xinference 1.4.0__py3-none-any.whl → 1.5.0__py3-none-any.whl

xinference/model/utils.py

@@ -11,17 +11,21 @@
 # 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 asyncio
 import json
 import logging
 import os
 import random
+import threading
 from json import JSONDecodeError
 from pathlib import Path
-from typing import Any, Callable, Dict, Optional, Tuple, Union
+from typing import Any, Callable, Dict, Optional, Set, Tuple, Type, Union
 
 import huggingface_hub
 import numpy as np
 import torch
+from tqdm.auto import tqdm
 
 from ..constants import (
     XINFERENCE_CACHE_DIR,
@@ -343,3 +347,116 @@ def set_all_random_seed(seed: int):
     np.random.seed(seed)
     torch.manual_seed(seed)
     torch.cuda.manual_seed_all(seed)
+
+
+class CancellableDownloader:
+    def __init__(
+        self,
+        cancel_error_cls: Type[BaseException] = asyncio.CancelledError,
+        cancelled_event: Optional[threading.Event] = None,
+    ):
+        self._cancelled = cancelled_event
+        if self._cancelled is None:
+            self._cancelled = threading.Event()
+        self._done_event = threading.Event()
+        self._cancel_error_cls = cancel_error_cls
+        self._original_update = None
+        # progress for tqdm that is main
+        self._main_progresses: Set[tqdm] = set()
+        # progress for file downloader
+        # mainly when tqdm unit is set
+        self._download_progresses: Set[tqdm] = set()
+        # tqdm original update
+        self._original_tqdm_update = None
+
+    def reset(self):
+        self._main_progresses.clear()
+        self._download_progresses.clear()
+
+    def get_progress(self) -> float:
+        if self.cancelled or self.done:
+            # directly return 1.0 when cancelled or finished
+            return 1.0
+
+        tasks = finished_tasks = 0
+        for main_progress in self._main_progresses:
+            tasks += main_progress.total or 0
+            finished_tasks += main_progress.n
+
+        if tasks == 0:
+            # we assumed at least 1 task
+            tasks = 1
+
+        finished_ratio = finished_tasks / tasks
+
+        all_download_progress = finished_download_progress = 0
+        for download_progress in self._download_progresses:
+            # we skip finished download
+            if download_progress.n == download_progress.total:
+                continue
+            all_download_progress += download_progress.total or (
+                download_progress.n * 10
+            )
+            finished_download_progress += download_progress.n
+
+        if all_download_progress > 0:
+            rest_ratio = (
+                (tasks - finished_tasks)
+                / tasks
+                * (finished_download_progress / all_download_progress)
+            )
+            return finished_ratio + rest_ratio
+        else:
+            return finished_ratio
+
+    def cancel(self):
+        self._cancelled.set()
+
+    @property
+    def cancelled(self):
+        return self._cancelled.is_set()
+
+    @property
+    def done(self):
+        return self._done_event.is_set()
+
+    def wait(self, timeout: float):
+        self._done_event.wait(timeout)
+
+    def raise_error(self, error_msg: str = "Download cancelled"):
+        raise self._cancel_error_cls(error_msg)
+
+    def patch_tqdm(self):
+        # patch tqdm
+        # raise error if cancelled
+        self._original_update = original_update = tqdm.update
+        downloader = self
+
+        def patched_update(self, n):
+            if downloader.cancelled:
+                downloader.raise_error()
+            if not self.disable:
+                progresses = (
+                    downloader._main_progresses
+                    if getattr(self, "unit", "it") == "it"
+                    else downloader._download_progresses
+                )
+                progresses.add(self)
+            return original_update(self, n)
+
+        tqdm.update = patched_update
+
+    def unpatch_tqdm(self):
+        from tqdm.auto import tqdm
+
+        if self._original_update:
+            tqdm.update = self._original_update
+
+    def __enter__(self):
+        self.patch_tqdm()
+        return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        self.unpatch_tqdm()
+        self._done_event.set()
+        self.reset()

xinference/model/video/core.py

@@ -17,7 +17,7 @@ from collections import defaultdict
 from typing import Any, Dict, List, Literal, Optional, Tuple
 
 from ...constants import XINFERENCE_CACHE_DIR
-from ..core import CacheableModelSpec, ModelDescription
+from ..core import CacheableModelSpec, ModelDescription, VirtualEnvSettings
 from ..utils import valid_model_revision
 from .diffusers import DiffUsersVideoModel
 
