforms. Note that the input waveform(s) will be padded by
+        `int(self.n_fft - self.hop_length) / 2` on both sides using the `reflect` padding mode.
+
+        Args:
+            waveform (`np.ndarray` of shape `(length,)`):
+                The input waveform. This must be a single real-valued, mono waveform.
+
+        Returns:
+            `numpy.ndarray`: Array containing a log-mel spectrogram of shape `(num_frames, num_mel_bins)`.
+        """
+        # Do custom padding based on the official MelGAN and Hifi-GAN implementations
+        # See https://github.com/maum-ai/univnet/blob/9bb2b54838bb6d7ce767131cc7b8b61198bc7558/utils/stft.py#L84-L86
+        waveform = np.pad(
+            waveform,
+            (int((self.n_fft - self.hop_length) / 2), int((self.n_fft - self.hop_length) / 2)),
+            mode="reflect",
+        )
+
+        # Get the complex spectrogram.
+        # Note: waveform must be unbatched currently due to the implementation of spectrogram(...).
+        complex_spectrogram = spectrogram(
+            waveform,
+            window=self.window,
+            frame_length=self.n_fft,
+            hop_length=self.hop_length,
+            fft_length=self.n_fft,
+            power=None,
+            center=self.center,
+            mel_filters=None,
+            mel_floor=None,
+        )
+
+        # Apply the MEL filter bank and MEL floor manually since UnivNet uses a slightly different implementation
+        amplitude_spectrogram = np.sqrt(
+            np.real(complex_spectrogram) ** 2 + np.imag(complex_spectrogram) ** 2 + self.mel_floor
+        )
+        mel_spectrogram = np.matmul(self.mel_filters.T, amplitude_spectrogram)
+
+        # Perform spectral normalization to get the log mel spectrogram.
+        log_mel_spectrogram = np.log(
+            np.clip(mel_spectrogram, a_min=self.compression_clip_val, a_max=None) * self.compression_factor
+        )
+
+        # Return spectrogram with num_mel_bins last
+        return log_mel_spectrogram.T
+
+    def generate_noise(
+        self,
+        noise_length: int,
+        generator: Optional[np.random.Generator] = None,
+    ) -> np.ndarray:
+        """
+        Generates a random noise sequence of standard Gaussian noise for use in the `noise_sequence` argument of
+        [`UnivNetModel.forward`].
+
+        Args:
+            spectrogram_length (`int`):
+                The length (dim 0) of the generated noise.
+            model_in_channels (`int`, *optional*, defaults to `None`):
+                The number of features (dim 1) of the generated noise. This should correspond to the
+                `model_in_channels` of the [`UnivNetGan`] model. If not set, this will default to
+                `self.config.model_in_channels`.
+            generator (`numpy.random.Generator`, *optional*, defaults to `None`)
+                An optional `numpy.random.Generator` random number generator to control noise generation. If not set, a
+                new generator with fresh entropy will be created.
+
+        Returns:
+            `numpy.ndarray`: Array containing random standard Gaussian noise of shape `(noise_length,
+            model_in_channels)`.
+        """
+        if generator is None:
+            generator = np.random.default_rng()
+
+        noise_shape = (noise_length, self.model_in_channels)
+        noise = generator.standard_normal(noise_shape, dtype=np.float32)
+
+        return noise
+
+    def batch_decode(self, waveforms, waveform_lengths=None) -> List[np.ndarray]:
+        r"""
+        Removes padding from generated audio after running [`UnivNetModel.forward`]. This returns a ragged list of 1D
+        audio waveform arrays and not a single tensor/array because in general the waveforms will have different
+        lengths after removing padding.
+
+        Args:
+            waveforms (`torch.FloatTensor` of shape `(batch_size, sequence_length)`):
+                The batched output waveforms from the [`UnivNetModel`].
+            waveform_lengths (`torch.FloatTensor` of shape `(batch_size,)`, *optional*):
+                The batched lengths of each waveform before padding.
+
+        Returns:
+            `List[np.ndarray]`: A ragged list of 1D waveform arrays with padding removed.
+        """
+        # Collapse the batched waveform tensor to a list of 1D audio waveforms
+        waveforms = [waveform.detach().clone().cpu().numpy() for waveform in waveforms]
+
+        if waveform_lengths is not None:
+            waveforms = [waveform[: waveform_lengths[i]] for i, waveform in enumerate(waveforms)]
+
+        return waveforms
+
+    def __call__(
+        self,
+        raw_speech: Union[np.ndarray, List[float], List[np.ndarray], List[List[float]]],
+        sampling_rate: Optional[int] = None,
+        padding: Union[bool, str, PaddingStrategy] = True,
+        max_length: Optional[int] = None,
+        truncation: bool = True,
+        pad_to_multiple_of: Optional[int] = None,
+        return_noise: bool = True,
+        generator: Optional[np.random.Generator] = None,
+        pad_end: bool = False,
+        pad_length: Optional[int] = None,
+        do_normalize: Optional[str] = None,
+        return_attention_mask: Optional[bool] = None,
+        return_tensors: Optional[Union[str, TensorType]] = None,
+    ) -> BatchFeature:
+        """
+        Main method to featurize and prepare for the model one or several sequence(s).
+
+        Args:
+            raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
+                The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
+                values, a list of numpy arrays or a list of list of float values. Must be mono channel audio, not
+                stereo, i.e. single float per timestep.
+            sampling_rate (`int`, *optional*):
+                The sampling rate at which the `raw_speech` input was sampled. It is strongly recommended to pass
+                `sampling_rate` at the forward call to prevent silent errors and allow automatic speech recognition
+                pipeline.
+            padding (`bool`, `str` or [`~utils.PaddingStrategy`], *optional*, defaults to `True`):
+                Select a strategy to pad the input `raw_speech` waveforms (according to the model's padding side and
+                padding index) among:
+
+                - `True` or `'longest'`: Pad to the longest sequence in the batch (or no padding if only a single
+                  sequence if provided).
+                - `'max_length'`: Pad to a maximum length specified with the argument `max_length` or to the maximum
+                  acceptable input length for the model if that argument is not provided.
+                - `False` or `'do_not_pad'` (default): No padding (i.e., can output a batch with sequences of different
+                  lengths).
+
+                If `pad_end = True`, that padding will occur before the `padding` strategy is applied.
+            max_length (`int`, *optional*):
+                Maximum length of the returned list and optionally padding length (see above).
+            truncation (`bool`, *optional*, defaults to `True`):
+                Activates truncation to cut input sequences longer than `max_length` to `max_length`.
+            pad_to_multiple_of (`int`, *optional*):
+                If set will pad the sequence to a multiple of the provided value.
+
+                This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability
+                `>= 7.5` (Volta), or on TPUs which benefit from having sequence lengths be a multiple of 128.
+            return_noise (`bool`, *optional*, defaults to `True`):
+                Whether to generate and return a noise waveform for use in [`UnivNetModel.forward`].
+            generator (`numpy.random.Generator`, *optional*, defaults to `None`):
+                An optional `numpy.random.Generator` random number generator to use when generating noise.
+            pad_end (`bool`, *optional*, defaults to `False`):
+                Whether to pad the end of each waveform with silence. This can help reduce artifacts at the end of the
+                generated audio sample; see https://github.com/seungwonpark/melgan/issues/8 for more details. This
+                padding will be done before the padding strategy specified in `padding` is performed.
+            pad_length (`int`, *optional*, defaults to `None`):
+                If padding the end of each waveform, the length of the padding in spectrogram frames. If not set, this
+                will default to `self.config.pad_end_length`.
+            do_normalize (`bool`, *optional*):
+                Whether to perform Tacotron 2 normalization on the input. Normalizing can help to significantly improve
+                the performance for some models. If not set, this will default to `self.config.do_normalize`.
+            return_attention_mask (`bool`, *optional*):
+                Whether to return the attention mask. If left to the default, will return the attention mask according
+                to the specific feature_extractor's default.
+
+                [What are attention masks?](../glossary#attention-mask)
+
+            return_tensors (`str` or [`~utils.TensorType`], *optional*):
+                If set, will return tensors instead of list of python integers. Acceptable values are:
+
+                - `'tf'`: Return TensorFlow `tf.constant` objects.
+                - `'pt'`: Return PyTorch `torch.np.array` objects.
+                - `'np'`: Return Numpy `np.ndarray` objects.
+        """
+        do_normalize = do_normalize if do_normalize is not None else self.do_normalize
+
+        if sampling_rate is not None:
+            if sampling_rate != self.sampling_rate:
+                raise ValueError(
+                    f"The model corresponding to this feature extractor: {self.__class__.__name__} was trained using a"
+                    f" sampling rate of {self.sampling_rate}. Please make sure that the provided `raw_speech` input"
+                    f" was sampled with {self.sampling_rate} and not {sampling_rate}."
+                )
+        else:
+            logger.warning(
+                "It is strongly recommended to pass the `sampling_rate` argument to this function. "
+                "Failing to do so can result in silent errors that might be hard to debug."
+            )
+
+        is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        is_batched = is_batched_numpy or (
+            isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
+        )
+
+        if is_batched:
+            raw_speech = [np.asarray(speech, dtype=np.float32) for speech in raw_speech]
+        elif not is_batched and not isinstance(raw_speech, np.ndarray):
+            raw_speech = np.asarray(raw_speech, dtype=np.float32)
+        elif isinstance(raw_speech, np.ndarray) and raw_speech.dtype is np.dtype(np.float64):
+            raw_speech = raw_speech.astype(np.float32)
+
+        # always return batch
+        if not is_batched:
+            raw_speech = [np.asarray(raw_speech, dtype=np.float32)]
+
+        # Pad end to reduce artifacts
+        if pad_end:
+            pad_length = pad_length if pad_length is not None else self.pad_end_length
+            raw_speech = [
+                np.pad(waveform, (0, pad_length * self.hop_length), constant_values=self.padding_value)
+                for waveform in raw_speech
+            ]
+
+        batched_speech = BatchFeature({"input_features": raw_speech})
+
+        padded_inputs = self.pad(
+            batched_speech,
+            padding=padding,
+            max_length=max_length if max_length is not None else self.num_max_samples,
+            truncation=truncation,
+            pad_to_multiple_of=pad_to_multiple_of,
+            return_attention_mask=return_attention_mask,
+        )
+
+        # make sure list is in array format
+        # input_features = padded_inputs.get("input_features").transpose(2, 0, 1)
+        input_features = padded_inputs.get("input_features")
+
+        mel_spectrograms = [self.mel_spectrogram(waveform) for waveform in input_features]
+
+        if isinstance(input_features[0], List):
+            batched_speech["input_features"] = [np.asarray(mel, dtype=np.float32) for mel in mel_spectrograms]
+        else:
+            batched_speech["input_features"] = [mel.astype(np.float32) for mel in mel_spectrograms]
+
+        # convert attention_mask to correct format
+        attention_mask = padded_inputs.get("attention_mask")
+        if attention_mask is not None:
+            batched_speech["padding_mask"] = [np.asarray(array, dtype=np.int32) for array in attention_mask]
+
+        if return_noise:
+            noise = [
+                self.generate_noise(spectrogram.shape[0], generator)
+                for spectrogram in batched_speech["input_features"]
+            ]
+            batched_speech["noise_sequence"] = noise
+
+        if do_normalize:
+            batched_speech["input_features"] = [
+                self.normalize(spectrogram) for spectrogram in batched_speech["input_features"]
+            ]
+
+        if return_tensors is not None:
+            batched_speech = batched_speech.convert_to_tensors(return_tensors)
+
+        return batched_speech
+
+    def to_dict(self) -> Dict[str, Any]:
+        output = super().to_dict()
+
+        # Don't serialize these as they are derived from the other properties.
+        names = ["window", "mel_filters", "n_fft", "n_freqs", "num_max_samples"]
+        for name in names:
+            if name in output:
+                del output[name]
+
+        return output

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/univnet/modeling_univnet.py

