Upload folder using huggingface_hub

README.md

@@ -0,0 +1,255 @@
+---
+library_name: transformers
+base_model:
+- zai-org/GLM-OCR
+---
+
+This tiny model is intended for debugging. It is randomly initialized using the configuration adapted from [zai-org/GLM-OCR](https://huggingface.co/zai-org/GLM-OCR).
+
+| File path | Size |
+|------|------|
+| model.safetensors | 3.8MB |
+
+
+### Example usage:
+
+```python
+import torch
+from transformers import AutoModelForImageTextToText, AutoProcessor
+
+model_id = "tiny-random/glm-ocr"
+model = AutoModelForImageTextToText.from_pretrained(
+    model_id, dtype=torch.bfloat16, device_map="cuda",
+)
+processor = AutoProcessor.from_pretrained(model_id)
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {
+                "type": "image",
+                "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg",
+            },
+            {"type": "text", "text": "Describe this image."},
+        ],
+    }
+]
+
+# Preparation for inference
+inputs = processor.apply_chat_template(
+    messages,
+    tokenize=True,
+    add_generation_prompt=True,
+    return_dict=True,
+    return_tensors="pt"
+).to(model.device)
+inputs.pop("token_type_ids", None)
+generated_ids = model.generate(**inputs, max_new_tokens=16)
+output_text = processor.decode(generated_ids[0], skip_special_tokens=False)
+print(output_text)
+```
+
+### Codes to create this repo:
+
+<details>
+<summary>Click to expand</summary>
+
+```python
+import json
+from copy import deepcopy
+from pathlib import Path
+
+import accelerate
+import torch
+import torch.nn as nn
+from huggingface_hub import file_exists, hf_hub_download
+from transformers import (
+    AutoConfig,
+    AutoModelForCausalLM,
+    AutoProcessor,
+    GenerationConfig,
+    GlmOcrForConditionalGeneration,
+    set_seed,
+)
+
+source_model_id = "zai-org/GLM-OCR"
+save_folder = "/tmp/tiny-random/glm-ocr"
+
+processor = AutoProcessor.from_pretrained(
+    source_model_id, trust_remote_code=True)
+processor.save_pretrained(save_folder)
+
+with open(hf_hub_download(source_model_id, filename='config.json', repo_type='model'), 'r', encoding='utf-8') as f:
+    config_json: dict = json.load(f)
+
+config_json['text_config'].update({
+    "head_dim": 32,
+    "hidden_size": 8,
+    "intermediate_size": 64,
+    "num_attention_heads": 8,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 4,
+    "rope_parameters": {
+        "rope_type": "default",
+        "mrope_section": [4, 4, 8],
+        "partial_rotary_factor": 1.0,
+        "rope_theta": 10000,
+    },
+})
+config_json['vision_config'].update({
+    "hidden_size": 32,
+    "depth": 2,
+    "num_heads": 1,
+    "intermediate_size": 64,
+    "out_hidden_size": 8,
+})
+with open(f"{save_folder}/config.json", "w", encoding='utf-8') as f:
+    json.dump(config_json, f, indent=2)
+
+config = AutoConfig.from_pretrained(
+    save_folder,
+    trust_remote_code=True,
+)
+print(config)
+torch.set_default_dtype(torch.bfloat16)
+model = GlmOcrForConditionalGeneration(config)
+torch.set_default_dtype(torch.float32)
+
