[INFO|configuration_utils.py:672] 2024-10-07 10:46:19,862 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/config.json [WARNING|modeling_rope_utils.py:379] 2024-10-07 10:46:19,863 >> Unrecognized keys in `rope_scaling` for 'rope_type'='default': {'mrope_section'} [INFO|configuration_utils.py:739] 2024-10-07 10:46:19,865 >> Model config Qwen2VLConfig { "_name_or_path": "Qwen/Qwen2-VL-2B-Instruct", "architectures": [ "Qwen2VLForConditionalGeneration" ], "attention_dropout": 0.0, "bos_token_id": 151643, "eos_token_id": 151645, "hidden_act": "silu", "hidden_size": 1536, "image_token_id": 151655, "initializer_range": 0.02, "intermediate_size": 8960, "max_position_embeddings": 32768, "max_window_layers": 28, "model_type": "qwen2_vl", "num_attention_heads": 12, "num_hidden_layers": 28, "num_key_value_heads": 2, "rms_norm_eps": 1e-06, "rope_scaling": { "mrope_section": [ 16, 24, 24 ], "rope_type": "default", "type": "default" }, "rope_theta": 1000000.0, "sliding_window": 32768, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "transformers_version": "4.45.0", "use_cache": true, "use_sliding_window": false, "video_token_id": 151656, "vision_config": { "hidden_size": 1536, "in_chans": 3, "model_type": "qwen2_vl", "spatial_patch_size": 14 }, "vision_end_token_id": 151653, "vision_start_token_id": 151652, "vision_token_id": 151654, "vocab_size": 151936 } [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:24,575 >> loading file vocab.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/vocab.json [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:24,576 >> loading file merges.txt from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/merges.txt [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:24,576 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/tokenizer.json [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:24,576 >> loading file added_tokens.json from cache at None [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:24,576 >> loading file special_tokens_map.json from cache at None [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:24,576 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/tokenizer_config.json [INFO|tokenization_utils_base.py:2478] 2024-10-07 10:46:24,943 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. [INFO|image_processing_base.py:375] 2024-10-07 10:46:25,869 >> loading configuration file preprocessor_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/preprocessor_config.json [INFO|image_processing_base.py:375] 2024-10-07 10:46:26,105 >> loading configuration file preprocessor_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/preprocessor_config.json [INFO|image_processing_base.py:429] 2024-10-07 10:46:26,105 >> Image processor Qwen2VLImageProcessor { "do_convert_rgb": true, "do_normalize": true, "do_rescale": true, "do_resize": true, "image_mean": [ 0.48145466, 0.4578275, 0.40821073 ], "image_processor_type": "Qwen2VLImageProcessor", "image_std": [ 0.26862954, 0.26130258, 0.27577711 ], "max_pixels": 12845056, "merge_size": 2, "min_pixels": 3136, "patch_size": 14, "processor_class": "Qwen2VLProcessor", "resample": 3, "rescale_factor": 0.00392156862745098, "size": { "max_pixels": 12845056, "min_pixels": 3136 }, "temporal_patch_size": 2 } [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:26,342 >> loading file vocab.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/vocab.json [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:26,342 >> loading file merges.txt from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/merges.txt [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:26,342 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/tokenizer.json [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:26,342 >> loading file added_tokens.json from cache at None [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:26,342 >> loading file special_tokens_map.json from cache at None [INFO|tokenization_utils_base.py:2214] 2024-10-07 10:46:26,342 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/tokenizer_config.json [INFO|tokenization_utils_base.py:2478] 2024-10-07 10:46:26,691 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. [INFO|processing_utils.py:744] 2024-10-07 10:46:27,998 >> Processor Qwen2VLProcessor: - image_processor: Qwen2VLImageProcessor { "do_convert_rgb": true, "do_normalize": true, "do_rescale": true, "do_resize": true, "image_mean": [ 0.48145466, 0.4578275, 0.40821073 ], "image_processor_type": "Qwen2VLImageProcessor", "image_std": [ 0.26862954, 0.26130258, 0.27577711 ], "max_pixels": 12845056, "merge_size": 2, "min_pixels": 3136, "patch_size": 14, "processor_class": "Qwen2VLProcessor", "resample": 3, "rescale_factor": 0.00392156862745098, "size": { "max_pixels": 12845056, "min_pixels": 3136 }, "temporal_patch_size": 2 } - tokenizer: Qwen2TokenizerFast(name_or_path='Qwen/Qwen2-VL-2B-Instruct', vocab_size=151643, model_max_length=32768, is_fast=True, padding_side='left', truncation_side='right', special_tokens={'eos_token': '<|im_end|>', 'pad_token': '<|endoftext|>', 'additional_special_tokens': ['<|im_start|>', '<|im_end|>', '<|object_ref_start|>', '<|object_ref_end|>', '<|box_start|>', '<|box_end|>', '<|quad_start|>', '<|quad_end|>', '<|vision_start|>', '<|vision_end|>', '<|vision_pad|>', '<|image_pad|>', '<|video_pad|>']}, clean_up_tokenization_spaces=False), added_tokens_decoder={ 151643: AddedToken("<|endoftext|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151644: AddedToken("<|im_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151645: AddedToken("<|im_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151646: AddedToken("<|object_ref_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151647: AddedToken("<|object_ref_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151648: AddedToken("<|box_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151649: AddedToken("<|box_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151650: AddedToken("<|quad_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151651: AddedToken("<|quad_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151652: AddedToken("<|vision_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151653: AddedToken("<|vision_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151654: AddedToken("<|vision_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151655: AddedToken("<|image_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), 151656: AddedToken("<|video_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), } { "processor_class": "Qwen2VLProcessor" } [INFO|configuration_utils.py:672] 2024-10-07 10:46:33,485 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/config.json [WARNING|modeling_rope_utils.py:379] 2024-10-07 10:46:33,486 >> Unrecognized keys in `rope_scaling` for 'rope_type'='default': {'mrope_section'} [INFO|configuration_utils.py:739] 2024-10-07 10:46:33,488 >> Model config Qwen2VLConfig { "_name_or_path": "Qwen/Qwen2-VL-2B-Instruct", "architectures": [ "Qwen2VLForConditionalGeneration" ], "attention_dropout": 0.0, "bos_token_id": 151643, "eos_token_id": 151645, "hidden_act": "silu", "hidden_size": 1536, "image_token_id": 151655, "initializer_range": 0.02, "intermediate_size": 8960, "max_position_embeddings": 32768, "max_window_layers": 28, "model_type": "qwen2_vl", "num_attention_heads": 12, "num_hidden_layers": 28, "num_key_value_heads": 2, "rms_norm_eps": 1e-06, "rope_scaling": { "mrope_section": [ 16, 24, 24 ], "rope_type": "default", "type": "default" }, "rope_theta": 1000000.0, "sliding_window": 32768, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "transformers_version": "4.45.0", "use_cache": true, "use_sliding_window": false, "video_token_id": 151656, "vision_config": { "hidden_size": 1536, "in_chans": 3, "model_type": "qwen2_vl", "spatial_patch_size": 14 }, "vision_end_token_id": 151653, "vision_start_token_id": 151652, "vision_token_id": 151654, "vocab_size": 151936 } [INFO|modeling_utils.py:3726] 2024-10-07 10:46:34,699 >> loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/model.safetensors.index.json [INFO|modeling_utils.py:1622] 2024-10-07 10:46:49,097 >> Instantiating Qwen2VLForConditionalGeneration model under default dtype torch.bfloat16. [INFO|configuration_utils.py:1099] 2024-10-07 10:46:49,099 >> Generate config GenerationConfig { "bos_token_id": 151643, "eos_token_id": 151645 } [WARNING|logging.py:328] 2024-10-07 10:46:49,381 >> `Qwen2VLRotaryEmbedding` can now be fully parameterized by passing the model config through the `config` argument. All other arguments will be removed in v4.46 [INFO|modeling_utils.py:4568] 2024-10-07 10:46:51,825 >> All model checkpoint weights were used when