@@ -44,6 +44,7 @@ class VideoModelFamilyV1(CacheableModelSpec):
     model_ability: Optional[List[str]]
     default_model_config: Optional[Dict[str, Any]]
     default_generate_config: Optional[Dict[str, Any]]
+    virtualenv: Optional[VirtualEnvSettings]
 
 
 class VideoModelDescription(ModelDescription):
@@ -57,6 +58,10 @@ class VideoModelDescription(ModelDescription):
         super().__init__(address, devices, model_path=model_path)
         self._model_spec = model_spec
 
+    @property
+    def spec(self):
+        return self._model_spec
+
     def to_dict(self):
         return {
             "model_type": "video",

xinference/thirdparty/deepseek_vl2/__init__.py

@@ -0,0 +1,31 @@
+# Copyright (c) 2023-2024 DeepSeek.
+#
+# Permission is hereby granted, free of charge, to any person obtaining a copy of
+# this software and associated documentation files (the "Software"), to deal in
+# the Software without restriction, including without limitation the rights to
+# use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
+# the Software, and to permit persons to whom the Software is furnished to do so,
+# subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+# FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+# COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
+# IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+# CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+
+# check if python version is above 3.10
+import sys
+
+if sys.version_info >= (3, 10):
+    print("Python version is above 3.10, patching the collections module.")
+    # Monkey patch collections
+    import collections
+    import collections.abc
+
+    for type_name in collections.abc.__all__:
+        setattr(collections, type_name, getattr(collections.abc, type_name))

xinference/thirdparty/deepseek_vl2/models/__init__.py

@@ -0,0 +1,26 @@
+# Copyright (c) 2023-2024 DeepSeek.
+#
+# Permission is hereby granted, free of charge, to any person obtaining a copy of
+# this software and associated documentation files (the "Software"), to deal in
+# the Software without restriction, including without limitation the rights to
+# use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
+# the Software, and to permit persons to whom the Software is furnished to do so,
+# subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+# FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+# COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
+# IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+# CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+from .processing_deepseek_vl_v2 import DeepseekVLV2Processor
+from .modeling_deepseek_vl_v2 import DeepseekVLV2ForCausalLM
+
+__all__ = [
+    "DeepseekVLV2Processor",
+    "DeepseekVLV2ForCausalLM",
+]