+if file_exists(filename="generation_config.json", repo_id=source_model_id, repo_type='model'):
+    model.generation_config = GenerationConfig.from_pretrained(
+        source_model_id, trust_remote_code=True,
+    )
+    model.generation_config.do_sample = True
+    print(model.generation_config)
+
+model = model.cpu()
+set_seed(42)
+n_params = sum(p.numel() for p in model.parameters())
+with torch.no_grad():
+    for name, p in sorted(model.named_parameters()):
+        torch.nn.init.normal_(p, 0, 0.1)
+        print(name, p.shape, p.numel() / n_params * 100, '%')
+# MTP
+set_seed(42)
+config = config.get_text_config()
+model.model.language_model.layers.append(nn.ModuleDict(dict(
+    shared_head=nn.ModuleDict(dict(
+        norm=nn.RMSNorm(config.hidden_size),
+        head=deepcopy(model.model.language_model.embed_tokens),
+    )),
+    embed_tokens=deepcopy(model.model.language_model.embed_tokens),
+    eh_proj=nn.Linear(config.hidden_size * 2,
+                      config.hidden_size, bias=False),
+    enorm=nn.RMSNorm(config.hidden_size),
+    hnorm=nn.RMSNorm(config.hidden_size),
+    input_layernorm=nn.RMSNorm(config.hidden_size),
+    post_attention_layernorm=nn.RMSNorm(config.hidden_size),
+    post_self_attn_layernorm=nn.RMSNorm(config.hidden_size),
+    self_attn=deepcopy(model.model.language_model.layers[1].self_attn),
+    mlp=deepcopy(model.model.language_model.layers[1].mlp),
+)))
+# for i in range(1, len(model.model.language_model.layers)):
+#     model.model.language_model.layers[i].mlp.gate.e_score_correction_bias = torch.rand_like(
+#         model.model.language_model.layers[i].mlp.gate.e_score_correction_bias).float()
+model.save_pretrained(save_folder)
+print(model)
+```
+
+</details>
+
+### Printing the model:
+
+<details><summary>Click to expand</summary>
+
+```text
+GlmOcrForConditionalGeneration(
+  (model): GlmOcrModel(
+    (visual): GlmOcrVisionModel(
+      (patch_embed): GlmOcrVisionPatchEmbed(
+        (proj): Conv3d(3, 32, kernel_size=(2, 14, 14), stride=(2, 14, 14))
+      )
+      (rotary_pos_emb): GlmOcrVisionRotaryEmbedding()
+      (blocks): ModuleList(
+        (0-1): 2 x GlmOcrVisionBlock(
+          (norm1): GlmOcrRMSNorm((32,), eps=1e-05)
+          (norm2): GlmOcrRMSNorm((32,), eps=1e-05)
+          (attn): GlmOcrVisionAttention(
+            (qkv): Linear(in_features=32, out_features=96, bias=True)
+            (proj): Linear(in_features=32, out_features=32, bias=True)
+            (q_norm): GlmOcrRMSNorm((32,), eps=1e-05)
+            (k_norm): GlmOcrRMSNorm((32,), eps=1e-05)
+          )
+          (mlp): GlmOcrVisionMlp(
+            (gate_proj): Linear(in_features=32, out_features=64, bias=True)
+            (up_proj): Linear(in_features=32, out_features=64, bias=True)
+            (down_proj): Linear(in_features=64, out_features=32, bias=True)
+            (act_fn): SiLUActivation()
+          )
+        )
+      )
+      (merger): GlmOcrVisionPatchMerger(
+        (proj): Linear(in_features=8, out_features=8, bias=False)
+        (post_projection_norm): LayerNorm((8,), eps=1e-05, elementwise_affine=True)