@@ -0,0 +1,636 @@
+# Copyright 2023 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch UnivNetModel model."""
+
+from dataclasses import dataclass
+from typing import Optional, Tuple, Union
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+
+from ...modeling_utils import ModelOutput, PreTrainedModel
+from ...utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
+from .configuration_univnet import UnivNetConfig
+
+
+logger = logging.get_logger(__name__)
+
+# General docstring
+_CONFIG_FOR_DOC = "UnivNetConfig"
+
+_CHECKPOINT_FOR_DOC = "dg845/univnet-dev"
+
+UNIVNET_PRETRAINED_MODEL_ARCHIVE_LIST = [
+    "dg845/univnet-dev",
+    # See all UnivNet models at https://huggingface.co/models?filter=univnet
+]
+
+
+@dataclass
+class UnivNetModelOutput(ModelOutput):
+    """
+    Output class for the [`UnivNetModel`], which includes the generated audio waveforms and the original unpadded
+    lengths of those waveforms (so that the padding can be removed by [`UnivNetModel.batch_decode`]).
+
+    Args:
+        waveforms (`torch.FloatTensor` of shape `(batch_size, sequence_length)`):
+            Batched 1D (mono-channel) output audio waveforms.
+        waveform_lengths (`torch.FloatTensor` of shape `(batch_size,)`):
+            The batched length in samples of each unpadded waveform in `waveforms`.
+    """
+
+    waveforms: torch.FloatTensor = None
+    waveform_lengths: torch.FloatTensor = None
+
+
+class UnivNetKernelPredictorResidualBlock(nn.Module):
+    """
+    Implementation of the residual block for the kernel predictor network inside each location variable convolution
+    block (LVCBlock).
+
+    Parameters:
+        config: (`UnivNetConfig`):
+            Config for the `UnivNetModel` model.
+    """
+
+    def __init__(
+        self,
+        config: UnivNetConfig,
+    ):
+        super().__init__()
+        self.channels = config.model_in_channels
+        self.kernel_size = config.kernel_predictor_conv_size
+        self.dropout_prob = config.kernel_predictor_dropout
+        self.leaky_relu_slope = config.leaky_relu_slope
+
+        padding = (self.kernel_size - 1) // 2
+
+        self.dropout = nn.Dropout(self.dropout_prob)
+        self.conv1 = nn.Conv1d(self.channels, self.channels, self.kernel_size, padding=padding, bias=True)
+        self.conv2 = nn.Conv1d(self.channels, self.channels, self.kernel_size, padding=padding, bias=True)
+
+    def forward(self, hidden_states: torch.FloatTensor):
+        # hidden_states should have shape (batch_size, channels, seq_length)
+        residual = hidden_states
+        hidden_states = self.dropout(hidden_states)
+        hidden_states = self.conv1(hidden_states)
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+        hidden_states = self.conv2(hidden_states)
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+        return hidden_states + residual
+
+    def apply_weight_norm(self):
+        nn.utils.weight_norm(self.conv1)
+        nn.utils.weight_norm(self.conv2)
+
+    def remove_weight_norm(self):
+        nn.utils.remove_weight_norm(self.conv1)
+        nn.utils.remove_weight_norm(self.conv2)
+
+
+class UnivNetKernelPredictor(nn.Module):
+    """
+    Implementation of the kernel predictor network which supplies the kernel and bias for the location variable
+    convolutional layers (LVCs) in each UnivNet LVCBlock.
+
+    Based on the KernelPredictor implementation in
+    [maum-ai/univnet](https://github.com/maum-ai/univnet/blob/9bb2b54838bb6d7ce767131cc7b8b61198bc7558/model/lvcnet.py#L7).
+
+    Parameters:
+        config: (`UnivNetConfig`):
+            Config for the `UnivNetModel` model.
+        conv_kernel_size (`int`, *optional*, defaults to 3):
+            The kernel size for the location variable convolutional layer kernels (convolutional weight tensor).
+        conv_layers (`int`, *optional*, defaults to 4):
+            The number of location variable convolutional layers to output kernels and biases for.
+    """
+
+    def __init__(
+        self,
+        config: UnivNetConfig,
+        conv_kernel_size: int = 3,
+        conv_layers: int = 4,
+    ):
+        super().__init__()
+
+        self.conv_in_channels = config.model_hidden_channels
+        self.conv_out_channels = 2 * config.model_hidden_channels
+        self.conv_kernel_size = conv_kernel_size
+        self.conv_layers = conv_layers
+
+        self.kernel_channels = (
+            self.conv_in_channels * self.conv_out_channels * self.conv_kernel_size * self.conv_layers
+        )
+        self.bias_channels = self.conv_out_channels * self.conv_layers
+
+        self.resnet_in_channels = config.num_mel_bins
+        self.resnet_hidden_channels = config.kernel_predictor_hidden_channels
+        self.resnet_kernel_size = config.kernel_predictor_conv_size
+        self.num_blocks = config.kernel_predictor_num_blocks
+
+        self.leaky_relu_slope = config.leaky_relu_slope
+
+        padding = (self.resnet_kernel_size - 1) // 2
+
+        self.input_conv = nn.Conv1d(self.resnet_in_channels, self.resnet_hidden_channels, 5, padding=2, bias=True)
+
+        self.resblocks = nn.ModuleList([UnivNetKernelPredictorResidualBlock(config) for _ in range(self.num_blocks)])
+
+        self.kernel_conv = nn.Conv1d(
+            self.resnet_hidden_channels, self.kernel_channels, self.resnet_kernel_size, padding=padding, bias=True
+        )
+        self.bias_conv = nn.Conv1d(
+            self.resnet_hidden_channels, self.bias_channels, self.resnet_kernel_size, padding=padding, bias=True
+        )
+
+    def forward(self, spectrogram: torch.FloatTensor):
+        """
+        Maps a conditioning log-mel spectrogram to a tensor of convolutional kernels and biases, for use in location
+        variable convolutional layers. Note that the input spectrogram should have shape (batch_size, input_channels,
+        seq_length).
+
+        Args:
+            spectrogram (`torch.FloatTensor` of shape `(batch_size, input_channels, seq_length)`):
+                Tensor containing the log-mel spectrograms.
+
+        Returns:
+            Tuple[`torch.FloatTensor, `torch.FloatTensor`]: tuple of tensors where the first element is the tensor of
+            location variable convolution kernels of shape `(batch_size, self.conv_layers, self.conv_in_channels,
+            self.conv_out_channels, self.conv_kernel_size, seq_length)` and the second element is the tensor of
+            location variable convolution biases of shape `(batch_size, self.conv_layers. self.conv_out_channels,
+            seq_length)`.
+        """
+        batch_size, _, seq_length = spectrogram.shape
+
+        hidden_states = self.input_conv(spectrogram)
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+
+        for resblock in self.resblocks:
+            hidden_states = resblock(hidden_states)
+
+        kernel_hidden_states = self.kernel_conv(hidden_states)
+        bias_hidden_states = self.bias_conv(hidden_states)
+
+        # Reshape kernels and biases to appropriate shape
+        kernels = kernel_hidden_states.view(
+            batch_size,
+            self.conv_layers,
+            self.conv_in_channels,
+            self.conv_out_channels,
+            self.conv_kernel_size,
+            seq_length,
+        ).contiguous()
+        biases = bias_hidden_states.view(
+            batch_size,
+            self.conv_layers,
+            self.conv_out_channels,
+            seq_length,
+        ).contiguous()
+
+        return kernels, biases
+
+    def apply_weight_norm(self):
+        nn.utils.weight_norm(self.input_conv)
+        for layer in self.resblocks:
+            layer.apply_weight_norm()
+        nn.utils.weight_norm(self.kernel_conv)
+        nn.utils.weight_norm(self.bias_conv)
+
+    def remove_weight_norm(self):
+        nn.utils.remove_weight_norm(self.input_conv)
+        for layer in self.resblocks:
+            layer.remove_weight_norm()
+        nn.utils.remove_weight_norm(self.kernel_conv)
+        nn.utils.remove_weight_norm(self.bias_conv)
+
+
+class UnivNetLvcResidualBlock(nn.Module):
+    """
+    Implementation of the location variable convolution (LVC) residual block for the UnivNet residual network.
+
+    Parameters:
+        config: (`UnivNetConfig`):
+            Config for the `UnivNetModel` model.
+        kernel_size (`int`):
+            The kernel size for the dilated 1D convolutional layer.
+        dilation (`int`):
+            The dilation for the dilated 1D convolutional layer.
+    """
+
+    def __init__(
+        self,
+        config: UnivNetConfig,
+        kernel_size: int,
+        dilation: int,
+    ):
+        super().__init__()
+        self.hidden_channels = config.model_hidden_channels
+        self.kernel_size = kernel_size
+        self.dilation = dilation
+        self.leaky_relu_slope = config.leaky_relu_slope
+
+        padding = self.dilation * (self.kernel_size - 1) // 2
+
+        self.conv = nn.Conv1d(
+            self.hidden_channels,
+            self.hidden_channels,
+            self.kernel_size,
+            padding=padding,
+            dilation=self.dilation,
+        )
+
+    def forward(self, hidden_states, kernel, bias, hop_size=256):
+        residual = hidden_states
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+        hidden_states = self.conv(hidden_states)
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+        hidden_states = self.location_variable_convolution(hidden_states, kernel, bias, hop_size=hop_size)
+        # Gated activation unit
+        hidden_states = torch.sigmoid(hidden_states[:, : self.hidden_channels, :]) * torch.tanh(
+            hidden_states[:, self.hidden_channels :, :]
+        )
+        # Skip connection
+        hidden_states = residual + hidden_states
+
+        return hidden_states
+
+    # Based on https://github.com/maum-ai/univnet/blob/9bb2b54838bb6d7ce767131cc7b8b61198bc7558/model/lvcnet.py#L171
+    def location_variable_convolution(
+        self,
+        hidden_states: torch.FloatTensor,
+        kernel: torch.FloatTensor,
+        bias: torch.FloatTensor,
+        dilation: int = 1,
+        hop_size: int = 256,
+    ):
+        """
+        Performs location-variable convolution operation on the input sequence (hidden_states) using the local
+        convolution kernel. This was introduced in [LVCNet: Efficient Condition-Dependent Modeling Network for Waveform
+        Generation](https://arxiv.org/abs/2102.10815) by Zhen Zheng, Jianzong Wang, Ning Cheng, and Jing Xiao.
+
+        Time: 414 μs ± 309 ns per loop (mean ± std. dev. of 7 runs, 1000 loops each), test on NVIDIA V100.
+
+        Args:
+            hidden_states (`torch.FloatTensor` of shape `(batch_size, in_channels, in_length)`):
+                The input sequence of shape (batch, in_channels, in_length).
+            kernel (`torch.FloatTensor` of shape `(batch_size, in_channels, out_channels, kernel_size, kernel_length)`):
+                The local convolution kernel of shape (batch, in_channels, out_channels, kernel_size, kernel_length).
+            bias (`torch.FloatTensor` of shape `(batch_size, out_channels, kernel_length)`):
+                The bias for the local convolution of shape (batch, out_channels, kernel_length).
+            dilation (`int`, *optional*, defaults to 1):
+                The dilation of convolution.
+            hop_size (`int`, *optional*, defaults to 256):
+                The hop_size of the conditioning sequence.
+        Returns:
+            `torch.FloatTensor`: the output sequence after performing local convolution with shape (batch_size,
+            out_channels, in_length).
+        """
+        batch, _, in_length = hidden_states.shape
+        batch, _, out_channels, kernel_size, kernel_length = kernel.shape
+        if in_length != (kernel_length * hop_size):
+            raise ValueError(
+                f"Dim 2 of `hidden_states` should be {kernel_length * hop_size}) but got {in_length}. Please check"
+                " `hidden_states` or `kernel` and `hop_size` to make sure they are correct."
+            )
+
+        padding = dilation * int((kernel_size - 1) / 2)
+
+        # (batch, in_channels, in_length + 2*padding)
+        hidden_states = nn.functional.pad(hidden_states, (padding, padding), "constant", 0)
+        # (batch, in_channels, kernel_length, hop_size + 2*padding)
+        hidden_states = hidden_states.unfold(2, hop_size + 2 * padding, hop_size)
+
+        if hop_size < dilation:
+            hidden_states = nn.functional.pad(hidden_states, (0, dilation), "constant", 0)
+        # (batch, in_channels, kernel_length, (hop_size + 2*padding)/dilation, dilation)
+        hidden_states = hidden_states.unfold(3, dilation, dilation)
+        hidden_states = hidden_states[:, :, :, :, :hop_size]
+        # (batch, in_channels, kernel_length, dilation, (hop_size + 2*padding)/dilation)
+        hidden_states = hidden_states.transpose(3, 4)
+        # (batch, in_channels, kernel_length, dilation, _, kernel_size)
+        hidden_states = hidden_states.unfold(4, kernel_size, 1)
+
+        # Apply local convolution kernel to hidden_states.
+        output_hidden_states = torch.einsum("bildsk,biokl->bolsd", hidden_states, kernel)
+
+        output_hidden_states = output_hidden_states.to(memory_format=torch.channels_last_3d)
+        bias = bias.unsqueeze(-1).unsqueeze(-1).to(memory_format=torch.channels_last_3d)
+        output_hidden_states = output_hidden_states + bias
+        output_hidden_states = output_hidden_states.contiguous().view(batch, out_channels, -1)
+
+        return output_hidden_states
+
+    def apply_weight_norm(self):
+        nn.utils.weight_norm(self.conv)
+
+    def remove_weight_norm(self):
+        nn.utils.remove_weight_norm(self.conv)
+
+
+class UnivNetLvcBlock(nn.Module):
+    """
+    Implementation of the location variable convolution (LVC) residual block of the UnivNet residual block. Includes a
+    `UnivNetKernelPredictor` inside to predict the kernels and biases of the LVC layers.
+
+    Based on LVCBlock in
+    [maum-ai/univnet](https://github.com/maum-ai/univnet/blob/9bb2b54838bb6d7ce767131cc7b8b61198bc7558/model/lvcnet.py#L98)
+
+    Parameters:
+        config (`UnivNetConfig`):
+            Config for the `UnivNetModel` model.
+        layer_id (`int`):
+            An integer corresponding to the index of the current LVC resnet block layer. This should be between 0 and
+            `len(config.resblock_stride_sizes) - 1)` inclusive.
+        lvc_hop_size (`int`, *optional*, defaults to 256):
+            The hop size for the location variable convolutional layers.
+    """
+
+    def __init__(
+        self,
+        config: UnivNetConfig,
+        layer_id: int,
+        lvc_hop_size: int = 256,
+    ):
+        super().__init__()
+        self.hidden_channels = config.model_hidden_channels
+        self.kernel_size = config.resblock_kernel_sizes[layer_id]
+        self.stride = config.resblock_stride_sizes[layer_id]
+        self.dilations = config.resblock_dilation_sizes[layer_id]
+        self.cond_hop_length = lvc_hop_size
+        self.leaky_relu_slope = config.leaky_relu_slope
+        self.num_blocks = len(self.dilations)
+
+        self.convt_pre = nn.ConvTranspose1d(
+            self.hidden_channels,
+            self.hidden_channels,
+            2 * self.stride,
+            stride=self.stride,
+            padding=self.stride // 2 + self.stride % 2,
+            output_padding=self.stride % 2,
+        )
+
+        self.kernel_predictor = UnivNetKernelPredictor(config, self.kernel_size, self.num_blocks)
+
+        self.resblocks = nn.ModuleList(
+            [UnivNetLvcResidualBlock(config, self.kernel_size, self.dilations[i]) for i in range(self.num_blocks)]
+        )
+
+    def forward(self, hidden_states: torch.FloatTensor, spectrogram: torch.FloatTensor):
+        # hidden_states: (batch_size, hidden_channels, seq_length)
+        # spectrogram: (batch_size, cond_channels, cond_length)
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+        hidden_states = self.convt_pre(hidden_states)
+
+        kernels, biases = self.kernel_predictor(spectrogram)
+
+        for i, resblock in enumerate(self.resblocks):
+            kernel = kernels[:, i, :, :, :, :]
+            bias = biases[:, i, :, :]
+            hidden_states = resblock(hidden_states, kernel, bias, hop_size=self.cond_hop_length)
+
+        return hidden_states
+
+    def apply_weight_norm(self):
+        nn.utils.weight_norm(self.convt_pre)
+        self.kernel_predictor.apply_weight_norm()
+        for layer in self.resblocks:
+            layer.apply_weight_norm()
+
+    def remove_weight_norm(self):
+        nn.utils.remove_weight_norm(self.convt_pre)
+        self.kernel_predictor.remove_weight_norm()
+        for layer in self.resblocks:
+            layer.remove_weight_norm()
+
+
+UNIVNET_START_DOCSTRING = r"""
+    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
+    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+    etc.)
+
+    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
+    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
+    and behavior.
+
+    Parameters:
+        config ([`UnivNetConfig`]):
+            Model configuration class with all the parameters of the model. Initializing with a config file does not
+            load the weights associated with the model, only the configuration. Check out the
+            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+
+UNIVNET_INPUTS_DOCSTRING = r"""
+    Converts a noise waveform and a conditioning spectrogram to a speech waveform. Passing a batch of log-mel
+    spectrograms returns a batch of speech waveforms. Passing a single, un-batched log-mel spectrogram returns a
+    single, un-batched speech waveform.
+
+    Args:
+        input_features (`torch.FloatTensor`):
+            Tensor containing the log-mel spectrograms. Can be batched and of shape `(batch_size, sequence_length,
+            config.num_mel_channels)`, or un-batched and of shape `(sequence_length, config.num_mel_channels)`.
+        noise_sequence (`torch.FloatTensor`, *optional*):
+            Tensor containing a noise sequence of standard Gaussian noise. Can be batched and of shape `(batch_size,
+            sequence_length, config.model_in_channels)`, or un-batched and of shape (sequence_length,
+            config.model_in_channels)`. If not supplied, will be randomly generated.
+        padding_mask (`torch.BoolTensor`, *optional*):
+            Mask indicating which parts of each sequence are padded. Mask values are selected in `[0, 1]`:
+
+            - 1 for tokens that are **not masked**
+            - 0 for tokens that are **masked**
+
+            The mask can be batched and of shape `(batch_size, sequence_length)` or un-batched and of shape
+            `(sequence_length,)`.
+        generator (`torch.Generator`, *optional*):
+            A [torch generator](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation
+            deterministic.
+        return_dict:
+            Whether to return a [`~utils.ModelOutput`] subclass instead of a plain tuple.
+"""
+
+
+@add_start_docstrings(
+    """UnivNet GAN vocoder.""",
+    UNIVNET_START_DOCSTRING,
+)
+class UnivNetModel(PreTrainedModel):
+    config_class = UnivNetConfig
+    main_input_name = "input_features"
+
+    def __init__(self, config: UnivNetConfig):
+        super().__init__(config)
+
+        self.num_kernels = len(config.resblock_kernel_sizes)
+        self.leaky_relu_slope = config.leaky_relu_slope
+
+        self.conv_pre = nn.Conv1d(
+            config.model_in_channels,
+            config.model_hidden_channels,
+            kernel_size=7,
+            stride=1,
+            padding=3,
+            padding_mode="reflect",
+        )
+
+        # Initialize location-variable convolution ResNet Blocks.
+        num_layers = len(config.resblock_stride_sizes)
+        hop_length = 1
+        hop_lengths = []
+        for stride in config.resblock_stride_sizes:
+            hop_length = hop_length * stride
+            hop_lengths.append(hop_length)
+
+        self.resblocks = nn.ModuleList(
+            [
+                UnivNetLvcBlock(
+                    config,
+                    layer_id=i,
+                    lvc_hop_size=hop_lengths[i],
+                )
+                for i in range(num_layers)
+            ]
+        )
+
+        self.conv_post = nn.Conv1d(config.model_hidden_channels, 1, 7, padding=3, padding_mode="reflect")
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @add_start_docstrings_to_model_forward(UNIVNET_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=UnivNetModelOutput, config_class=_CONFIG_FOR_DOC)
+    def forward(
+        self,
+        input_features: torch.FloatTensor,
+        noise_sequence: Optional[torch.FloatTensor] = None,
+        padding_mask: Optional[torch.FloatTensor] = None,
+        generator: Optional[torch.Generator] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.FloatTensor], UnivNetModelOutput]:
+        r"""
+        Returns:
+
+        Example:
+
+         ```python
+         >>> from transformers import UnivNetFeatureExtractor, UnivNetModel
+         >>> from datasets import load_dataset, Audio
+
+         >>> model = UnivNetModel.from_pretrained("dg845/univnet-dev")
+         >>> feature_extractor = UnivNetFeatureExtractor.from_pretrained("dg845/univnet-dev")
+
+         >>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
+         >>> # Resample the audio to the feature extractor's sampling rate.
+         >>> ds = ds.cast_column("audio", Audio(sampling_rate=feature_extractor.sampling_rate))
+         >>> inputs = feature_extractor(
+         ...     ds[0]["audio"]["array"], sampling_rate=ds[0]["audio"]["sampling_rate"], return_tensors="pt"
+         ... )
+         >>> audio = model(**inputs).waveforms
+         >>> list(audio.shape)
+         [1, 140288]
+         ```
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        # Resolve batch sizes for noise_sequence and spectrogram
+        spectrogram_batched = input_features.dim() == 3
+        if not spectrogram_batched:
+            input_features = input_features.unsqueeze(0)
+        spectrogram_batch_size, spectrogram_length, _ = input_features.shape
+
+        if noise_sequence is not None:
+            noise_sequence_batched = noise_sequence.dim() == 3
+            if not noise_sequence_batched:
+                noise_sequence = noise_sequence.unsqueeze(0)
+        else:
+            # Randomly generate noise_sequence
+            noise_sequence_shape = (spectrogram_batch_size, spectrogram_length, self.config.model_in_channels)
+            noise_sequence = torch.randn(
+                noise_sequence_shape, generator=generator, dtype=input_features.dtype, device=input_features.device
+            )
+        noise_sequence_batch_size = noise_sequence.shape[0]
+
+        if spectrogram_batch_size > 1 and noise_sequence_batch_size == 1:
+            # Repeat noise_sequence spectrogram_batch_size times
+            noise_sequence = noise_sequence.repeat(spectrogram_batch_size, 1, 1)
+        elif noise_sequence_batch_size > 1 and spectrogram_batch_size == 1:
+            # Repeat spectrogram noise_sequence_batch_size times
+            input_features = input_features.repeat(noise_sequence_batch_size, 1, 1)
+
+        if noise_sequence_batch_size != spectrogram_batch_size:
+            raise ValueError(
+                f"The batch size of `noise_sequence` is {noise_sequence_batch_size} and the batch size of"
+                f" `input_features` is {spectrogram_batch_size}, but the two are expected to be equal."
+            )
+
+        if padding_mask is not None:
+            if padding_mask.dim() == 1:
+                padding_mask = padding_mask.unsqueeze(0)
+            padding_mask_batch_size = padding_mask.shape[0]
+            if padding_mask_batch_size != spectrogram_batch_size:
+                raise ValueError(
+                    f"The batch size of `padding_mask` is {padding_mask_batch_size} and the batch size of"
+                    f" `input_features` is {spectrogram_batch_size}, but the two are expected to be equal."
+                )
+
+        # Change shapes to have channels before sequence lengths
+        hidden_states = noise_sequence.transpose(2, 1)
+        input_features = input_features.transpose(2, 1)
+
+        hidden_states = self.conv_pre(hidden_states)
+
+        for resblock in self.resblocks:
+            hidden_states = resblock(hidden_states, input_features)
+
+        hidden_states = nn.functional.leaky_relu(hidden_states, self.leaky_relu_slope)
+        hidden_states = self.conv_post(hidden_states)
+        hidden_states = torch.tanh(hidden_states)
+
+        # Remove sequence length dimension since this collapses to 1
+        # NOTE: keep waveforms batched even if there's only one
+        waveform = hidden_states.squeeze(1)
+
+        # Get sequence lengths for UnivNetFeatureExtractor.batch_decode.
+        waveform_lengths = None
+        if padding_mask is not None:
+            # Padding is always contiguous and added on the right
+            waveform_lengths = torch.sum(padding_mask, dim=1)
+
+        if not return_dict:
+            outputs = (waveform, waveform_lengths)
+            return outputs
+
+        return UnivNetModelOutput(
+            waveforms=waveform,
+            waveform_lengths=waveform_lengths,
+        )
+
+    def _init_weights(self, module):
+        """Initialize the weights."""
+        if isinstance(module, (nn.Linear, nn.Conv1d, nn.ConvTranspose1d)):
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+            if module.bias is not None:
+                module.bias.data.zero_()
+
+    def apply_weight_norm(self):
+        nn.utils.weight_norm(self.conv_pre)
+        for layer in self.resblocks:
+            layer.apply_weight_norm()
+        nn.utils.weight_norm(self.conv_post)
+
+    def remove_weight_norm(self):
+        nn.utils.remove_weight_norm(self.conv_pre)
+        for layer in self.resblocks:
+            layer.remove_weight_norm()
+        nn.utils.remove_weight_norm(self.conv_post)

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/yoso/convert_yoso_pytorch_to_pytorch.py

@@ -0,0 +1,108 @@
+# coding=utf-8
+# Copyright 2022 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Convert YOSO checkpoints from the original repository. URL: https://github.com/mlpen/YOSO"""
+
+import argparse
+
+import torch
+
+from transformers import YosoConfig, YosoForMaskedLM
+
+
+def rename_key(orig_key):
+    if "model" in orig_key:
+        orig_key = orig_key.replace("model.", "")
+    if "norm1" in orig_key:
+        orig_key = orig_key.replace("norm1", "attention.output.LayerNorm")
+    if "norm2" in orig_key:
+        orig_key = orig_key.replace("norm2", "output.LayerNorm")
+    if "norm" in orig_key:
+        orig_key = orig_key.replace("norm", "LayerNorm")
+    if "transformer" in orig_key:
+        layer_num = orig_key.split(".")[0].split("_")[-1]
+        orig_key = orig_key.replace(f"transformer_{layer_num}", f"encoder.layer.{layer_num}")
+    if "mha.attn" in orig_key:
+        orig_key = orig_key.replace("mha.attn", "attention.self")
+    if "mha" in orig_key:
+        orig_key = orig_key.replace("mha", "attention")
+    if "W_q" in orig_key:
+        orig_key = orig_key.replace("W_q", "self.query")
+    if "W_k" in orig_key:
+        orig_key = orig_key.replace("W_k", "self.key")
+    if "W_v" in orig_key:
+        orig_key = orig_key.replace("W_v", "self.value")
+    if "ff1" in orig_key:
+        orig_key = orig_key.replace("ff1", "intermediate.dense")
+    if "ff2" in orig_key:
+        orig_key = orig_key.replace("ff2", "output.dense")
+    if "ff" in orig_key:
+        orig_key = orig_key.replace("ff", "output.dense")
+    if "mlm_class" in orig_key:
+        orig_key = orig_key.replace("mlm.mlm_class", "cls.predictions.decoder")
+    if "mlm" in orig_key:
+        orig_key = orig_key.replace("mlm", "cls.predictions.transform")
+    if "cls" not in orig_key:
+        orig_key = "yoso." + orig_key
+
+    return orig_key
+
+
+def convert_checkpoint_helper(max_position_embeddings, orig_state_dict):
+    for key in orig_state_dict.copy().keys():
+        val = orig_state_dict.pop(key)
+
+        if ("pooler" in key) or ("sen_class" in key):
+            continue
+        else:
+            orig_state_dict[rename_key(key)] = val
+
+    orig_state_dict["cls.predictions.bias"] = orig_state_dict["cls.predictions.decoder.bias"]
+    orig_state_dict["yoso.embeddings.position_ids"] = torch.arange(max_position_embeddings).expand((1, -1)) + 2
+
+    return orig_state_dict
+
+
+def convert_yoso_checkpoint(checkpoint_path, yoso_config_file, pytorch_dump_path):
+    orig_state_dict = torch.load(checkpoint_path, map_location="cpu")["model_state_dict"]
+    config = YosoConfig.from_json_file(yoso_config_file)
+    model = YosoForMaskedLM(config)
+
+    new_state_dict = convert_checkpoint_helper(config.max_position_embeddings, orig_state_dict)
+
+    print(model.load_state_dict(new_state_dict))
+    model.eval()
+    model.save_pretrained(pytorch_dump_path)
+
+    print(f"Checkpoint successfuly converted. Model saved at {pytorch_dump_path}")
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    # Required parameters
+    parser.add_argument(
+        "--pytorch_model_path", default=None, type=str, required=True, help="Path to YOSO pytorch checkpoint."
+    )
+    parser.add_argument(
+        "--config_file",
+        default=None,
+        type=str,
+        required=True,
+        help="The json file for YOSO model config.",
+    )
+    parser.add_argument(
+        "--pytorch_dump_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
+    )
+    args = parser.parse_args()
+    convert_yoso_checkpoint(args.pytorch_model_path, args.config_file, args.pytorch_dump_path)

evalkit_cambrian/lib/python3.10/site-packages/transformers/models/yoso/modeling_yoso.py