xinference/thirdparty/deepseek_vl2/models/configuration_deepseek.py

@@ -0,0 +1,210 @@
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+DEEPSEEK_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
+class DeepseekV2Config(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`DeepseekV2Model`]. It is used to instantiate an DeepSeek
+    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+    defaults will yield a similar configuration to that of the DeepSeek-V2 with multi-latent attention.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 102400):
+            Vocabulary size of the Deep model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`DeepseekV2Model`]
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 11008):
+            Dimension of the MLP representations.
+        moe_intermediate_size (`int`, *optional*, defaults to 1407):
+            Dimension of the MoE representations.
+        num_hidden_layers (`int`, *optional*, defaults to 32):
+            Number of hidden layers in the Transformer decoder.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        n_shared_experts (`int`, *optional*, defaults to None):
+            Number of shared experts, None means dense model.
+        n_routed_experts (`int`, *optional*, defaults to None):
+            Number of routed experts, None means dense model.
+        routed_scaling_factor (`float`, *optional*, defaults to 1.0):
+            Scaling factor or routed experts.
+        topk_method (`str`, *optional*, defaults to `gready`):
+            Topk method used in routed gate.
+        n_group (`int`, *optional*, defaults to None):
+            Number of groups for routed experts.
+        topk_group (`int`, *optional*, defaults to None):
+            Number of selected groups for each token(for each token, ensuring the selected experts is only within `topk_group` groups).
+        num_experts_per_tok (`int`, *optional*, defaults to None):
+            Number of selected experts, None means dense model.
+        moe_layer_freq (`int`, *optional*, defaults to 1):
+            The frequency of the MoE layer: one expert layer for every `moe_layer_freq - 1` dense layers.
+        first_k_dense_replace (`int`, *optional*, defaults to 0):
+            Number of dense layers in shallow layers(embed->dense->dense->...->dense->moe->moe...->lm_head).
+                                                            \--k dense layers--/
+        norm_topk_prob (`bool`, *optional*, defaults to False):
+            Whether to normalize the weights of the routed experts.
+        scoring_func (`str`, *optional*, defaults to 'softmax'):
+            Method of computing expert weights.
+        aux_loss_alpha (`float`, *optional*, defaults to 0.001):
+            Auxiliary loss weight coefficient.
+        seq_aux = (`bool`, *optional*, defaults to True):
+            Whether to compute the auxiliary loss for each individual sample.
+        num_key_value_heads (`int`, *optional*):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details checkout [this
+            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
+            `num_attention_heads`.
+        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
+            The non-linear activation function (function or string) in the decoder.
+        max_position_embeddings (`int`, *optional*, defaults to 2048):
+            The maximum sequence length that this model might ever be used with.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
+            The epsilon used by the rms normalization layers.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        pad_token_id (`int`, *optional*):
+            Padding token id.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            Beginning of stream token id.
+        eos_token_id (`int`, *optional*, defaults to 2):
+            End of stream token id.
+        pretraining_tp (`int`, *optional*, defaults to 1):
+            Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
+            document](https://huggingface.co/docs/transformers/parallelism) to understand more about it. This value is
+            necessary to ensure exact reproducibility of the pretraining results. Please refer to [this
+            issue](https://github.com/pytorch/pytorch/issues/76232).
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether to tie weight embeddings
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        rope_scaling (`Dict`, *optional*):
+            Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
+            strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
+            `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
+            `max_position_embeddings` to the expected new maximum.
+        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
+            Whether to use a bias in the query, key, value and output projection layers during self-attention.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        use_mla (`bool`, *optional*, defaults to `True`): Use multi-latent attention or multi-head attention. If True,
+            the model will use multi-latent attention, otherwise, it will use multi-head attention.
+
+    ```python
+    >>> from transformers import DeepseekV2Model, DeepseekV2Config
+
+    >>> # Initializing a Deepseek-V2 style configuration
+    >>> configuration = DeepseekV2Config()
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "deepseek_v2"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=102400,
+        hidden_size=4096,
+        intermediate_size=11008,
+        moe_intermediate_size = 1407,
+        num_hidden_layers=30,
+        num_attention_heads=32,
+        num_key_value_heads=32,
+        n_shared_experts = None,
+        n_routed_experts = None,
+        ep_size = 1,
+        routed_scaling_factor = 1.0,
+        kv_lora_rank = 512,
+        q_lora_rank = 1536,
+        qk_rope_head_dim = 64,
+        v_head_dim = 128,
+        qk_nope_head_dim = 128,
+        topk_method = 'gready',
+        n_group = None,
+        topk_group = None,
+        num_experts_per_tok = None,
+        moe_layer_freq = 1,
+        first_k_dense_replace = 0,
+        norm_topk_prob = False,
+        scoring_func = 'softmax',
+        aux_loss_alpha = 0.001,
+        seq_aux = True,
+        hidden_act="silu",
+        max_position_embeddings=2048,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        pad_token_id=None,
+        bos_token_id=100000,
+        eos_token_id=100001,
+        pretraining_tp=1,
+        tie_word_embeddings=False,
+        rope_theta=10000.0,
+        rope_scaling=None,
+        attention_bias=False,
+        attention_dropout=0.0,
+        use_mla=True,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.moe_intermediate_size = moe_intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.n_shared_experts = n_shared_experts
+        self.n_routed_experts = n_routed_experts
+        self.ep_size = ep_size
+        self.routed_scaling_factor = routed_scaling_factor
+        self.kv_lora_rank = kv_lora_rank
+        self.q_lora_rank = q_lora_rank
+        self.qk_rope_head_dim = qk_rope_head_dim
+        self.v_head_dim = v_head_dim
+        self.qk_nope_head_dim = qk_nope_head_dim
+        self.topk_method = topk_method
+        self.n_group = n_group
+        self.topk_group = topk_group
+        self.num_experts_per_tok = num_experts_per_tok
+        self.moe_layer_freq = moe_layer_freq
+        self.first_k_dense_replace = first_k_dense_replace
+        self.norm_topk_prob = norm_topk_prob
+        self.scoring_func = scoring_func
+        self.aux_loss_alpha = aux_loss_alpha
+        self.seq_aux = seq_aux
+        # for backward compatibility
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = float(rms_norm_eps)
+        self.pretraining_tp = pretraining_tp
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+        self.use_mla = use_mla
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )