Update tiny models for PerceptionLMForConditionalGeneration

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

chat_template.jinja

@@ -0,0 +1,3 @@
+{{- bos_token }}{%- if messages[0]['role'] == 'system' -%}    {%- set system_message = messages[0]['content']|trim %}
+    {%- set messages = messages[1:] %}
+{%- else %}    {%- set system_message = 'You are a helpful language and vision assistant. You are able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.' %}{%- endif %}{{- '<|start_header_id|>system<|end_header_id|>\n\n' }}{{- system_message }}{{- '<|eot_id|>' }}{%- for message in messages %}{{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n' }}{%- for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<|image|>' }}{%- endfor %}{%- for content in message['content'] | selectattr('type', 'equalto', 'video') %}{{ '<|video|>' }}{%- endfor %}{%- for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{- content['text'] | trim }}{%- endfor %}{{'<|eot_id|>' }}{%- endfor %}{%- if add_generation_prompt %}{{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{%- endif %}

config.json

@@ -0,0 +1,86 @@
+{
+  "architectures": [
+    "PerceptionLMForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "image_token_id": 0,
+  "model_type": "perception_lm",
+  "projector_pooling_ratio": 1,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attention_probs_dropout_prob": 0.1,
+    "bos_token_id": 128000,
+    "eos_token_id": 128009,
+    "head_dim": 8,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.1,
+    "hidden_size": 32,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1",
+      "2": "LABEL_2"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 37,
+    "is_training": true,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1,
+      "LABEL_2": 2
+    },
+    "max_position_embeddings": 512,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 4,
+    "num_choices": 4,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 4,
+    "pad_token_id": 128001,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    },
+    "seq_length": 7,
+    "tie_word_embeddings": false,
+    "type_sequence_label_size": 2,
+    "type_vocab_size": 16,
+    "use_cache": true,
+    "use_input_mask": true,
+    "use_labels": true,
+    "use_token_type_ids": false,
+    "vocab_size": 128256
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.16.0.dev0",
+  "video_token_id": 2,
+  "vision_config": {
+    "architecture": "vit_pe_core_large_patch14_336",
+    "do_pooling": true,
+    "initializer_range": 0.02,
+    "label_names": [
+      "LABEL_0",
+      "LABEL_1"
+    ],
+    "model_args": {
+      "depth": 2,
+      "embed_dim": 64,
+      "global_pool": "",
+      "img_size": [
+        14,
+        14
+      ],
+      "init_values": 0.1,
+      "ref_feat_shape": [
+        1,
+        1
+      ],
+      "use_post_transformer_norm": false
+    },
+    "model_type": "timm_wrapper",
+    "num_classes": 2
+  },
+  "vision_use_cls_token": true
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "eos_token_id": 128009,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 128001,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db332541daf75b31afaa437875ed78f61bb7f5c9d897cf6fc7b5e3625583eb5b
+size 17050632

preprocessor_config.json

@@ -0,0 +1,30 @@
+{
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "do_center_crop": false,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "PerceptionLMImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_frame_tiles": 1,
+  "max_num_tiles": 36,
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 448,
+    "width": 448
+  },
+  "tile_size": 448,
+  "vision_input_type": "thumb+tile"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab0d969ef8cfbf1d640debe30a97c3fb11caf9b24b75fa89fbd4e9df1d51cac6
+size 17209923

tokenizer_config.json

@@ -0,0 +1,22 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "image_token": "<|image|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 11520,
+  "model_specific_special_tokens": {
+    "image_token": "<|image|>",
+    "video_token": "<|video|>"
+  },
+  "pad_token": "<|end_of_text|>",
+  "processor_class": "PerceptionLMProcessor",
+  "tokenizer_class": "TokenizersBackend",
+  "video_token": "<|video|>"
+}

video_preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "do_center_crop": false,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_metadata": false,
+  "size": {
+    "height": 448,
+    "width": 448
+  },
+  "video_processor_type": "PerceptionLMVideoProcessor"
+}