initializing Qwen2VLForConditionalGeneration. [INFO|modeling_utils.py:4576] 2024-10-07 10:46:51,825 >> All the weights of Qwen2VLForConditionalGeneration were initialized from the model checkpoint at Qwen/Qwen2-VL-2B-Instruct. If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2VLForConditionalGeneration for predictions without further training. [INFO|configuration_utils.py:1054] 2024-10-07 10:46:52,385 >> loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2-VL-2B-Instruct/snapshots/aca78372505e6cb469c4fa6a35c60265b00ff5a4/generation_config.json [INFO|configuration_utils.py:1099] 2024-10-07 10:46:52,385 >> Generate config GenerationConfig { "bos_token_id": 151643, "do_sample": true, "eos_token_id": [ 151645, 151643 ], "pad_token_id": 151643, "temperature": 0.01, "top_k": 1, "top_p": 0.001 } [INFO|trainer.py:667] 2024-10-07 10:46:52,481 >> Using auto half precision backend [INFO|trainer.py:2243] 2024-10-07 10:46:53,044 >> ***** Running training ***** [INFO|trainer.py:2244] 2024-10-07 10:46:53,044 >> Num examples = 100 [INFO|trainer.py:2245] 2024-10-07 10:46:53,044 >> Num Epochs = 3 [INFO|trainer.py:2246] 2024-10-07 10:46:53,044 >> Instantaneous batch size per device = 2 [INFO|trainer.py:2249] 2024-10-07 10:46:53,045 >> Total train batch size (w. parallel, distributed & accumulation) = 16 [INFO|trainer.py:2250] 2024-10-07 10:46:53,045 >> Gradient Accumulation steps = 8 [INFO|trainer.py:2251] 2024-10-07 10:46:53,045 >> Total optimization steps = 18 [INFO|trainer.py:2252] 2024-10-07 10:46:53,047 >> Number of trainable parameters = 1,543,714,304 [INFO|trainer.py:3705] 2024-10-07 10:47:41,126 >> Saving model checkpoint to saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18 [INFO|configuration_utils.py:407] 2024-10-07 10:47:41,128 >> Configuration saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18/config.json [INFO|configuration_utils.py:868] 2024-10-07 10:47:41,129 >> Configuration saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18/generation_config.json [INFO|modeling_utils.py:2838] 2024-10-07 10:48:08,436 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 2 checkpoint shards. You can find where each parameters has been saved in the index located at saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18/model.safetensors.index.json. [INFO|tokenization_utils_base.py:2649] 2024-10-07 10:48:08,438 >> tokenizer config file saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18/tokenizer_config.json [INFO|tokenization_utils_base.py:2658] 2024-10-07 10:48:08,439 >> Special tokens file saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18/special_tokens_map.json [INFO|image_processing_base.py:258] 2024-10-07 10:48:52,717 >> Image processor saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/checkpoint-18/preprocessor_config.json [INFO|trainer.py:2505] 2024-10-07 10:48:52,717 >> Training completed. Do not forget to share your model on huggingface.co/models =) [INFO|image_processing_base.py:258] 2024-10-07 10:48:52,720 >> Image processor saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/preprocessor_config.json [INFO|trainer.py:3705] 2024-10-07 10:48:52,720 >> Saving model checkpoint to saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38 [INFO|configuration_utils.py:407] 2024-10-07 10:48:52,723 >> Configuration saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/config.json [INFO|configuration_utils.py:868] 2024-10-07 10:48:52,723 >> Configuration saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/generation_config.json [INFO|modeling_utils.py:2838] 2024-10-07 10:49:26,625 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 2 checkpoint shards. You can find where each parameters has been saved in the index located at saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/model.safetensors.index.json. [INFO|tokenization_utils_base.py:2649] 2024-10-07 10:49:26,627 >> tokenizer config file saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/tokenizer_config.json [INFO|tokenization_utils_base.py:2658] 2024-10-07 10:49:26,628 >> Special tokens file saved in saves/Qwen2-VL-2B-Instruct/full/train_2024-10-07-10-45-38/special_tokens_map.json [INFO|modelcard.py:449] 2024-10-07 10:49:26,987 >> Dropping the following result as it does not have all the necessary fields: {'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}