@@ -0,0 +1,1304 @@
+# coding=utf-8
+# Copyright 2022 University of Wisconsin-Madison and The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch YOSO model."""
+
+
+import math
+from pathlib import Path
+from typing import Optional, Tuple, Union
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+
+from ...activations import ACT2FN
+from ...modeling_outputs import (
+    BaseModelOutputWithCrossAttentions,
+    MaskedLMOutput,
+    MultipleChoiceModelOutput,
+    QuestionAnsweringModelOutput,
+    SequenceClassifierOutput,
+    TokenClassifierOutput,
+)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
+from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
+from .configuration_yoso import YosoConfig
+
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "uw-madison/yoso-4096"
+_CONFIG_FOR_DOC = "YosoConfig"
+
+YOSO_PRETRAINED_MODEL_ARCHIVE_LIST = [
+    "uw-madison/yoso-4096",
+    # See all YOSO models at https://huggingface.co/models?filter=yoso
+]
+
+
+def load_cuda_kernels():
+    global lsh_cumulation
+    try:
+        from torch.utils.cpp_extension import load
+
+        def append_root(files):
+            src_folder = Path(__file__).resolve().parent.parent.parent / "kernels" / "yoso"
+            return [src_folder / file for file in files]
+
+        src_files = append_root(
+            ["fast_lsh_cumulation_torch.cpp", "fast_lsh_cumulation.cu", "fast_lsh_cumulation_cuda.cu"]
+        )
+
+        load("fast_lsh_cumulation", src_files, verbose=True)
+
+        import fast_lsh_cumulation as lsh_cumulation
+
+        return True
+    except Exception:
+        lsh_cumulation = None
+        return False
+
+
+def to_contiguous(input_tensors):
+    if isinstance(input_tensors, list):
+        out = []
+        for tensor in input_tensors:
+            if not tensor.is_contiguous():
+                tensor = tensor.contiguous()
+            out.append(tensor)
+        return out
+    else:
+        if not input_tensors.is_contiguous():
+            input_tensors = input_tensors.contiguous()
+        return input_tensors
+
+
+def normalize(input_tensors):
+    if isinstance(input_tensors, list):
+        out = []
+        for tensor in input_tensors:
+            out.append(nn.functional.normalize(tensor, p=2, dim=-1))
+        return out
+    else:
+        return nn.functional.normalize(input_tensors, p=2, dim=-1)
+
+
+def hashing(query, key, num_hash, hash_len):
+    if len(query.size()) != 3:
+        raise ValueError("Query has incorrect size.")
+    if len(key.size()) != 3:
+        raise ValueError("Key has incorrect size.")
+
+    rmat = torch.randn(query.size(0), query.size(2), num_hash * hash_len, device=query.device)
+    raise_pow = 2 ** torch.arange(hash_len, device=query.device)
+
+    query_projection = torch.matmul(query, rmat).reshape(query.size(0), query.size(1), num_hash, hash_len)
+    key_projection = torch.matmul(key, rmat).reshape(key.size(0), key.size(1), num_hash, hash_len)
+    query_binary = (query_projection > 0).int()
+    key_binary = (key_projection > 0).int()
+    query_hash = torch.sum(query_binary * raise_pow, dim=-1)
+    query_hash = torch.sum(key_binary * raise_pow, dim=-1)
+
+    return query_hash.int(), query_hash.int()
+
+
+class YosoCumulation(torch.autograd.Function):
+    @staticmethod
+    def forward(ctx, query_mask, key_mask, query, key, value, config):
+        hash_code_len = config["hash_code_len"]
+
+        expectation = (1 - torch.acos(torch.matmul(query, key.transpose(-1, -2))) / math.pi) ** hash_code_len
+        expectation = expectation * query_mask[:, :, None] * key_mask[:, None, :]
+        cumulation_value = torch.matmul(expectation, value)
+
+        ctx.save_for_backward(query_mask, key_mask, expectation, query, key, value)
+        ctx.config = config
+
+        return cumulation_value
+
+    @staticmethod
+    def backward(ctx, grad):
+        grad = to_contiguous(grad)
+
+        query_mask, key_mask, expectation, query, key, value = ctx.saved_tensors
+        config = ctx.config
+
+        hash_code_len = config["hash_code_len"]
+
+        weighted_exp = torch.matmul(grad, value.transpose(-1, -2)) * expectation
+        grad_query = torch.matmul(weighted_exp, (hash_code_len / 2) * key)
+        grad_key = torch.matmul(weighted_exp.transpose(-1, -2), (hash_code_len / 2) * query)
+        grad_value = torch.matmul(expectation.transpose(-1, -2), grad)
+
+        return None, None, grad_query, grad_key, grad_value, None
+
+
+class YosoLSHCumulation(torch.autograd.Function):
+    @staticmethod
+    def forward(ctx, query_mask, key_mask, query, key, value, config):
+        if query_mask.size(0) != key_mask.size(0):
+            raise ValueError("Query mask and Key mask differ in sizes in dimension 0")
+        if query_mask.size(0) != query.size(0):
+            raise ValueError("Query mask and Query differ in sizes in dimension 0")
+        if query_mask.size(0) != key.size(0):
+            raise ValueError("Query mask and Key differ in sizes in dimension 0")
+        if query_mask.size(0) != value.size(0):
+            raise ValueError("Query mask and Value mask differ in sizes in dimension 0")
+        if key.size(1) != value.size(1):
+            raise ValueError("Key and Value differ in sizes in dimension 1")
+        if query.size(2) != key.size(2):
+            raise ValueError("Query and Key differ in sizes in dimension 2")
+
+        query_mask, key_mask, query, key, value = to_contiguous([query_mask, key_mask, query, key, value])
+
+        use_cuda = query_mask.is_cuda
+        num_hash = config["num_hash"]
+        hash_code_len = config["hash_code_len"]
+        hashtable_capacity = int(2**hash_code_len)
+
+        if config["use_fast_hash"]:
+            query_hash_code, key_hash_code = lsh_cumulation.fast_hash(
+                query_mask, query, key_mask, key, num_hash, hash_code_len, use_cuda, 1
+            )
+        else:
+            query_hash_code, key_hash_code = hashing(query, key, num_hash, hash_code_len)
+
+        cumulation_value = lsh_cumulation.lsh_cumulation(
+            query_mask, query_hash_code, key_mask, key_hash_code, value, hashtable_capacity, use_cuda, 1
+        )
+
+        ctx.save_for_backward(query_mask, key_mask, query_hash_code, key_hash_code, query, key, value)
+        ctx.config = config
+
+        return cumulation_value
+
+    @staticmethod
+    def backward(ctx, grad):
+        grad = to_contiguous(grad)
+
+        query_mask, key_mask, query_hash_code, key_hash_code, query, key, value = ctx.saved_tensors
+        config = ctx.config
+
+        use_cuda = grad.is_cuda
+        hash_code_len = config["hash_code_len"]
+        hashtable_capacity = int(2**hash_code_len)
+
+        if config["lsh_backward"]:
+            grad_value = lsh_cumulation.lsh_cumulation(
+                key_mask, key_hash_code, query_mask, query_hash_code, grad, hashtable_capacity, use_cuda, 1
+            )
+            grad_query = lsh_cumulation.lsh_weighted_cumulation(
+                query_mask,
+                query_hash_code,
+                grad,
+                key_mask,
+                key_hash_code,
+                value,
+                (hash_code_len / 2) * key,
+                hashtable_capacity,
+                use_cuda,
+                4,
+            )
+            grad_key = lsh_cumulation.lsh_weighted_cumulation(
+                key_mask,
+                key_hash_code,
+                value,
+                query_mask,
+                query_hash_code,
+                grad,
+                (hash_code_len / 2) * query,
+                hashtable_capacity,
+                use_cuda,
+                4,
+            )
+        else:
+            expectation = (1 - torch.acos(torch.matmul(query, key.transpose(-1, -2))) / math.pi) ** hash_code_len
+            expectation = expectation * query_mask[:, :, None] * key_mask[:, None, :]
+            weighted_exp = torch.matmul(grad, value.transpose(-1, -2)) * expectation
+            grad_query = torch.matmul(weighted_exp, (hash_code_len / 2) * key)
+            grad_key = torch.matmul(weighted_exp.transpose(-1, -2), (hash_code_len / 2) * query)
+            grad_value = torch.matmul(expectation.transpose(-1, -2), grad)
+
+        return None, None, grad_query, grad_key, grad_value, None
+
+
+# Copied from transformers.models.nystromformer.modeling_nystromformer.NystromformerEmbeddings
+class YosoEmbeddings(nn.Module):
+    """Construct the embeddings from word, position and token_type embeddings."""
+
+    def __init__(self, config):
+        super().__init__()
+        self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
+        self.position_embeddings = nn.Embedding(config.max_position_embeddings + 2, config.hidden_size)
+        self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)
+
+        # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
+        # any TensorFlow checkpoint file
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
+        self.register_buffer(
+            "position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)) + 2, persistent=False
+        )
+        self.position_embedding_type = getattr(config, "position_embedding_type", "absolute")
+        self.register_buffer(
+            "token_type_ids",
+            torch.zeros(self.position_ids.size(), dtype=torch.long, device=self.position_ids.device),
+            persistent=False,
+        )
+
+    def forward(self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None):
+        if input_ids is not None:
+            input_shape = input_ids.size()
+        else:
+            input_shape = inputs_embeds.size()[:-1]
+
+        seq_length = input_shape[1]
+
+        if position_ids is None:
+            position_ids = self.position_ids[:, :seq_length]
+
+        # Setting the token_type_ids to the registered buffer in constructor where it is all zeros, which usually occurs
+        # when its auto-generated, registered buffer helps users when tracing the model without passing token_type_ids, solves
+        # issue #5664
+        if token_type_ids is None:
+            if hasattr(self, "token_type_ids"):
+                buffered_token_type_ids = self.token_type_ids[:, :seq_length]
+                buffered_token_type_ids_expanded = buffered_token_type_ids.expand(input_shape[0], seq_length)
+                token_type_ids = buffered_token_type_ids_expanded
+            else:
+                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=self.position_ids.device)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.word_embeddings(input_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+        embeddings = inputs_embeds + token_type_embeddings
+        if self.position_embedding_type == "absolute":
+            position_embeddings = self.position_embeddings(position_ids)
+            embeddings += position_embeddings
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings)
+        return embeddings
+
+
+class YosoSelfAttention(nn.Module):
+    def __init__(self, config, position_embedding_type=None):
+        super().__init__()
+        if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
+            raise ValueError(
+                f"The hidden size ({config.hidden_size}) is not a multiple of the number of attention "
+                f"heads ({config.num_attention_heads})"
+            )
+
+        self.num_attention_heads = config.num_attention_heads
+        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+        self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+        self.query = nn.Linear(config.hidden_size, self.all_head_size)
+        self.key = nn.Linear(config.hidden_size, self.all_head_size)
+        self.value = nn.Linear(config.hidden_size, self.all_head_size)
+
+        self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
+        self.position_embedding_type = (
+            position_embedding_type if position_embedding_type is not None else config.position_embedding_type
+        )
+
+        self.use_expectation = config.use_expectation
+        self.hash_code_len = config.hash_code_len
+        self.use_conv = config.conv_window is not None
+        self.use_fast_hash = config.use_fast_hash
+        self.num_hash = config.num_hash
+        self.lsh_backward = config.lsh_backward
+
+        self.lsh_config = {
+            "hash_code_len": self.hash_code_len,
+            "use_fast_hash": self.use_fast_hash,
+            "num_hash": self.num_hash,
+            "lsh_backward": self.lsh_backward,
+        }
+
+        if config.conv_window is not None:
+            self.conv = nn.Conv2d(
+                in_channels=config.num_attention_heads,
+                out_channels=config.num_attention_heads,
+                kernel_size=(config.conv_window, 1),
+                padding=(config.conv_window // 2, 0),
+                bias=False,
+                groups=config.num_attention_heads,
+            )
+
+    def transpose_for_scores(self, layer):
+        new_layer_shape = layer.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
+        layer = layer.view(*new_layer_shape)
+        return layer.permute(0, 2, 1, 3)
+
+    def forward(self, hidden_states, attention_mask=None, output_attentions=False):
+        mixed_query_layer = self.query(hidden_states)
+
+        key_layer = self.transpose_for_scores(self.key(hidden_states))
+        value_layer = self.transpose_for_scores(self.value(hidden_states))
+        query_layer = self.transpose_for_scores(mixed_query_layer)
+
+        if self.use_conv:
+            conv_value_layer = self.conv(value_layer * attention_mask[:, None, :, None])
+
+        batch_size, num_heads, seq_len, head_dim = query_layer.size()
+
+        query_layer = query_layer.reshape(batch_size * num_heads, seq_len, head_dim)
+        key_layer = key_layer.reshape(batch_size * num_heads, seq_len, head_dim)
+        value_layer = value_layer.reshape(batch_size * num_heads, seq_len, head_dim)
+
+        # revert changes made by get_extended_attention_mask
+        attention_mask = 1.0 + attention_mask / 10000.0
+        attention_mask = (
+            attention_mask.squeeze().repeat(1, num_heads, 1).reshape(batch_size * num_heads, seq_len).int()
+        )
+
+        # The CUDA kernels are most efficient with inputs whose size is a multiple of a GPU's warp size (32). Inputs
+        # smaller than this are padded with zeros.
+        gpu_warp_size = 32
+
+        if (not self.use_expectation) and head_dim < gpu_warp_size:
+            pad_size = batch_size * num_heads, seq_len, gpu_warp_size - head_dim
+
+            query_layer = torch.cat(
+                [
+                    query_layer,
+                    torch.zeros(pad_size, device=query_layer.device),
+                ],
+                dim=-1,
+            )
+            key_layer = torch.cat(
+                [
+                    key_layer,
+                    torch.zeros(pad_size, device=key_layer.device),
+                ],
+                dim=-1,
+            )
+            value_layer = torch.cat(
+                [
+                    value_layer,
+                    torch.zeros(pad_size, device=value_layer.device),
+                ],
+                dim=-1,
+            )
+
+        if self.use_expectation or self.training:
+            query_layer, key_layer = normalize([query_layer, key_layer])
+
+        if self.use_expectation:
+            context_layer = YosoCumulation.apply(
+                attention_mask, attention_mask, query_layer, key_layer, value_layer, self.lsh_config
+            )
+        else:
+            context_layer = YosoLSHCumulation.apply(
+                attention_mask, attention_mask, query_layer, key_layer, value_layer, self.lsh_config
+            )
+
+        if (not self.use_expectation) and head_dim < gpu_warp_size:
+            context_layer = context_layer[:, :, :head_dim]
+
+        context_layer = normalize(context_layer)
+
+        context_layer = context_layer.reshape(batch_size, num_heads, seq_len, head_dim)
+
+        if self.use_conv:
+            context_layer += conv_value_layer
+
+        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+        context_layer = context_layer.view(*new_context_layer_shape)
+
+        outputs = (context_layer, context_layer) if output_attentions else (context_layer,)
+
+        return outputs
+
+
+# Copied from transformers.models.bert.modeling_bert.BertSelfOutput
+class YosoSelfOutput(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+    def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states + input_tensor)
+        return hidden_states
+
+
+class YosoAttention(nn.Module):
+    def __init__(self, config, position_embedding_type=None):
+        super().__init__()
+        self.self = YosoSelfAttention(config, position_embedding_type=position_embedding_type)
+        self.output = YosoSelfOutput(config)
+        self.pruned_heads = set()
+
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        heads, index = find_pruneable_heads_and_indices(
+            heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
+        )
+
+        # Prune linear layers
+        self.self.query = prune_linear_layer(self.self.query, index)
+        self.self.key = prune_linear_layer(self.self.key, index)
+        self.self.value = prune_linear_layer(self.self.value, index)
+        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+        # Update hyper params and store pruned heads
+        self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
+        self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
+
+    def forward(self, hidden_states, attention_mask=None, output_attentions=False):
+        self_outputs = self.self(hidden_states, attention_mask, output_attentions)
+        attention_output = self.output(self_outputs[0], hidden_states)
+        outputs = (attention_output,) + self_outputs[1:]  # add attentions if we output them
+        return outputs
+
+
+# Copied from transformers.models.bert.modeling_bert.BertIntermediate
+class YosoIntermediate(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.intermediate_size)
+        if isinstance(config.hidden_act, str):
+            self.intermediate_act_fn = ACT2FN[config.hidden_act]
+        else:
+            self.intermediate_act_fn = config.hidden_act
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.intermediate_act_fn(hidden_states)
+        return hidden_states
+
+
+# Copied from transformers.models.bert.modeling_bert.BertOutput
+class YosoOutput(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+    def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states + input_tensor)
+        return hidden_states
+
+
+class YosoLayer(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.chunk_size_feed_forward = config.chunk_size_feed_forward
+        self.seq_len_dim = 1
+        self.attention = YosoAttention(config)
+        self.add_cross_attention = config.add_cross_attention
+        self.intermediate = YosoIntermediate(config)
+        self.output = YosoOutput(config)
+
+    def forward(self, hidden_states, attention_mask=None, output_attentions=False):
+        self_attention_outputs = self.attention(hidden_states, attention_mask, output_attentions=output_attentions)
+        attention_output = self_attention_outputs[0]
+
+        outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights
+
+        layer_output = apply_chunking_to_forward(
+            self.feed_forward_chunk, self.chunk_size_feed_forward, self.seq_len_dim, attention_output
+        )
+        outputs = (layer_output,) + outputs
+
+        return outputs
+
+    def feed_forward_chunk(self, attention_output):
+        intermediate_output = self.intermediate(attention_output)
+        layer_output = self.output(intermediate_output, attention_output)
+        return layer_output
+
+
+class YosoEncoder(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.layer = nn.ModuleList([YosoLayer(config) for _ in range(config.num_hidden_layers)])
+        self.gradient_checkpointing = False
+
+    def forward(
+        self,
+        hidden_states,
+        attention_mask=None,
+        head_mask=None,
+        output_attentions=False,
+        output_hidden_states=False,
+        return_dict=True,
+    ):
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attentions = () if output_attentions else None
+
+        for i, layer_module in enumerate(self.layer):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+                layer_outputs = self._gradient_checkpointing_func(
+                    layer_module.__call__,
+                    hidden_states,
+                    attention_mask,
+                    output_attentions,
+                )
+            else:
+                layer_outputs = layer_module(hidden_states, attention_mask, output_attentions)
+
+            hidden_states = layer_outputs[0]
+            if output_attentions:
+                all_self_attentions = all_self_attentions + (layer_outputs[1],)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, all_hidden_states, all_self_attentions] if v is not None)
+        return BaseModelOutputWithCrossAttentions(
+            last_hidden_state=hidden_states,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+        )
+
+
+# Copied from transformers.models.bert.modeling_bert.BertPredictionHeadTransform
+class YosoPredictionHeadTransform(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        if isinstance(config.hidden_act, str):
+            self.transform_act_fn = ACT2FN[config.hidden_act]
+        else:
+            self.transform_act_fn = config.hidden_act
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.transform_act_fn(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states)
+        return hidden_states
+
+
+# Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead with Bert->Yoso
+class YosoLMPredictionHead(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.transform = YosoPredictionHeadTransform(config)
+
+        # The output weights are the same as the input embeddings, but there is
+        # an output-only bias for each token.
+        self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        self.bias = nn.Parameter(torch.zeros(config.vocab_size))
+
+        # Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
+        self.decoder.bias = self.bias
+
+    def forward(self, hidden_states):
+        hidden_states = self.transform(hidden_states)
+        hidden_states = self.decoder(hidden_states)
+        return hidden_states
+
+
+# Copied from transformers.models.bert.modeling_bert.BertOnlyMLMHead with Bert->Yoso
+class YosoOnlyMLMHead(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.predictions = YosoLMPredictionHead(config)
+
+    def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
+        prediction_scores = self.predictions(sequence_output)
+        return prediction_scores
+
+
+class YosoPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+    models.
+    """
+
+    config_class = YosoConfig
+    base_model_prefix = "yoso"
+    supports_gradient_checkpointing = True
+
+    def _init_weights(self, module):
+        """Initialize the weights"""
+        if isinstance(module, nn.Linear):
+            # Slightly different from the TF version which uses truncated_normal for initialization
+            # cf https://github.com/pytorch/pytorch/pull/5617
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+        elif isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+
+
+YOSO_START_DOCSTRING = r"""
+    This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use
+    it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and
+    behavior.
+
+    Parameters:
+        config ([`YosoConfig`]): Model configuration class with all the parameters of the model.
+            Initializing with a config file does not load the weights associated with the model, only the
+            configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+YOSO_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (`torch.LongTensor` of shape `({0})`):
+            Indices of input sequence tokens in the vocabulary.
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            [What are input IDs?](../glossary#input-ids)
+        attention_mask (`torch.FloatTensor` of shape `({0})`, *optional*):
+            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            [What are attention masks?](../glossary#attention-mask)
+        token_type_ids (`torch.LongTensor` of shape `({0})`, *optional*):
+            Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0,
+            1]`:
+
+            - 0 corresponds to a *sentence A* token,
+            - 1 corresponds to a *sentence B* token.
+
+            [What are token type IDs?](../glossary#token-type-ids)
+        position_ids (`torch.LongTensor` of shape `({0})`, *optional*):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+            config.max_position_embeddings - 1]`.
+
+            [What are position IDs?](../glossary#position-ids)
+        head_mask (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+            Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
+
+            - 1 indicates the head is **not masked**,
+            - 0 indicates the head is **masked**.
+
+        inputs_embeds (`torch.FloatTensor` of shape `({0}, hidden_size)`, *optional*):
+            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+            is useful if you want more control over how to convert *input_ids* indices into associated vectors than the
+            model's internal embedding lookup matrix.
+        output_attentions (`bool`, *optional*):
+            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+            tensors for more detail.
+        output_hidden_states (`bool`, *optional*):
+            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+            more detail.
+        return_dict (`bool`, *optional*):
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+
+@add_start_docstrings(
+    "The bare YOSO Model transformer outputting raw hidden-states without any specific head on top.",
+    YOSO_START_DOCSTRING,
+)
+class YosoModel(YosoPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.config = config
+
+        self.embeddings = YosoEmbeddings(config)
+        self.encoder = YosoEncoder(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
+    def _prune_heads(self, heads_to_prune):
+        """
+        Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
+        class PreTrainedModel
+        """
+        for layer, heads in heads_to_prune.items():
+            self.encoder.layer[layer].attention.prune_heads(heads)
+
+    @add_start_docstrings_to_model_forward(YOSO_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=BaseModelOutputWithCrossAttentions,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, BaseModelOutputWithCrossAttentions]:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        batch_size, seq_length = input_shape
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if attention_mask is None:
+            attention_mask = torch.ones(((batch_size, seq_length)), device=device)
+
+        if token_type_ids is None:
+            if hasattr(self.embeddings, "token_type_ids"):
+                buffered_token_type_ids = self.embeddings.token_type_ids[:, :seq_length]
+                buffered_token_type_ids_expanded = buffered_token_type_ids.expand(batch_size, seq_length)
+                token_type_ids = buffered_token_type_ids_expanded
+            else:
+                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
+
+        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+        # ourselves in which case we just need to make it broadcastable to all heads.
+        extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(attention_mask, input_shape)
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+        embedding_output = self.embeddings(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            token_type_ids=token_type_ids,
+            inputs_embeds=inputs_embeds,
+        )
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_mask=extended_attention_mask,
+            head_mask=head_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        sequence_output = encoder_outputs[0]
+
+        if not return_dict:
+            return (sequence_output,) + encoder_outputs[1:]
+
+        return BaseModelOutputWithCrossAttentions(
+            last_hidden_state=sequence_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+            cross_attentions=encoder_outputs.cross_attentions,
+        )
+
+
+@add_start_docstrings("""YOSO Model with a `language modeling` head on top.""", YOSO_START_DOCSTRING)
+class YosoForMaskedLM(YosoPreTrainedModel):
+    _tied_weights_keys = ["cls.predictions.decoder.weight", "cls.predictions.decoder.bias"]
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.yoso = YosoModel(config)
+        self.cls = YosoOnlyMLMHead(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_output_embeddings(self):
+        return self.cls.predictions.decoder
+
+    def set_output_embeddings(self, new_embeddings):
+        self.cls.predictions.decoder = new_embeddings
+
+    @add_start_docstrings_to_model_forward(YOSO_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=MaskedLMOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, MaskedLMOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
+            config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
+            loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.yoso(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+        prediction_scores = self.cls(sequence_output)
+
+        masked_lm_loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()  # -100 index = padding token
+            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
+
+        if not return_dict:
+            output = (prediction_scores,) + outputs[1:]
+            return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output
+
+        return MaskedLMOutput(
+            loss=masked_lm_loss,
+            logits=prediction_scores,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+class YosoClassificationHead(nn.Module):
+    """Head for sentence-level classification tasks."""
+
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.out_proj = nn.Linear(config.hidden_size, config.num_labels)
+
+        self.config = config
+
+    def forward(self, features, **kwargs):
+        x = features[:, 0, :]  # take <s> token (equiv. to [CLS])
+        x = self.dropout(x)
+        x = self.dense(x)
+        x = ACT2FN[self.config.hidden_act](x)
+        x = self.dropout(x)
+        x = self.out_proj(x)
+        return x
+
+
+@add_start_docstrings(
+    """YOSO Model transformer with a sequence classification/regression head on top (a linear layer on top of
+    the pooled output) e.g. for GLUE tasks.""",
+    YOSO_START_DOCSTRING,
+)
+class YosoForSequenceClassification(YosoPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.yoso = YosoModel(config)
+        self.classifier = YosoClassificationHead(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @add_start_docstrings_to_model_forward(YOSO_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=SequenceClassifierOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, SequenceClassifierOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.yoso(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+        logits = self.classifier(sequence_output)
+
+        loss = None
+        if labels is not None:
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(logits, labels)
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@add_start_docstrings(
+    """YOSO Model with a multiple choice classification head on top (a linear layer on top of
+    the pooled output and a softmax) e.g. for RocStories/SWAG tasks.""",
+    YOSO_START_DOCSTRING,
+)
+class YosoForMultipleChoice(YosoPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.yoso = YosoModel(config)
+        self.pre_classifier = nn.Linear(config.hidden_size, config.hidden_size)
+        self.classifier = nn.Linear(config.hidden_size, 1)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @add_start_docstrings_to_model_forward(YOSO_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
+    @add_code_sample_docstrings(
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=MultipleChoiceModelOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, MultipleChoiceModelOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
+            num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See
+            `input_ids` above)
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+        num_choices = input_ids.shape[1] if input_ids is not None else inputs_embeds.shape[1]
+
+        input_ids = input_ids.view(-1, input_ids.size(-1)) if input_ids is not None else None
+        attention_mask = attention_mask.view(-1, attention_mask.size(-1)) if attention_mask is not None else None
+        token_type_ids = token_type_ids.view(-1, token_type_ids.size(-1)) if token_type_ids is not None else None
+        position_ids = position_ids.view(-1, position_ids.size(-1)) if position_ids is not None else None
+        inputs_embeds = (
+            inputs_embeds.view(-1, inputs_embeds.size(-2), inputs_embeds.size(-1))
+            if inputs_embeds is not None
+            else None
+        )
+
+        outputs = self.yoso(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_state = outputs[0]  # (bs * num_choices, seq_len, dim)
+        pooled_output = hidden_state[:, 0]  # (bs * num_choices, dim)
+        pooled_output = self.pre_classifier(pooled_output)  # (bs * num_choices, dim)
+        pooled_output = nn.ReLU()(pooled_output)  # (bs * num_choices, dim)
+        logits = self.classifier(pooled_output)
+
+        reshaped_logits = logits.view(-1, num_choices)
+
+        loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()
+            loss = loss_fct(reshaped_logits, labels)
+
+        if not return_dict:
+            output = (reshaped_logits,) + outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return MultipleChoiceModelOutput(
+            loss=loss,
+            logits=reshaped_logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@add_start_docstrings(
+    """YOSO Model with a token classification head on top (a linear layer on top of
+    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks.""",
+    YOSO_START_DOCSTRING,
+)
+class YosoForTokenClassification(YosoPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.yoso = YosoModel(config)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @add_start_docstrings_to_model_forward(YOSO_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=TokenClassifierOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, TokenClassifierOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.yoso(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+
+        sequence_output = self.dropout(sequence_output)
+        logits = self.classifier(sequence_output)
+
+        loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()
+            # Only keep active parts of the loss
+            if attention_mask is not None:
+                active_loss = attention_mask.view(-1) == 1
+                active_logits = logits.view(-1, self.num_labels)
+                active_labels = torch.where(
+                    active_loss, labels.view(-1), torch.tensor(loss_fct.ignore_index).type_as(labels)
+                )
+                loss = loss_fct(active_logits, active_labels)
+            else:
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return TokenClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@add_start_docstrings(
+    """YOSO Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear
+    layers on top of the hidden-states output to compute `span start logits` and `span end logits`).""",
+    YOSO_START_DOCSTRING,
+)
+class YosoForQuestionAnswering(YosoPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+
+        config.num_labels = 2
+        self.num_labels = config.num_labels
+
+        self.yoso = YosoModel(config)
+        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @add_start_docstrings_to_model_forward(YOSO_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=QuestionAnsweringModelOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        start_positions: Optional[torch.Tensor] = None,
+        end_positions: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, QuestionAnsweringModelOutput]:
+        r"""
+        start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for position (index) of the start of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+            are not taken into account for computing the loss.
+        end_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for position (index) of the end of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+            are not taken into account for computing the loss.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.yoso(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+
+        logits = self.qa_outputs(sequence_output)
+        start_logits, end_logits = logits.split(1, dim=-1)
+        start_logits = start_logits.squeeze(-1)
+        end_logits = end_logits.squeeze(-1)
+
+        total_loss = None
+        if start_positions is not None and end_positions is not None:
+            # If we are on multi-GPU, split add a dimension
+            if len(start_positions.size()) > 1:
+                start_positions = start_positions.squeeze(-1)
+            if len(end_positions.size()) > 1:
+                end_positions = end_positions.squeeze(-1)
+            # sometimes the start/end positions are outside our model inputs, we ignore these terms
+            ignored_index = start_logits.size(1)
+            start_positions = start_positions.clamp(0, ignored_index)
+            end_positions = end_positions.clamp(0, ignored_index)
+
+            loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
+            start_loss = loss_fct(start_logits, start_positions)
+            end_loss = loss_fct(end_logits, end_positions)
+            total_loss = (start_loss + end_loss) / 2
+
+        if not return_dict:
+            output = (start_logits, end_logits) + outputs[1:]
+            return ((total_loss,) + output) if total_loss is not None else output
+
+        return QuestionAnsweringModelOutput(
+            loss=total_loss,
+            start_logits=start_logits,
+            end_logits=end_logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )

infer_4_47_1/lib/python3.10/site-packages/fontTools/feaLib/__main__.py

@@ -0,0 +1,78 @@
+from fontTools.ttLib import TTFont
+from fontTools.feaLib.builder import addOpenTypeFeatures, Builder
+from fontTools.feaLib.error import FeatureLibError
+from fontTools import configLogger
+from fontTools.misc.cliTools import makeOutputFileName
+import sys
+import argparse
+import logging
+
+
+log = logging.getLogger("fontTools.feaLib")
+
+
+def main(args=None):
+    """Add features from a feature file (.fea) into an OTF font"""
+    parser = argparse.ArgumentParser(
+        description="Use fontTools to compile OpenType feature files (*.fea)."
+    )
+    parser.add_argument(
+        "input_fea", metavar="FEATURES", help="Path to the feature file"
+    )
+    parser.add_argument(
+        "input_font", metavar="INPUT_FONT", help="Path to the input font"
+    )
+    parser.add_argument(
+        "-o",
+        "--output",
+        dest="output_font",
+        metavar="OUTPUT_FONT",
+        help="Path to the output font.",
+    )
+    parser.add_argument(
+        "-t",
+        "--tables",
+        metavar="TABLE_TAG",
+        choices=Builder.supportedTables,
+        nargs="+",
+        help="Specify the table(s) to be built.",
+    )
+    parser.add_argument(
+        "-d",
+        "--debug",
+        action="store_true",
+        help="Add source-level debugging information to font.",
+    )
+    parser.add_argument(
+        "-v",
+        "--verbose",
+        help="Increase the logger verbosity. Multiple -v " "options are allowed.",
+        action="count",
+        default=0,
+    )
+    parser.add_argument(
+        "--traceback", help="show traceback for exceptions.", action="store_true"
+    )
+    options = parser.parse_args(args)
+
+    levels = ["WARNING", "INFO", "DEBUG"]
+    configLogger(level=levels[min(len(levels) - 1, options.verbose)])
+
+    output_font = options.output_font or makeOutputFileName(options.input_font)
+    log.info("Compiling features to '%s'" % (output_font))
+
+    font = TTFont(options.input_font)
+    try:
+        addOpenTypeFeatures(
+            font, options.input_fea, tables=options.tables, debug=options.debug
+        )
+    except FeatureLibError as e:
+        if options.traceback:
+            raise
+        log.error(e)
+        sys.exit(1)
+    font.save(output_font)
+
+
+if __name__ == "__main__":
+    sys.exit(main())

infer_4_47_1/lib/python3.10/site-packages/fontTools/feaLib/error.py

@@ -0,0 +1,22 @@
+class FeatureLibError(Exception):
+    def __init__(self, message, location):
+        Exception.__init__(self, message)
+        self.location = location
+
+    def __str__(self):
+        message = Exception.__str__(self)
+        if self.location:
+            return f"{self.location}: {message}"
+        else:
+            return message
+
+
+class IncludedFeaNotFound(FeatureLibError):
+    def __str__(self):
+        assert self.location is not None
+
+        message = (
+            "The following feature file should be included but cannot be found: "
+            f"{Exception.__str__(self)}"
+        )
+        return f"{self.location}: {message}"

infer_4_47_1/lib/python3.10/site-packages/fontTools/feaLib/lookupDebugInfo.py

@@ -0,0 +1,12 @@
+from typing import NamedTuple
+
+LOOKUP_DEBUG_INFO_KEY = "com.github.fonttools.feaLib"
+LOOKUP_DEBUG_ENV_VAR = "FONTTOOLS_LOOKUP_DEBUGGING"
+
+
+class LookupDebugInfo(NamedTuple):
+    """Information about where a lookup came from, to be embedded in a font"""
+
+    location: str
+    name: str
+    feature: list

infer_4_47_1/lib/python3.10/site-packages/fontTools/feaLib/variableScalar.py

@@ -0,0 +1,113 @@
+from fontTools.varLib.models import VariationModel, normalizeValue, piecewiseLinearMap
+
+
+def Location(loc):
+    return tuple(sorted(loc.items()))
+
+
+class VariableScalar:
+    """A scalar with different values at different points in the designspace."""
+
+    def __init__(self, location_value={}):
+        self.values = {}
+        self.axes = {}
+        for location, value in location_value.items():
+            self.add_value(location, value)
+
+    def __repr__(self):
+        items = []
+        for location, value in self.values.items():
+            loc = ",".join(["%s=%i" % (ax, loc) for ax, loc in location])
+            items.append("%s:%i" % (loc, value))
+        return "(" + (" ".join(items)) + ")"
+
+    @property
+    def does_vary(self):
+        values = list(self.values.values())
+        return any(v != values[0] for v in values[1:])
+
+    @property
+    def axes_dict(self):
+        if not self.axes:
+            raise ValueError(
+                ".axes must be defined on variable scalar before interpolating"
+            )
+        return {ax.axisTag: ax for ax in self.axes}
+
+    def _normalized_location(self, location):
+        location = self.fix_location(location)
+        normalized_location = {}
+        for axtag in location.keys():
+            if axtag not in self.axes_dict:
+                raise ValueError("Unknown axis %s in %s" % (axtag, location))
+            axis = self.axes_dict[axtag]
+            normalized_location[axtag] = normalizeValue(
+                location[axtag], (axis.minValue, axis.defaultValue, axis.maxValue)
+            )
+
+        return Location(normalized_location)
+
+    def fix_location(self, location):
+        location = dict(location)
+        for tag, axis in self.axes_dict.items():
+            if tag not in location:
+                location[tag] = axis.defaultValue
+        return location
+
+    def add_value(self, location, value):
+        if self.axes:
+            location = self.fix_location(location)
+
+        self.values[Location(location)] = value
+
+    def fix_all_locations(self):
+        self.values = {
+            Location(self.fix_location(l)): v for l, v in self.values.items()
+        }
+
+    @property
+    def default(self):
+        self.fix_all_locations()
+        key = Location({ax.axisTag: ax.defaultValue for ax in self.axes})
+        if key not in self.values:
+            raise ValueError("Default value could not be found")
+            # I *guess* we could interpolate one, but I don't know how.
+        return self.values[key]
+
+    def value_at_location(self, location, model_cache=None, avar=None):
+        loc = Location(location)
+        if loc in self.values.keys():
+            return self.values[loc]
+        values = list(self.values.values())
+        loc = dict(self._normalized_location(loc))
+        return self.model(model_cache, avar).interpolateFromMasters(loc, values)
+
+    def model(self, model_cache=None, avar=None):
+        if model_cache is not None:
+            key = tuple(self.values.keys())
+            if key in model_cache:
+                return model_cache[key]
+        locations = [dict(self._normalized_location(k)) for k in self.values.keys()]
+        if avar is not None:
+            mapping = avar.segments
+            locations = [
+                {
+                    k: piecewiseLinearMap(v, mapping[k]) if k in mapping else v
+                    for k, v in location.items()
+                }
+                for location in locations
+            ]
+        m = VariationModel(locations)
+        if model_cache is not None:
+            model_cache[key] = m
+        return m
+
+    def get_deltas_and_supports(self, model_cache=None, avar=None):
+        values = list(self.values.values())
+        return self.model(model_cache, avar).getDeltasAndSupports(values)
+
+    def add_to_variation_store(self, store_builder, model_cache=None, avar=None):
+        deltas, supports = self.get_deltas_and_supports(model_cache, avar)
+        store_builder.setSupports(supports)
+        index = store_builder.storeDeltas(deltas)
+        return int(self.default), index

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/__init__.py

@@ -0,0 +1,248 @@
+# Copyright 2013 Google, Inc. All Rights Reserved.
+#
+# Google Author(s): Behdad Esfahbod, Roozbeh Pournader
+
+from fontTools import ttLib
+import fontTools.merge.base
+from fontTools.merge.cmap import (
+    computeMegaGlyphOrder,
+    computeMegaCmap,
+    renameCFFCharStrings,
+)
+from fontTools.merge.layout import layoutPreMerge, layoutPostMerge
+from fontTools.merge.options import Options
+import fontTools.merge.tables
+from fontTools.misc.loggingTools import Timer
+from functools import reduce
+import sys
+import logging
+
+
+log = logging.getLogger("fontTools.merge")
+timer = Timer(logger=logging.getLogger(__name__ + ".timer"), level=logging.INFO)
+
+
+class Merger(object):
+    """Font merger.
+
+    This class merges multiple files into a single OpenType font, taking into
+    account complexities such as OpenType layout (``GSUB``/``GPOS``) tables and
+    cross-font metrics (for example ``hhea.ascent`` is set to the maximum value
+    across all the fonts).
+
+    If multiple glyphs map to the same Unicode value, and the glyphs are considered
+    sufficiently different (that is, they differ in any of paths, widths, or
+    height), then subsequent glyphs are renamed and a lookup in the ``locl``
+    feature will be created to disambiguate them. For example, if the arguments
+    are an Arabic font and a Latin font and both contain a set of parentheses,
+    the Latin glyphs will be renamed to ``parenleft.1`` and ``parenright.1``,
+    and a lookup will be inserted into the to ``locl`` feature (creating it if
+    necessary) under the ``latn`` script to substitute ``parenleft`` with
+    ``parenleft.1`` etc.
+
+    Restrictions:
+
+    - All fonts must have the same units per em.
+    - If duplicate glyph disambiguation takes place as described above then the
+      fonts must have a ``GSUB`` table.
+
+    Attributes:
+            options: Currently unused.
+    """
+
+    def __init__(self, options=None):
+        if not options:
+            options = Options()
+
+        self.options = options
+
+    def _openFonts(self, fontfiles):
+        fonts = [ttLib.TTFont(fontfile) for fontfile in fontfiles]
+        for font, fontfile in zip(fonts, fontfiles):
+            font._merger__fontfile = fontfile
+            font._merger__name = font["name"].getDebugName(4)
+        return fonts
+
+    def merge(self, fontfiles):
+        """Merges fonts together.
+
+        Args:
+                fontfiles: A list of file names to be merged
+
+        Returns:
+                A :class:`fontTools.ttLib.TTFont` object. Call the ``save`` method on
+                this to write it out to an OTF file.
+        """
+        #
+        # Settle on a mega glyph order.
+        #
+        fonts = self._openFonts(fontfiles)
+        glyphOrders = [list(font.getGlyphOrder()) for font in fonts]
+        computeMegaGlyphOrder(self, glyphOrders)
+
+        # Take first input file sfntVersion
+        sfntVersion = fonts[0].sfntVersion
+
+        # Reload fonts and set new glyph names on them.
+        fonts = self._openFonts(fontfiles)
+        for font, glyphOrder in zip(fonts, glyphOrders):
+            font.setGlyphOrder(glyphOrder)
+            if "CFF " in font:
+                renameCFFCharStrings(self, glyphOrder, font["CFF "])
+
+        cmaps = [font["cmap"] for font in fonts]
+        self.duplicateGlyphsPerFont = [{} for _ in fonts]
+        computeMegaCmap(self, cmaps)
+
+        mega = ttLib.TTFont(sfntVersion=sfntVersion)
+        mega.setGlyphOrder(self.glyphOrder)
+
+        for font in fonts:
+            self._preMerge(font)
+
+        self.fonts = fonts
+
+        allTags = reduce(set.union, (list(font.keys()) for font in fonts), set())
+        allTags.remove("GlyphOrder")
+
+        for tag in sorted(allTags):
+            if tag in self.options.drop_tables:
+                continue
+
+            with timer("merge '%s'" % tag):
+                tables = [font.get(tag, NotImplemented) for font in fonts]
+
+                log.info("Merging '%s'.", tag)
+                clazz = ttLib.getTableClass(tag)
+                table = clazz(tag).merge(self, tables)
+                # XXX Clean this up and use:  table = mergeObjects(tables)
+
+                if table is not NotImplemented and table is not False:
+                    mega[tag] = table
+                    log.info("Merged '%s'.", tag)
+                else:
+                    log.info("Dropped '%s'.", tag)
+
+        del self.duplicateGlyphsPerFont
+        del self.fonts
+
+        self._postMerge(mega)
+
+        return mega
+
+    def mergeObjects(self, returnTable, logic, tables):
+        # Right now we don't use self at all.  Will use in the future
+        # for options and logging.
+
+        allKeys = set.union(
+            set(),
+            *(vars(table).keys() for table in tables if table is not NotImplemented),
+        )
+        for key in allKeys:
+            log.info(" %s", key)
+            try:
+                mergeLogic = logic[key]
+            except KeyError:
+                try:
+                    mergeLogic = logic["*"]
+                except KeyError:
+                    raise Exception(
+                        "Don't know how to merge key %s of class %s"
+                        % (key, returnTable.__class__.__name__)
+                    )
+            if mergeLogic is NotImplemented:
+                continue
+            value = mergeLogic(getattr(table, key, NotImplemented) for table in tables)
+            if value is not NotImplemented:
+                setattr(returnTable, key, value)
+
+        return returnTable
+
+    def _preMerge(self, font):
+        layoutPreMerge(font)
+
+    def _postMerge(self, font):
+        layoutPostMerge(font)
+
+        if "OS/2" in font:
+            # https://github.com/fonttools/fonttools/issues/2538
+            # TODO: Add an option to disable this?
+            font["OS/2"].recalcAvgCharWidth(font)
+
+
+__all__ = ["Options", "Merger", "main"]
+
+
+@timer("make one with everything (TOTAL TIME)")
+def main(args=None):
+    """Merge multiple fonts into one"""
+    from fontTools import configLogger
+
+    if args is None:
+        args = sys.argv[1:]
+
+    options = Options()
+    args = options.parse_opts(args)
+    fontfiles = []
+    if options.input_file:
+        with open(options.input_file) as inputfile:
+            fontfiles = [
+                line.strip()
+                for line in inputfile.readlines()
+                if not line.lstrip().startswith("#")
+            ]
+    for g in args:
+        fontfiles.append(g)
+
+    if len(fontfiles) < 1:
+        print(
+            "usage: pyftmerge [font1 ... fontN] [--input-file=filelist.txt] [--output-file=merged.ttf] [--import-file=tables.ttx]",
+            file=sys.stderr,
+        )
+        print(
+            "                                   [--drop-tables=tags] [--verbose] [--timing]",
+            file=sys.stderr,
+        )
+        print("", file=sys.stderr)
+        print(" font1 ... fontN              Files to merge.", file=sys.stderr)
+        print(
+            " --input-file=<filename>      Read files to merge from a text file, each path new line. # Comment lines allowed.",
+            file=sys.stderr,
+        )
+        print(
+            " --output-file=<filename>     Specify output file name (default: merged.ttf).",
+            file=sys.stderr,
+        )
+        print(
+            " --import-file=<filename>     TTX file to import after merging. This can be used to set metadata.",
+            file=sys.stderr,
+        )
+        print(
+            " --drop-tables=<table tags>   Comma separated list of table tags to skip, case sensitive.",
+            file=sys.stderr,
+        )
+        print(
+            " --verbose                    Output progress information.",
+            file=sys.stderr,
+        )
+        print(" --timing                     Output progress timing.", file=sys.stderr)
+        return 1
+
+    configLogger(level=logging.INFO if options.verbose else logging.WARNING)
+    if options.timing:
+        timer.logger.setLevel(logging.DEBUG)
+    else:
+        timer.logger.disabled = True
+
+    merger = Merger(options=options)
+    font = merger.merge(fontfiles)
+
+    if options.import_file:
+        font.importXML(options.import_file)
+
+    with timer("compile and save font"):
+        font.save(options.output_file)
+
+
+if __name__ == "__main__":
+    sys.exit(main())