+        (gate_proj): Linear(in_features=8, out_features=24, bias=False)
+        (up_proj): Linear(in_features=8, out_features=24, bias=False)
+        (down_proj): Linear(in_features=24, out_features=8, bias=False)
+        (act1): GELU(approximate='none')
+        (act_fn): SiLUActivation()
+      )
+      (downsample): Conv2d(32, 8, kernel_size=(2, 2), stride=(2, 2))
+      (post_layernorm): GlmOcrRMSNorm((32,), eps=1e-05)
+    )
+    (language_model): GlmOcrTextModel(
+      (embed_tokens): Embedding(59392, 8, padding_idx=59246)
+      (layers): ModuleList(
+        (0-1): 2 x GlmOcrTextDecoderLayer(
+          (self_attn): GlmOcrTextAttention(
+            (q_proj): Linear(in_features=8, out_features=256, bias=False)
+            (k_proj): Linear(in_features=8, out_features=128, bias=False)
+            (v_proj): Linear(in_features=8, out_features=128, bias=False)
+            (o_proj): Linear(in_features=256, out_features=8, bias=False)
+          )
+          (mlp): GlmOcrTextMLP(
+            (gate_up_proj): Linear(in_features=8, out_features=128, bias=False)
+            (down_proj): Linear(in_features=64, out_features=8, bias=False)
+            (activation_fn): SiLUActivation()
+          )
+          (input_layernorm): GlmOcrRMSNorm((8,), eps=1e-05)
+          (post_attention_layernorm): GlmOcrRMSNorm((8,), eps=1e-05)
+          (post_self_attn_layernorm): GlmOcrRMSNorm((8,), eps=1e-05)
+          (post_mlp_layernorm): GlmOcrRMSNorm((8,), eps=1e-05)
+        )
+        (2): ModuleDict(
+          (shared_head): ModuleDict(
+            (norm): RMSNorm((8,), eps=None, elementwise_affine=True)
+            (head): Embedding(59392, 8, padding_idx=59246)
+          )
+          (embed_tokens): Embedding(59392, 8, padding_idx=59246)
+          (eh_proj): Linear(in_features=16, out_features=8, bias=False)
+          (enorm): RMSNorm((8,), eps=None, elementwise_affine=True)
+          (hnorm): RMSNorm((8,), eps=None, elementwise_affine=True)
+          (input_layernorm): RMSNorm((8,), eps=None, elementwise_affine=True)
+          (post_attention_layernorm): RMSNorm((8,), eps=None, elementwise_affine=True)
+          (post_self_attn_layernorm): RMSNorm((8,), eps=None, elementwise_affine=True)
+          (self_attn): GlmOcrTextAttention(
+            (q_proj): Linear(in_features=8, out_features=256, bias=False)
+            (k_proj): Linear(in_features=8, out_features=128, bias=False)
+            (v_proj): Linear(in_features=8, out_features=128, bias=False)
+            (o_proj): Linear(in_features=256, out_features=8, bias=False)
+          )
+          (mlp): GlmOcrTextMLP(
+            (gate_up_proj): Linear(in_features=8, out_features=128, bias=False)
+            (down_proj): Linear(in_features=64, out_features=8, bias=False)
+            (activation_fn): SiLUActivation()
+          )
+        )
+      )
+      (norm): GlmOcrRMSNorm((8,), eps=1e-05)
+      (rotary_emb): GlmOcrTextRotaryEmbedding()
+    )
+  )
+  (lm_head): Linear(in_features=8, out_features=59392, bias=False)
+)
+```
+
+</details>