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/__main__.py

@@ -0,0 +1,6 @@
+import sys
+from fontTools.merge import main
+
+
+if __name__ == "__main__":
+    sys.exit(main())

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/base.py

@@ -0,0 +1,81 @@
+# Copyright 2013 Google, Inc. All Rights Reserved.
+#
+# Google Author(s): Behdad Esfahbod, Roozbeh Pournader
+
+from fontTools.ttLib.tables.DefaultTable import DefaultTable
+import logging
+
+
+log = logging.getLogger("fontTools.merge")
+
+
+def add_method(*clazzes, **kwargs):
+    """Returns a decorator function that adds a new method to one or
+    more classes."""
+    allowDefault = kwargs.get("allowDefaultTable", False)
+
+    def wrapper(method):
+        done = []
+        for clazz in clazzes:
+            if clazz in done:
+                continue  # Support multiple names of a clazz
+            done.append(clazz)
+            assert allowDefault or clazz != DefaultTable, "Oops, table class not found."
+            assert (
+                method.__name__ not in clazz.__dict__
+            ), "Oops, class '%s' has method '%s'." % (clazz.__name__, method.__name__)
+            setattr(clazz, method.__name__, method)
+        return None
+
+    return wrapper
+
+
+def mergeObjects(lst):
+    lst = [item for item in lst if item is not NotImplemented]
+    if not lst:
+        return NotImplemented
+    lst = [item for item in lst if item is not None]
+    if not lst:
+        return None
+
+    clazz = lst[0].__class__
+    assert all(type(item) == clazz for item in lst), lst
+
+    logic = clazz.mergeMap
+    returnTable = clazz()
+    returnDict = {}
+
+    allKeys = set.union(set(), *(vars(table).keys() for table in lst))
+    for key in allKeys:
+        try:
+            mergeLogic = logic[key]
+        except KeyError:
+            try:
+                mergeLogic = logic["*"]
+            except KeyError:
+                raise Exception(
+                    "Don't know how to merge key %s of class %s" % (key, clazz.__name__)
+                )
+        if mergeLogic is NotImplemented:
+            continue
+        value = mergeLogic(getattr(table, key, NotImplemented) for table in lst)
+        if value is not NotImplemented:
+            returnDict[key] = value
+
+    returnTable.__dict__ = returnDict
+
+    return returnTable
+
+
+@add_method(DefaultTable, allowDefaultTable=True)
+def merge(self, m, tables):
+    if not hasattr(self, "mergeMap"):
+        log.info("Don't know how to merge '%s'.", self.tableTag)
+        return NotImplemented
+
+    logic = self.mergeMap
+
+    if isinstance(logic, dict):
+        return m.mergeObjects(self, self.mergeMap, tables)
+    else:
+        return logic(tables)

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/cmap.py

@@ -0,0 +1,141 @@
+# Copyright 2013 Google, Inc. All Rights Reserved.
+#
+# Google Author(s): Behdad Esfahbod, Roozbeh Pournader
+
+from fontTools.merge.unicode import is_Default_Ignorable
+from fontTools.pens.recordingPen import DecomposingRecordingPen
+import logging
+
+
+log = logging.getLogger("fontTools.merge")
+
+
+def computeMegaGlyphOrder(merger, glyphOrders):
+    """Modifies passed-in glyphOrders to reflect new glyph names.
+    Stores merger.glyphOrder."""
+    megaOrder = {}
+    for glyphOrder in glyphOrders:
+        for i, glyphName in enumerate(glyphOrder):
+            if glyphName in megaOrder:
+                n = megaOrder[glyphName]
+                while (glyphName + "." + repr(n)) in megaOrder:
+                    n += 1
+                megaOrder[glyphName] = n
+                glyphName += "." + repr(n)
+                glyphOrder[i] = glyphName
+            megaOrder[glyphName] = 1
+    merger.glyphOrder = megaOrder = list(megaOrder.keys())
+
+
+def _glyphsAreSame(
+    glyphSet1,
+    glyphSet2,
+    glyph1,
+    glyph2,
+    advanceTolerance=0.05,
+    advanceToleranceEmpty=0.20,
+):
+    pen1 = DecomposingRecordingPen(glyphSet1)
+    pen2 = DecomposingRecordingPen(glyphSet2)
+    g1 = glyphSet1[glyph1]
+    g2 = glyphSet2[glyph2]
+    g1.draw(pen1)
+    g2.draw(pen2)
+    if pen1.value != pen2.value:
+        return False
+    # Allow more width tolerance for glyphs with no ink
+    tolerance = advanceTolerance if pen1.value else advanceToleranceEmpty
+    # TODO Warn if advances not the same but within tolerance.
+    if abs(g1.width - g2.width) > g1.width * tolerance:
+        return False
+    if hasattr(g1, "height") and g1.height is not None:
+        if abs(g1.height - g2.height) > g1.height * tolerance:
+            return False
+    return True
+
+
+# Valid (format, platformID, platEncID) triplets for cmap subtables containing
+# Unicode BMP-only and Unicode Full Repertoire semantics.
+# Cf. OpenType spec for "Platform specific encodings":
+# https://docs.microsoft.com/en-us/typography/opentype/spec/name
+class _CmapUnicodePlatEncodings:
+    BMP = {(4, 3, 1), (4, 0, 3), (4, 0, 4), (4, 0, 6)}
+    FullRepertoire = {(12, 3, 10), (12, 0, 4), (12, 0, 6)}
+
+
+def computeMegaCmap(merger, cmapTables):
+    """Sets merger.cmap and merger.glyphOrder."""
+
+    # TODO Handle format=14.
+    # Only merge format 4 and 12 Unicode subtables, ignores all other subtables
+    # If there is a format 12 table for a font, ignore the format 4 table of it
+    chosenCmapTables = []
+    for fontIdx, table in enumerate(cmapTables):
+        format4 = None
+        format12 = None
+        for subtable in table.tables:
+            properties = (subtable.format, subtable.platformID, subtable.platEncID)
+            if properties in _CmapUnicodePlatEncodings.BMP:
+                format4 = subtable
+            elif properties in _CmapUnicodePlatEncodings.FullRepertoire:
+                format12 = subtable
+            else:
+                log.warning(
+                    "Dropped cmap subtable from font '%s':\t"
+                    "format %2s, platformID %2s, platEncID %2s",
+                    fontIdx,
+                    subtable.format,
+                    subtable.platformID,
+                    subtable.platEncID,
+                )
+        if format12 is not None:
+            chosenCmapTables.append((format12, fontIdx))
+        elif format4 is not None:
+            chosenCmapTables.append((format4, fontIdx))
+
+    # Build the unicode mapping
+    merger.cmap = cmap = {}
+    fontIndexForGlyph = {}
+    glyphSets = [None for f in merger.fonts] if hasattr(merger, "fonts") else None
+
+    for table, fontIdx in chosenCmapTables:
+        # handle duplicates
+        for uni, gid in table.cmap.items():
+            oldgid = cmap.get(uni, None)
+            if oldgid is None:
+                cmap[uni] = gid
+                fontIndexForGlyph[gid] = fontIdx
+            elif is_Default_Ignorable(uni) or uni in (0x25CC,):  # U+25CC DOTTED CIRCLE
+                continue
+            elif oldgid != gid:
+                # Char previously mapped to oldgid, now to gid.
+                # Record, to fix up in GSUB 'locl' later.
+                if merger.duplicateGlyphsPerFont[fontIdx].get(oldgid) is None:
+                    if glyphSets is not None:
+                        oldFontIdx = fontIndexForGlyph[oldgid]
+                        for idx in (fontIdx, oldFontIdx):
+                            if glyphSets[idx] is None:
+                                glyphSets[idx] = merger.fonts[idx].getGlyphSet()
+                        # if _glyphsAreSame(glyphSets[oldFontIdx], glyphSets[fontIdx], oldgid, gid):
+                        # 	continue
+                    merger.duplicateGlyphsPerFont[fontIdx][oldgid] = gid
+                elif merger.duplicateGlyphsPerFont[fontIdx][oldgid] != gid:
+                    # Char previously mapped to oldgid but oldgid is already remapped to a different
+                    # gid, because of another Unicode character.
+                    # TODO: Try harder to do something about these.
+                    log.warning(
+                        "Dropped mapping from codepoint %#06X to glyphId '%s'", uni, gid
+                    )
+
+
+def renameCFFCharStrings(merger, glyphOrder, cffTable):
+    """Rename topDictIndex charStrings based on glyphOrder."""
+    td = cffTable.cff.topDictIndex[0]
+
+    charStrings = {}
+    for i, v in enumerate(td.CharStrings.charStrings.values()):
+        glyphName = glyphOrder[i]
+        charStrings[glyphName] = v
+    td.CharStrings.charStrings = charStrings
+
+    td.charset = list(glyphOrder)

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/layout.py

@@ -0,0 +1,526 @@
+# Copyright 2013 Google, Inc. All Rights Reserved.
+#
+# Google Author(s): Behdad Esfahbod, Roozbeh Pournader
+
+from fontTools import ttLib
+from fontTools.ttLib.tables.DefaultTable import DefaultTable
+from fontTools.ttLib.tables import otTables
+from fontTools.merge.base import add_method, mergeObjects
+from fontTools.merge.util import *
+import logging
+
+
+log = logging.getLogger("fontTools.merge")
+
+
+def mergeLookupLists(lst):
+    # TODO Do smarter merge.
+    return sumLists(lst)
+
+
+def mergeFeatures(lst):
+    assert lst
+    self = otTables.Feature()
+    self.FeatureParams = None
+    self.LookupListIndex = mergeLookupLists(
+        [l.LookupListIndex for l in lst if l.LookupListIndex]
+    )
+    self.LookupCount = len(self.LookupListIndex)
+    return self
+
+
+def mergeFeatureLists(lst):
+    d = {}
+    for l in lst:
+        for f in l:
+            tag = f.FeatureTag
+            if tag not in d:
+                d[tag] = []
+            d[tag].append(f.Feature)
+    ret = []
+    for tag in sorted(d.keys()):
+        rec = otTables.FeatureRecord()
+        rec.FeatureTag = tag
+        rec.Feature = mergeFeatures(d[tag])
+        ret.append(rec)
+    return ret
+
+
+def mergeLangSyses(lst):
+    assert lst
+
+    # TODO Support merging ReqFeatureIndex
+    assert all(l.ReqFeatureIndex == 0xFFFF for l in lst)
+
+    self = otTables.LangSys()
+    self.LookupOrder = None
+    self.ReqFeatureIndex = 0xFFFF
+    self.FeatureIndex = mergeFeatureLists(
+        [l.FeatureIndex for l in lst if l.FeatureIndex]
+    )
+    self.FeatureCount = len(self.FeatureIndex)
+    return self
+
+
+def mergeScripts(lst):
+    assert lst
+
+    if len(lst) == 1:
+        return lst[0]
+    langSyses = {}
+    for sr in lst:
+        for lsr in sr.LangSysRecord:
+            if lsr.LangSysTag not in langSyses:
+                langSyses[lsr.LangSysTag] = []
+            langSyses[lsr.LangSysTag].append(lsr.LangSys)
+    lsrecords = []
+    for tag, langSys_list in sorted(langSyses.items()):
+        lsr = otTables.LangSysRecord()
+        lsr.LangSys = mergeLangSyses(langSys_list)
+        lsr.LangSysTag = tag
+        lsrecords.append(lsr)
+
+    self = otTables.Script()
+    self.LangSysRecord = lsrecords
+    self.LangSysCount = len(lsrecords)
+    dfltLangSyses = [s.DefaultLangSys for s in lst if s.DefaultLangSys]
+    if dfltLangSyses:
+        self.DefaultLangSys = mergeLangSyses(dfltLangSyses)
+    else:
+        self.DefaultLangSys = None
+    return self
+
+
+def mergeScriptRecords(lst):
+    d = {}
+    for l in lst:
+        for s in l:
+            tag = s.ScriptTag
+            if tag not in d:
+                d[tag] = []
+            d[tag].append(s.Script)
+    ret = []
+    for tag in sorted(d.keys()):
+        rec = otTables.ScriptRecord()
+        rec.ScriptTag = tag
+        rec.Script = mergeScripts(d[tag])
+        ret.append(rec)
+    return ret
+
+
+otTables.ScriptList.mergeMap = {
+    "ScriptCount": lambda lst: None,  # TODO
+    "ScriptRecord": mergeScriptRecords,
+}
+otTables.BaseScriptList.mergeMap = {
+    "BaseScriptCount": lambda lst: None,  # TODO
+    # TODO: Merge duplicate entries
+    "BaseScriptRecord": lambda lst: sorted(
+        sumLists(lst), key=lambda s: s.BaseScriptTag
+    ),
+}
+
+otTables.FeatureList.mergeMap = {
+    "FeatureCount": sum,
+    "FeatureRecord": lambda lst: sorted(sumLists(lst), key=lambda s: s.FeatureTag),
+}
+
+otTables.LookupList.mergeMap = {
+    "LookupCount": sum,
+    "Lookup": sumLists,
+}
+
+otTables.Coverage.mergeMap = {
+    "Format": min,
+    "glyphs": sumLists,
+}
+
+otTables.ClassDef.mergeMap = {
+    "Format": min,
+    "classDefs": sumDicts,
+}
+
+otTables.LigCaretList.mergeMap = {
+    "Coverage": mergeObjects,
+    "LigGlyphCount": sum,
+    "LigGlyph": sumLists,
+}
+
+otTables.AttachList.mergeMap = {
+    "Coverage": mergeObjects,
+    "GlyphCount": sum,
+    "AttachPoint": sumLists,
+}
+
+# XXX Renumber MarkFilterSets of lookups
+otTables.MarkGlyphSetsDef.mergeMap = {
+    "MarkSetTableFormat": equal,
+    "MarkSetCount": sum,
+    "Coverage": sumLists,
+}
+
+otTables.Axis.mergeMap = {
+    "*": mergeObjects,
+}
+
+# XXX Fix BASE table merging
+otTables.BaseTagList.mergeMap = {
+    "BaseTagCount": sum,
+    "BaselineTag": sumLists,
+}
+
+otTables.GDEF.mergeMap = otTables.GSUB.mergeMap = otTables.GPOS.mergeMap = (
+    otTables.BASE.mergeMap
+) = otTables.JSTF.mergeMap = otTables.MATH.mergeMap = {
+    "*": mergeObjects,
+    "Version": max,
+}
+
+ttLib.getTableClass("GDEF").mergeMap = ttLib.getTableClass("GSUB").mergeMap = (
+    ttLib.getTableClass("GPOS").mergeMap
+) = ttLib.getTableClass("BASE").mergeMap = ttLib.getTableClass(
+    "JSTF"
+).mergeMap = ttLib.getTableClass(
+    "MATH"
+).mergeMap = {
+    "tableTag": onlyExisting(equal),  # XXX clean me up
+    "table": mergeObjects,
+}
+
+
+@add_method(ttLib.getTableClass("GSUB"))
+def merge(self, m, tables):
+    assert len(tables) == len(m.duplicateGlyphsPerFont)
+    for i, (table, dups) in enumerate(zip(tables, m.duplicateGlyphsPerFont)):
+        if not dups:
+            continue
+        if table is None or table is NotImplemented:
+            log.warning(
+                "Have non-identical duplicates to resolve for '%s' but no GSUB. Are duplicates intended?: %s",
+                m.fonts[i]._merger__name,
+                dups,
+            )
+            continue
+
+        synthFeature = None
+        synthLookup = None
+        for script in table.table.ScriptList.ScriptRecord:
+            if script.ScriptTag == "DFLT":
+                continue  # XXX
+            for langsys in [script.Script.DefaultLangSys] + [
+                l.LangSys for l in script.Script.LangSysRecord
+            ]:
+                if langsys is None:
+                    continue  # XXX Create!
+                feature = [v for v in langsys.FeatureIndex if v.FeatureTag == "locl"]
+                assert len(feature) <= 1
+                if feature:
+                    feature = feature[0]
+                else:
+                    if not synthFeature:
+                        synthFeature = otTables.FeatureRecord()
+                        synthFeature.FeatureTag = "locl"
+                        f = synthFeature.Feature = otTables.Feature()
+                        f.FeatureParams = None
+                        f.LookupCount = 0
+                        f.LookupListIndex = []
+                        table.table.FeatureList.FeatureRecord.append(synthFeature)
+                        table.table.FeatureList.FeatureCount += 1
+                    feature = synthFeature
+                    langsys.FeatureIndex.append(feature)
+                    langsys.FeatureIndex.sort(key=lambda v: v.FeatureTag)
+
+                if not synthLookup:
+                    subtable = otTables.SingleSubst()
+                    subtable.mapping = dups
+                    synthLookup = otTables.Lookup()
+                    synthLookup.LookupFlag = 0
+                    synthLookup.LookupType = 1
+                    synthLookup.SubTableCount = 1
+                    synthLookup.SubTable = [subtable]
+                    if table.table.LookupList is None:
+                        # mtiLib uses None as default value for LookupList,
+                        # while feaLib points to an empty array with count 0
+                        # TODO: make them do the same
+                        table.table.LookupList = otTables.LookupList()
+                        table.table.LookupList.Lookup = []
+                        table.table.LookupList.LookupCount = 0
+                    table.table.LookupList.Lookup.append(synthLookup)
+                    table.table.LookupList.LookupCount += 1
+
+                if feature.Feature.LookupListIndex[:1] != [synthLookup]:
+                    feature.Feature.LookupListIndex[:0] = [synthLookup]
+                    feature.Feature.LookupCount += 1
+
+    DefaultTable.merge(self, m, tables)
+    return self
+
+
+@add_method(
+    otTables.SingleSubst,
+    otTables.MultipleSubst,
+    otTables.AlternateSubst,
+    otTables.LigatureSubst,
+    otTables.ReverseChainSingleSubst,
+    otTables.SinglePos,
+    otTables.PairPos,
+    otTables.CursivePos,
+    otTables.MarkBasePos,
+    otTables.MarkLigPos,
+    otTables.MarkMarkPos,
+)
+def mapLookups(self, lookupMap):
+    pass
+
+
+# Copied and trimmed down from subset.py
+@add_method(
+    otTables.ContextSubst,
+    otTables.ChainContextSubst,
+    otTables.ContextPos,
+    otTables.ChainContextPos,
+)
+def __merge_classify_context(self):
+    class ContextHelper(object):
+        def __init__(self, klass, Format):
+            if klass.__name__.endswith("Subst"):
+                Typ = "Sub"
+                Type = "Subst"
+            else:
+                Typ = "Pos"
+                Type = "Pos"
+            if klass.__name__.startswith("Chain"):
+                Chain = "Chain"
+            else:
+                Chain = ""
+            ChainTyp = Chain + Typ
+
+            self.Typ = Typ
+            self.Type = Type
+            self.Chain = Chain
+            self.ChainTyp = ChainTyp
+
+            self.LookupRecord = Type + "LookupRecord"
+
+            if Format == 1:
+                self.Rule = ChainTyp + "Rule"
+                self.RuleSet = ChainTyp + "RuleSet"
+            elif Format == 2:
+                self.Rule = ChainTyp + "ClassRule"
+                self.RuleSet = ChainTyp + "ClassSet"
+
+    if self.Format not in [1, 2, 3]:
+        return None  # Don't shoot the messenger; let it go
+    if not hasattr(self.__class__, "_merge__ContextHelpers"):
+        self.__class__._merge__ContextHelpers = {}
+    if self.Format not in self.__class__._merge__ContextHelpers:
+        helper = ContextHelper(self.__class__, self.Format)
+        self.__class__._merge__ContextHelpers[self.Format] = helper
+    return self.__class__._merge__ContextHelpers[self.Format]
+
+
+@add_method(
+    otTables.ContextSubst,
+    otTables.ChainContextSubst,
+    otTables.ContextPos,
+    otTables.ChainContextPos,
+)
+def mapLookups(self, lookupMap):
+    c = self.__merge_classify_context()
+
+    if self.Format in [1, 2]:
+        for rs in getattr(self, c.RuleSet):
+            if not rs:
+                continue
+            for r in getattr(rs, c.Rule):
+                if not r:
+                    continue
+                for ll in getattr(r, c.LookupRecord):
+                    if not ll:
+                        continue
+                    ll.LookupListIndex = lookupMap[ll.LookupListIndex]
+    elif self.Format == 3:
+        for ll in getattr(self, c.LookupRecord):
+            if not ll:
+                continue
+            ll.LookupListIndex = lookupMap[ll.LookupListIndex]
+    else:
+        assert 0, "unknown format: %s" % self.Format
+
+
+@add_method(otTables.ExtensionSubst, otTables.ExtensionPos)
+def mapLookups(self, lookupMap):
+    if self.Format == 1:
+        self.ExtSubTable.mapLookups(lookupMap)
+    else:
+        assert 0, "unknown format: %s" % self.Format
+
+
+@add_method(otTables.Lookup)
+def mapLookups(self, lookupMap):
+    for st in self.SubTable:
+        if not st:
+            continue
+        st.mapLookups(lookupMap)
+
+
+@add_method(otTables.LookupList)
+def mapLookups(self, lookupMap):
+    for l in self.Lookup:
+        if not l:
+            continue
+        l.mapLookups(lookupMap)
+
+
+@add_method(otTables.Lookup)
+def mapMarkFilteringSets(self, markFilteringSetMap):
+    if self.LookupFlag & 0x0010:
+        self.MarkFilteringSet = markFilteringSetMap[self.MarkFilteringSet]
+
+
+@add_method(otTables.LookupList)
+def mapMarkFilteringSets(self, markFilteringSetMap):
+    for l in self.Lookup:
+        if not l:
+            continue
+        l.mapMarkFilteringSets(markFilteringSetMap)
+
+
+@add_method(otTables.Feature)
+def mapLookups(self, lookupMap):
+    self.LookupListIndex = [lookupMap[i] for i in self.LookupListIndex]
+
+
+@add_method(otTables.FeatureList)
+def mapLookups(self, lookupMap):
+    for f in self.FeatureRecord:
+        if not f or not f.Feature:
+            continue
+        f.Feature.mapLookups(lookupMap)
+
+
+@add_method(otTables.DefaultLangSys, otTables.LangSys)
+def mapFeatures(self, featureMap):
+    self.FeatureIndex = [featureMap[i] for i in self.FeatureIndex]
+    if self.ReqFeatureIndex != 65535:
+        self.ReqFeatureIndex = featureMap[self.ReqFeatureIndex]
+
+
+@add_method(otTables.Script)
+def mapFeatures(self, featureMap):
+    if self.DefaultLangSys:
+        self.DefaultLangSys.mapFeatures(featureMap)
+    for l in self.LangSysRecord:
+        if not l or not l.LangSys:
+            continue
+        l.LangSys.mapFeatures(featureMap)
+
+
+@add_method(otTables.ScriptList)
+def mapFeatures(self, featureMap):
+    for s in self.ScriptRecord:
+        if not s or not s.Script:
+            continue
+        s.Script.mapFeatures(featureMap)
+
+
+def layoutPreMerge(font):
+    # Map indices to references
+
+    GDEF = font.get("GDEF")
+    GSUB = font.get("GSUB")
+    GPOS = font.get("GPOS")
+
+    for t in [GSUB, GPOS]:
+        if not t:
+            continue
+
+        if t.table.LookupList:
+            lookupMap = {i: v for i, v in enumerate(t.table.LookupList.Lookup)}
+            t.table.LookupList.mapLookups(lookupMap)
+            t.table.FeatureList.mapLookups(lookupMap)
+
+            if (
+                GDEF
+                and GDEF.table.Version >= 0x00010002
+                and GDEF.table.MarkGlyphSetsDef
+            ):
+                markFilteringSetMap = {
+                    i: v for i, v in enumerate(GDEF.table.MarkGlyphSetsDef.Coverage)
+                }
+                t.table.LookupList.mapMarkFilteringSets(markFilteringSetMap)
+
+        if t.table.FeatureList and t.table.ScriptList:
+            featureMap = {i: v for i, v in enumerate(t.table.FeatureList.FeatureRecord)}
+            t.table.ScriptList.mapFeatures(featureMap)
+
+    # TODO FeatureParams nameIDs
+
+
+def layoutPostMerge(font):
+    # Map references back to indices
+
+    GDEF = font.get("GDEF")
+    GSUB = font.get("GSUB")
+    GPOS = font.get("GPOS")
+
+    for t in [GSUB, GPOS]:
+        if not t:
+            continue
+
+        if t.table.FeatureList and t.table.ScriptList:
+            # Collect unregistered (new) features.
+            featureMap = GregariousIdentityDict(t.table.FeatureList.FeatureRecord)
+            t.table.ScriptList.mapFeatures(featureMap)
+
+            # Record used features.
+            featureMap = AttendanceRecordingIdentityDict(
+                t.table.FeatureList.FeatureRecord
+            )
+            t.table.ScriptList.mapFeatures(featureMap)
+            usedIndices = featureMap.s
+
+            # Remove unused features
+            t.table.FeatureList.FeatureRecord = [
+                f
+                for i, f in enumerate(t.table.FeatureList.FeatureRecord)
+                if i in usedIndices
+            ]
+
+            # Map back to indices.
+            featureMap = NonhashableDict(t.table.FeatureList.FeatureRecord)
+            t.table.ScriptList.mapFeatures(featureMap)
+
+            t.table.FeatureList.FeatureCount = len(t.table.FeatureList.FeatureRecord)
+
+        if t.table.LookupList:
+            # Collect unregistered (new) lookups.
+            lookupMap = GregariousIdentityDict(t.table.LookupList.Lookup)
+            t.table.FeatureList.mapLookups(lookupMap)
+            t.table.LookupList.mapLookups(lookupMap)
+
+            # Record used lookups.
+            lookupMap = AttendanceRecordingIdentityDict(t.table.LookupList.Lookup)
+            t.table.FeatureList.mapLookups(lookupMap)
+            t.table.LookupList.mapLookups(lookupMap)
+            usedIndices = lookupMap.s
+
+            # Remove unused lookups
+            t.table.LookupList.Lookup = [
+                l for i, l in enumerate(t.table.LookupList.Lookup) if i in usedIndices
+            ]
+
+            # Map back to indices.
+            lookupMap = NonhashableDict(t.table.LookupList.Lookup)
+            t.table.FeatureList.mapLookups(lookupMap)
+            t.table.LookupList.mapLookups(lookupMap)
+
+            t.table.LookupList.LookupCount = len(t.table.LookupList.Lookup)
+
+            if GDEF and GDEF.table.Version >= 0x00010002:
+                markFilteringSetMap = NonhashableDict(
+                    GDEF.table.MarkGlyphSetsDef.Coverage
+                )
+                t.table.LookupList.mapMarkFilteringSets(markFilteringSetMap)
+
+    # TODO FeatureParams nameIDs

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/options.py

@@ -0,0 +1,85 @@
+# Copyright 2013 Google, Inc. All Rights Reserved.
+#
+# Google Author(s): Behdad Esfahbod, Roozbeh Pournader
+
+
+class Options(object):
+    class UnknownOptionError(Exception):
+        pass
+
+    def __init__(self, **kwargs):
+        self.verbose = False
+        self.timing = False
+        self.drop_tables = []
+        self.input_file = None
+        self.output_file = "merged.ttf"
+        self.import_file = None
+
+        self.set(**kwargs)
+
+    def set(self, **kwargs):
+        for k, v in kwargs.items():
+            if not hasattr(self, k):
+                raise self.UnknownOptionError("Unknown option '%s'" % k)
+            setattr(self, k, v)
+
+    def parse_opts(self, argv, ignore_unknown=[]):
+        ret = []
+        opts = {}
+        for a in argv:
+            orig_a = a
+            if not a.startswith("--"):
+                ret.append(a)
+                continue
+            a = a[2:]
+            i = a.find("=")
+            op = "="
+            if i == -1:
+                if a.startswith("no-"):
+                    k = a[3:]
+                    v = False
+                else:
+                    k = a
+                    v = True
+            else:
+                k = a[:i]
+                if k[-1] in "-+":
+                    op = k[-1] + "="  # Ops is '-=' or '+=' now.
+                    k = k[:-1]
+                v = a[i + 1 :]
+            ok = k
+            k = k.replace("-", "_")
+            if not hasattr(self, k):
+                if ignore_unknown is True or ok in ignore_unknown:
+                    ret.append(orig_a)
+                    continue
+                else:
+                    raise self.UnknownOptionError("Unknown option '%s'" % a)
+
+            ov = getattr(self, k)
+            if isinstance(ov, bool):
+                v = bool(v)
+            elif isinstance(ov, int):
+                v = int(v)
+            elif isinstance(ov, list):
+                vv = v.split(",")
+                if vv == [""]:
+                    vv = []
+                vv = [int(x, 0) if len(x) and x[0] in "0123456789" else x for x in vv]
+                if op == "=":
+                    v = vv
+                elif op == "+=":
+                    v = ov
+                    v.extend(vv)
+                elif op == "-=":
+                    v = ov
+                    for x in vv:
+                        if x in v:
+                            v.remove(x)
+                else:
+                    assert 0
+
+            opts[k] = v
+        self.set(**opts)
+
+        return ret

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/tables.py

@@ -0,0 +1,341 @@
+# Copyright 2013 Google, Inc. All Rights Reserved.
+#
+# Google Author(s): Behdad Esfahbod, Roozbeh Pournader
+
+from fontTools import ttLib, cffLib
+from fontTools.misc.psCharStrings import T2WidthExtractor
+from fontTools.ttLib.tables.DefaultTable import DefaultTable
+from fontTools.merge.base import add_method, mergeObjects
+from fontTools.merge.cmap import computeMegaCmap
+from fontTools.merge.util import *
+import logging
+
+
+log = logging.getLogger("fontTools.merge")
+
+
+ttLib.getTableClass("maxp").mergeMap = {
+    "*": max,
+    "tableTag": equal,
+    "tableVersion": equal,
+    "numGlyphs": sum,
+    "maxStorage": first,
+    "maxFunctionDefs": first,
+    "maxInstructionDefs": first,
+    # TODO When we correctly merge hinting data, update these values:
+    # maxFunctionDefs, maxInstructionDefs, maxSizeOfInstructions
+}
+
+headFlagsMergeBitMap = {
+    "size": 16,
+    "*": bitwise_or,
+    1: bitwise_and,  # Baseline at y = 0
+    2: bitwise_and,  # lsb at x = 0
+    3: bitwise_and,  # Force ppem to integer values. FIXME?
+    5: bitwise_and,  # Font is vertical
+    6: lambda bit: 0,  # Always set to zero
+    11: bitwise_and,  # Font data is 'lossless'
+    13: bitwise_and,  # Optimized for ClearType
+    14: bitwise_and,  # Last resort font. FIXME? equal or first may be better
+    15: lambda bit: 0,  # Always set to zero
+}
+
+ttLib.getTableClass("head").mergeMap = {
+    "tableTag": equal,
+    "tableVersion": max,
+    "fontRevision": max,
+    "checkSumAdjustment": lambda lst: 0,  # We need *something* here
+    "magicNumber": equal,
+    "flags": mergeBits(headFlagsMergeBitMap),
+    "unitsPerEm": equal,
+    "created": current_time,
+    "modified": current_time,
+    "xMin": min,
+    "yMin": min,
+    "xMax": max,
+    "yMax": max,
+    "macStyle": first,
+    "lowestRecPPEM": max,
+    "fontDirectionHint": lambda lst: 2,
+    "indexToLocFormat": first,
+    "glyphDataFormat": equal,
+}
+
+ttLib.getTableClass("hhea").mergeMap = {
+    "*": equal,
+    "tableTag": equal,
+    "tableVersion": max,
+    "ascent": max,
+    "descent": min,
+    "lineGap": max,
+    "advanceWidthMax": max,
+    "minLeftSideBearing": min,
+    "minRightSideBearing": min,
+    "xMaxExtent": max,
+    "caretSlopeRise": first,
+    "caretSlopeRun": first,
+    "caretOffset": first,
+    "numberOfHMetrics": recalculate,
+}
+
+ttLib.getTableClass("vhea").mergeMap = {
+    "*": equal,
+    "tableTag": equal,
+    "tableVersion": max,
+    "ascent": max,
+    "descent": min,
+    "lineGap": max,
+    "advanceHeightMax": max,
+    "minTopSideBearing": min,
+    "minBottomSideBearing": min,
+    "yMaxExtent": max,
+    "caretSlopeRise": first,
+    "caretSlopeRun": first,
+    "caretOffset": first,
+    "numberOfVMetrics": recalculate,
+}
+
+os2FsTypeMergeBitMap = {
+    "size": 16,
+    "*": lambda bit: 0,
+    1: bitwise_or,  # no embedding permitted
+    2: bitwise_and,  # allow previewing and printing documents
+    3: bitwise_and,  # allow editing documents
+    8: bitwise_or,  # no subsetting permitted
+    9: bitwise_or,  # no embedding of outlines permitted
+}
+
+
+def mergeOs2FsType(lst):
+    lst = list(lst)
+    if all(item == 0 for item in lst):
+        return 0
+
+    # Compute least restrictive logic for each fsType value
+    for i in range(len(lst)):
+        # unset bit 1 (no embedding permitted) if either bit 2 or 3 is set
+        if lst[i] & 0x000C:
+            lst[i] &= ~0x0002
+        # set bit 2 (allow previewing) if bit 3 is set (allow editing)
+        elif lst[i] & 0x0008:
+            lst[i] |= 0x0004
+        # set bits 2 and 3 if everything is allowed
+        elif lst[i] == 0:
+            lst[i] = 0x000C
+
+    fsType = mergeBits(os2FsTypeMergeBitMap)(lst)
+    # unset bits 2 and 3 if bit 1 is set (some font is "no embedding")
+    if fsType & 0x0002:
+        fsType &= ~0x000C
+    return fsType
+
+
+ttLib.getTableClass("OS/2").mergeMap = {
+    "*": first,
+    "tableTag": equal,
+    "version": max,
+    "xAvgCharWidth": first,  # Will be recalculated at the end on the merged font
+    "fsType": mergeOs2FsType,  # Will be overwritten
+    "panose": first,  # FIXME: should really be the first Latin font
+    "ulUnicodeRange1": bitwise_or,
+    "ulUnicodeRange2": bitwise_or,
+    "ulUnicodeRange3": bitwise_or,
+    "ulUnicodeRange4": bitwise_or,
+    "fsFirstCharIndex": min,
+    "fsLastCharIndex": max,
+    "sTypoAscender": max,
+    "sTypoDescender": min,
+    "sTypoLineGap": max,
+    "usWinAscent": max,
+    "usWinDescent": max,
+    # Version 1
+    "ulCodePageRange1": onlyExisting(bitwise_or),
+    "ulCodePageRange2": onlyExisting(bitwise_or),
+    # Version 2, 3, 4
+    "sxHeight": onlyExisting(max),
+    "sCapHeight": onlyExisting(max),
+    "usDefaultChar": onlyExisting(first),
+    "usBreakChar": onlyExisting(first),
+    "usMaxContext": onlyExisting(max),
+    # version 5
+    "usLowerOpticalPointSize": onlyExisting(min),
+    "usUpperOpticalPointSize": onlyExisting(max),
+}
+
+
+@add_method(ttLib.getTableClass("OS/2"))
+def merge(self, m, tables):
+    DefaultTable.merge(self, m, tables)
+    if self.version < 2:
+        # bits 8 and 9 are reserved and should be set to zero
+        self.fsType &= ~0x0300
+    if self.version >= 3:
+        # Only one of bits 1, 2, and 3 may be set. We already take
+        # care of bit 1 implications in mergeOs2FsType. So unset
+        # bit 2 if bit 3 is already set.
+        if self.fsType & 0x0008:
+            self.fsType &= ~0x0004
+    return self
+
+
+ttLib.getTableClass("post").mergeMap = {
+    "*": first,
+    "tableTag": equal,
+    "formatType": max,
+    "isFixedPitch": min,
+    "minMemType42": max,
+    "maxMemType42": lambda lst: 0,
+    "minMemType1": max,
+    "maxMemType1": lambda lst: 0,
+    "mapping": onlyExisting(sumDicts),
+    "extraNames": lambda lst: [],
+}
+
+ttLib.getTableClass("vmtx").mergeMap = ttLib.getTableClass("hmtx").mergeMap = {
+    "tableTag": equal,
+    "metrics": sumDicts,
+}
+
+ttLib.getTableClass("name").mergeMap = {
+    "tableTag": equal,
+    "names": first,  # FIXME? Does mixing name records make sense?
+}
+
+ttLib.getTableClass("loca").mergeMap = {
+    "*": recalculate,
+    "tableTag": equal,
+}
+
+ttLib.getTableClass("glyf").mergeMap = {
+    "tableTag": equal,
+    "glyphs": sumDicts,
+    "glyphOrder": sumLists,
+    "_reverseGlyphOrder": recalculate,
+    "axisTags": equal,
+}
+
+
+@add_method(ttLib.getTableClass("glyf"))
+def merge(self, m, tables):
+    for i, table in enumerate(tables):
+        for g in table.glyphs.values():
+            if i:
+                # Drop hints for all but first font, since
+                # we don't map functions / CVT values.
+                g.removeHinting()
+            # Expand composite glyphs to load their
+            # composite glyph names.
+            if g.isComposite():
+                g.expand(table)
+    return DefaultTable.merge(self, m, tables)
+
+
+ttLib.getTableClass("prep").mergeMap = lambda self, lst: first(lst)
+ttLib.getTableClass("fpgm").mergeMap = lambda self, lst: first(lst)
+ttLib.getTableClass("cvt ").mergeMap = lambda self, lst: first(lst)
+ttLib.getTableClass("gasp").mergeMap = lambda self, lst: first(
+    lst
+)  # FIXME? Appears irreconcilable
+
+
+@add_method(ttLib.getTableClass("CFF "))
+def merge(self, m, tables):
+    if any(hasattr(table.cff[0], "FDSelect") for table in tables):
+        raise NotImplementedError("Merging CID-keyed CFF tables is not supported yet")
+
+    for table in tables:
+        table.cff.desubroutinize()
+
+    newcff = tables[0]
+    newfont = newcff.cff[0]
+    private = newfont.Private
+    newDefaultWidthX, newNominalWidthX = private.defaultWidthX, private.nominalWidthX
+    storedNamesStrings = []
+    glyphOrderStrings = []
+    glyphOrder = set(newfont.getGlyphOrder())
+
+    for name in newfont.strings.strings:
+        if name not in glyphOrder:
+            storedNamesStrings.append(name)
+        else:
+            glyphOrderStrings.append(name)
+
+    chrset = list(newfont.charset)
+    newcs = newfont.CharStrings
+    log.debug("FONT 0 CharStrings: %d.", len(newcs))
+
+    for i, table in enumerate(tables[1:], start=1):
+        font = table.cff[0]
+        defaultWidthX, nominalWidthX = (
+            font.Private.defaultWidthX,
+            font.Private.nominalWidthX,
+        )
+        widthsDiffer = (
+            defaultWidthX != newDefaultWidthX or nominalWidthX != newNominalWidthX
+        )
+        font.Private = private
+        fontGlyphOrder = set(font.getGlyphOrder())
+        for name in font.strings.strings:
+            if name in fontGlyphOrder:
+                glyphOrderStrings.append(name)
+        cs = font.CharStrings
+        gs = table.cff.GlobalSubrs
+        log.debug("Font %d CharStrings: %d.", i, len(cs))
+        chrset.extend(font.charset)
+        if newcs.charStringsAreIndexed:
+            for i, name in enumerate(cs.charStrings, start=len(newcs)):
+                newcs.charStrings[name] = i
+                newcs.charStringsIndex.items.append(None)
+        for name in cs.charStrings:
+            if widthsDiffer:
+                c = cs[name]
+                defaultWidthXToken = object()
+                extractor = T2WidthExtractor([], [], nominalWidthX, defaultWidthXToken)
+                extractor.execute(c)
+                width = extractor.width
+                if width is not defaultWidthXToken:
+                    # The following will be wrong if the width is added
+                    # by a subroutine. Ouch!
+                    c.program.pop(0)
+                else:
+                    width = defaultWidthX
+                if width != newDefaultWidthX:
+                    c.program.insert(0, width - newNominalWidthX)
+            newcs[name] = cs[name]
+
+    newfont.charset = chrset
+    newfont.numGlyphs = len(chrset)
+    newfont.strings.strings = glyphOrderStrings + storedNamesStrings
+
+    return newcff
+
+
+@add_method(ttLib.getTableClass("cmap"))
+def merge(self, m, tables):
+    # TODO Handle format=14.
+    if not hasattr(m, "cmap"):
+        computeMegaCmap(m, tables)
+    cmap = m.cmap
+
+    cmapBmpOnly = {uni: gid for uni, gid in cmap.items() if uni <= 0xFFFF}
+    self.tables = []
+    module = ttLib.getTableModule("cmap")
+    if len(cmapBmpOnly) != len(cmap):
+        # format-12 required.
+        cmapTable = module.cmap_classes[12](12)
+        cmapTable.platformID = 3
+        cmapTable.platEncID = 10
+        cmapTable.language = 0
+        cmapTable.cmap = cmap
+        self.tables.append(cmapTable)
+    # always create format-4
+    cmapTable = module.cmap_classes[4](4)
+    cmapTable.platformID = 3
+    cmapTable.platEncID = 1
+    cmapTable.language = 0
+    cmapTable.cmap = cmapBmpOnly
+    # ordered by platform then encoding
+    self.tables.insert(0, cmapTable)
+    self.tableVersion = 0
+    self.numSubTables = len(self.tables)
+    return self

infer_4_47_1/lib/python3.10/site-packages/fontTools/merge/unicode.py

@@ -0,0 +1,78 @@
+# Copyright 2021 Behdad Esfahbod. All Rights Reserved.
+
+
+def is_Default_Ignorable(u):
+    # http://www.unicode.org/reports/tr44/#Default_Ignorable_Code_Point
+    #
+    # TODO Move me to unicodedata module and autogenerate.
+    #
+    # Unicode 14.0:
+    # $ grep '; Default_Ignorable_Code_Point ' DerivedCoreProperties.txt | sed 's/;.*#/#/'
+    # 00AD          # Cf       SOFT HYPHEN
+    # 034F          # Mn       COMBINING GRAPHEME JOINER
+    # 061C          # Cf       ARABIC LETTER MARK
+    # 115F..1160    # Lo   [2] HANGUL CHOSEONG FILLER..HANGUL JUNGSEONG FILLER
+    # 17B4..17B5    # Mn   [2] KHMER VOWEL INHERENT AQ..KHMER VOWEL INHERENT AA
+    # 180B..180D    # Mn   [3] MONGOLIAN FREE VARIATION SELECTOR ONE..MONGOLIAN FREE VARIATION SELECTOR THREE
+    # 180E          # Cf       MONGOLIAN VOWEL SEPARATOR
+    # 180F          # Mn       MONGOLIAN FREE VARIATION SELECTOR FOUR
+    # 200B..200F    # Cf   [5] ZERO WIDTH SPACE..RIGHT-TO-LEFT MARK
+    # 202A..202E    # Cf   [5] LEFT-TO-RIGHT EMBEDDING..RIGHT-TO-LEFT OVERRIDE
+    # 2060..2064    # Cf   [5] WORD JOINER..INVISIBLE PLUS
+    # 2065          # Cn       <reserved-2065>
+    # 2066..206F    # Cf  [10] LEFT-TO-RIGHT ISOLATE..NOMINAL DIGIT SHAPES
+    # 3164          # Lo       HANGUL FILLER
+    # FE00..FE0F    # Mn  [16] VARIATION SELECTOR-1..VARIATION SELECTOR-16
+    # FEFF          # Cf       ZERO WIDTH NO-BREAK SPACE
+    # FFA0          # Lo       HALFWIDTH HANGUL FILLER
+    # FFF0..FFF8    # Cn   [9] <reserved-FFF0>..<reserved-FFF8>
+    # 1BCA0..1BCA3  # Cf   [4] SHORTHAND FORMAT LETTER OVERLAP..SHORTHAND FORMAT UP STEP
+    # 1D173..1D17A  # Cf   [8] MUSICAL SYMBOL BEGIN BEAM..MUSICAL SYMBOL END PHRASE
+    # E0000         # Cn       <reserved-E0000>
+    # E0001         # Cf       LANGUAGE TAG
+    # E0002..E001F  # Cn  [30] <reserved-E0002>..<reserved-E001F>
+    # E0020..E007F  # Cf  [96] TAG SPACE..CANCEL TAG
+    # E0080..E00FF  # Cn [128] <reserved-E0080>..<reserved-E00FF>
+    # E0100..E01EF  # Mn [240] VARIATION SELECTOR-17..VARIATION SELECTOR-256
+    # E01F0..E0FFF  # Cn [3600] <reserved-E01F0>..<reserved-E0FFF>
+    return (
+        u == 0x00AD
+        or u == 0x034F  # Cf       SOFT HYPHEN
+        or u == 0x061C  # Mn       COMBINING GRAPHEME JOINER
+        or 0x115F <= u <= 0x1160  # Cf       ARABIC LETTER MARK
+        or 0x17B4  # Lo   [2] HANGUL CHOSEONG FILLER..HANGUL JUNGSEONG FILLER
+        <= u
+        <= 0x17B5
+        or 0x180B  # Mn   [2] KHMER VOWEL INHERENT AQ..KHMER VOWEL INHERENT AA
+        <= u
+        <= 0x180D
+        or u  # Mn   [3] MONGOLIAN FREE VARIATION SELECTOR ONE..MONGOLIAN FREE VARIATION SELECTOR THREE
+        == 0x180E
+        or u == 0x180F  # Cf       MONGOLIAN VOWEL SEPARATOR
+        or 0x200B <= u <= 0x200F  # Mn       MONGOLIAN FREE VARIATION SELECTOR FOUR
+        or 0x202A <= u <= 0x202E  # Cf   [5] ZERO WIDTH SPACE..RIGHT-TO-LEFT MARK
+        or 0x2060  # Cf   [5] LEFT-TO-RIGHT EMBEDDING..RIGHT-TO-LEFT OVERRIDE
+        <= u
+        <= 0x2064
+        or u == 0x2065  # Cf   [5] WORD JOINER..INVISIBLE PLUS
+        or 0x2066 <= u <= 0x206F  # Cn       <reserved-2065>
+        or u == 0x3164  # Cf  [10] LEFT-TO-RIGHT ISOLATE..NOMINAL DIGIT SHAPES
+        or 0xFE00 <= u <= 0xFE0F  # Lo       HANGUL FILLER
+        or u == 0xFEFF  # Mn  [16] VARIATION SELECTOR-1..VARIATION SELECTOR-16
+        or u == 0xFFA0  # Cf       ZERO WIDTH NO-BREAK SPACE
+        or 0xFFF0 <= u <= 0xFFF8  # Lo       HALFWIDTH HANGUL FILLER
+        or 0x1BCA0 <= u <= 0x1BCA3  # Cn   [9] <reserved-FFF0>..<reserved-FFF8>
+        or 0x1D173  # Cf   [4] SHORTHAND FORMAT LETTER OVERLAP..SHORTHAND FORMAT UP STEP
+        <= u
+        <= 0x1D17A
+        or u == 0xE0000  # Cf   [8] MUSICAL SYMBOL BEGIN BEAM..MUSICAL SYMBOL END PHRASE
+        or u == 0xE0001  # Cn       <reserved-E0000>
+        or 0xE0002 <= u <= 0xE001F  # Cf       LANGUAGE TAG
+        or 0xE0020 <= u <= 0xE007F  # Cn  [30] <reserved-E0002>..<reserved-E001F>
+        or 0xE0080 <= u <= 0xE00FF  # Cf  [96] TAG SPACE..CANCEL TAG
+        or 0xE0100 <= u <= 0xE01EF  # Cn [128] <reserved-E0080>..<reserved-E00FF>
+        or 0xE01F0  # Mn [240] VARIATION SELECTOR-17..VARIATION SELECTOR-256
+        <= u
+        <= 0xE0FFF
+        or False  # Cn [3600] <reserved-E01F0>..<reserved-E0FFF>
+    )