chat_template.jinja

@@ -0,0 +1,140 @@
+[gMASK]<sop>
+{%- if tools -%}
+<|system|>
+# Tools
+
+You may call one or more functions to assist with the user query.
+
+You are provided with function signatures within <tools></tools> XML tags:
+<tools>
+{% for tool in tools %}
+{{ tool | tojson(ensure_ascii=False) }}
+{% endfor %}
+</tools>
+
+For each function call, output the function name and arguments within the following XML format:
+<tool_call>{function-name}
+<arg_key>{arg-key-1}</arg_key>
+<arg_value>{arg-value-1}</arg_value>
+<arg_key>{arg-key-2}</arg_key>
+<arg_value>{arg-value-2}</arg_value>
+...
+</tool_call>{%- endif -%}
+{%- macro visible_text(content) -%}
+    {%- if content is string -%}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping -%}
+        {%- for item in content -%}
+            {%- if item is mapping and item.type == 'text' -%}
+                {{- item.text }}
+            {%- elif item is mapping and (item.type == 'image' or 'image' in item) -%}
+                <|begin_of_image|><|image|><|end_of_image|>
+            {%- elif item is mapping and (item.type == 'video' or 'video' in item) -%}
+                <|begin_of_video|><|video|><|end_of_video|>
+            {%- elif item is string -%}
+                {{- item }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{- content }}
+    {%- endif -%}
+{%- endmacro -%}
+{%- set ns = namespace(last_user_index=-1) %}
+{%- for m in messages %}
+    {%- if m.role == 'user' %}
+        {% set ns.last_user_index = loop.index0 -%}
+    {%- endif %}
+{%- endfor %}
+{% for m in messages %}
+{%- if m.role == 'user' -%}<|user|>
+{% if m.content is string %}
+{{ m.content }}
+{%- else %}
+{%- for item in m.content %}
+{% if item.type == 'video' or 'video' in item %}
+<|begin_of_video|><|video|><|end_of_video|>{% elif item.type == 'image' or 'image' in item %}
+<|begin_of_image|><|image|><|end_of_image|>{% elif item.type == 'text' %}
+{{ item.text }}
+{%- endif %}
+{%- endfor %}
+{%- endif %}
+{{- '/nothink' if (enable_thinking is defined and not enable_thinking and not visible_text(m.content).endswith("/nothink")) else '' -}}
+{%- elif m.role == 'assistant' -%}
+<|assistant|>
+{%- set reasoning_content = '' %}
+{%- set content = visible_text(m.content) %}
+{%- if m.reasoning_content is string %}
+    {%- set reasoning_content = m.reasoning_content %}
+{%- else %}
+    {%- if '</think>' in content %}
+        {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+        {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+    {%- endif %}
+{%- endif %}
+{%- if loop.index0 > ns.last_user_index and reasoning_content -%}
+{{ '\n<think>' + reasoning_content.strip() +  '</think>'}}
+{%- else -%}
+{{ '\n<think></think>' }}
+{%- endif -%}
+{%- if content.strip() -%}
+{{ '\n' + content.strip() }}
+{%- endif -%}
+{% if m.tool_calls %}
+{% for tc in m.tool_calls %}
+{%- if tc.function %}
+    {%- set tc = tc.function %}
+{%- endif %}
+{{ '\n<tool_call>' + tc.name }}
+{% set _args = tc.arguments %}
+{% for k, v in _args.items() %}
+<arg_key>{{ k }}</arg_key>
+<arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>
+{% endfor %}
+</tool_call>{% endfor %}
+{% endif %}
+{%- elif m.role == 'tool' -%}
+{%- if m.content is string -%}
+{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+    {{- '<|observation|>' }}
+{%- endif %}
+{{- '\n<tool_response>\n' }}
+{{- m.content }}
+{{- '\n</tool_response>' }}
+{% elif m.content is iterable and m.content is not mapping %}
+{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+{{- '<|observation|>' }}
+{%- endif %}
+{{- '\n<tool_response>\n' }}
+{%- for tr in m.content -%}
+  {%- if tr is mapping and tr.type is defined -%}
+    {%- set t = tr.type | lower -%}
+    {%- if t == 'text' and tr.text is defined -%}
+{{ tr.text }}
+    {%- elif t in ['image', 'image_url'] -%}
+<|begin_of_image|><|image|><|end_of_image|>
+    {%- elif t in ['video', 'video_url'] -%}
+<|begin_of_video|><|video|><|end_of_video|>
+    {%- else -%}
+{{ tr | tojson(ensure_ascii=False) }}
+    {%- endif -%}
+  {%- else -%}
+{{ tr.output if tr.output is defined else tr }}
+  {%- endif -%}
+{%- endfor -%}
+{{- '\n</tool_response>' }}
+{%- else -%}
+<|observation|>{% for tr in m.content %}
+
+<tool_response>
+{{ tr.output if tr.output is defined else tr }}
+</tool_response>{% endfor -%}
+{% endif -%}
+{%- elif m.role == 'system' -%}
+<|system|>
+{{ visible_text(m.content) }}
+{%- endif -%}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+<|assistant|>
+{{'<think></think>\n' if (enable_thinking is defined and not enable_thinking) else ''}}
+{%- endif -%}

config.json

@@ -0,0 +1,69 @@
+{
+  "architectures": [
+    "GlmOcrForConditionalGeneration"
+  ],
+  "dtype": "bfloat16",
+  "image_end_token_id": 59257,
+  "image_start_token_id": 59256,
+  "image_token_id": 59280,
+  "model_type": "glm_ocr",
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "dtype": "bfloat16",
+    "eos_token_id": [
+      59246,
+      59253
+    ],
+    "head_dim": 32,
+    "hidden_act": "silu",
+    "hidden_size": 8,
+    "initializer_range": 0.02,
+    "intermediate_size": 64,
+    "max_position_embeddings": 131072,
+    "model_type": "glm_ocr_text",
+    "num_attention_heads": 8,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 4,
+    "num_nextn_predict_layers": 1,
+    "pad_token_id": 59246,
+    "rms_norm_eps": 1e-05,
+    "rope_parameters": {
+      "mrope_section": [
+        4,
+        4,
+        8
+      ],
+      "partial_rotary_factor": 1.0,
+      "rope_theta": 10000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 59392
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.2.0.dev0",
+  "video_end_token_id": 59259,
+  "video_start_token_id": 59258,
+  "video_token_id": 59281,
+  "vision_config": {
+    "attention_bias": true,
+    "attention_dropout": 0.0,
+    "depth": 2,
+    "hidden_act": "silu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 32,
+    "image_size": 336,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 64,
+    "model_type": "glm_ocr_vision",
+    "num_heads": 1,
+    "out_hidden_size": 8,
+    "patch_size": 14,
+    "rms_norm_eps": 1e-05,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  }
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "do_sample": true,
+  "eos_token_id": [
+    59246,
+    59253
+  ],
+  "pad_token_id": 59246,
+  "transformers_version": "5.2.0.dev0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1887b001e63419d877771c4e5a93fdb1738b31b16b16fb511dc637e12b9b7a4
+size 3977976

processor_config.json

@@ -0,0 +1,67 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.48145466,
+      0.4578275,
+      0.40821073
+    ],
+    "image_processor_type": "Glm46VImageProcessorFast",
+    "image_std": [
+      0.26862954,
+      0.26130258,
+      0.27577711
+    ],
+    "merge_size": 2,
+    "patch_size": 14,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 9633792,
+      "shortest_edge": 12544
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Glm46VProcessor",
+  "video_processor": {
+    "data_format": "channels_first",
+    "default_to_square": true,
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.48145466,
+      0.4578275,
+      0.40821073
+    ],
+    "image_processor_type": "Glm46VImageProcessor",
+    "image_std": [
+      0.26862954,
+      0.26130258,
+      0.27577711
+    ],
+    "max_duration": 300,
+    "max_image_size": {
+      "longest_edge": 47040000
+    },
+    "merge_size": 2,
+    "num_frames": 16,
+    "patch_size": 14,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 9633792,
+      "shortest_edge": 12544
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Glm46VVideoProcessor"
+  }
+}

tokenizer_config.json

@@ -0,0 +1,49 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>",
+    "<|begin_of_audio|>",
+    "<|end_of_audio|>",
+    "<|begin_of_transcription|>",
+    "<|end_of_transcription|>",
+    "<|code_prefix|>",
+    "<|code_middle|>",
+    "<|code_suffix|>",
+    "<think>",
+    "</think>",
+    "<tool_call>",
+    "</tool_call>",
+    "<tool_response>",
+    "</tool_response>",
+    "<arg_key>",
+    "</arg_key>",
+    "<arg_value>",
+    "</arg_value>",
+    "/nothink",
+    "<|begin_of_box|>",
+    "<|end_of_box|>",
+    "<|image|>",
+    "<|video|>"
+  ],
+  "is_local": false,
+  "model_max_length": 655380,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "Glm46VProcessor",
+  "tokenizer_class": "TokenizersBackend"
+}