Update tiny models for Blip2ForImageTextRetrieval

config.json

@@ -0,0 +1,81 @@
+{
+  "architectures": [
+    "Blip2ForImageTextRetrieval"
+  ],
+  "dtype": "float32",
+  "image_text_hidden_size": 256,
+  "image_token_index": 4,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "is_encoder_decoder": true,
+  "model_type": "blip-2",
+  "num_query_tokens": 10,
+  "qformer_config": {
+    "attention_dropout": 0.1,
+    "attention_probs_dropout_prob": 0.1,
+    "bos_token_id": 0,
+    "cross_attention_frequency": 2,
+    "dropout": 0.1,
+    "encoder_hidden_size": 32,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.1,
+    "hidden_size": 32,
+    "initializer_range": 0.02,
+    "intermediate_size": 37,
+    "layer_norm_eps": 1e-12,
+    "max_position_embeddings": 512,
+    "model_type": "blip_2_qformer",
+    "num_attention_heads": 4,
+    "num_hidden_layers": 2,
+    "pad_token_id": 0,
+    "projection_dim": 32,
+    "use_qformer_text_input": false,
+    "vocab_size": 99
+  },
+  "text_config": {
+    "bos_token_id": 2,
+    "classifier_dropout": 0.0,
+    "d_ff": 37,
+    "d_kv": 8,
+    "d_model": 32,
+    "decoder_start_token_id": 0,
+    "dense_act_fn": "relu",
+    "dropout_rate": 0.1,
+    "eos_token_id": 2,
+    "feed_forward_proj": "relu",
+    "initializer_factor": 0.002,
+    "is_decoder": false,
+    "is_gated_act": false,
+    "layer_norm_epsilon": 1e-06,
+    "model_type": "t5",
+    "num_decoder_layers": 2,
+    "num_heads": 4,
+    "num_layers": 2,
+    "pad_token_id": 1,
+    "relative_attention_max_distance": 128,
+    "relative_attention_num_buckets": 8,
+    "scale_decoder_outputs": true,
+    "tie_word_embeddings": true,
+    "use_cache": true,
+    "vocab_size": 50266
+  },
+  "transformers_version": "5.16.0.dev0",
+  "use_decoder_only_language_model": false,
+  "vision_config": {
+    "attention_dropout": 0.1,
+    "dropout": 0.1,
+    "hidden_act": "gelu",
+    "hidden_size": 32,
+    "image_size": 30,
+    "initializer_range": 1e-10,
+    "intermediate_size": 37,
+    "layer_norm_eps": 1e-06,
+    "model_type": "blip_2_vision_model",
+    "num_attention_heads": 4,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": 2,
+    "projection_dim": 32,
+    "qkv_bias": true
+  }
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b25aa7e42ea74fa552e815045d07020040b4f9a952136ffc39cde9a354714a5
+size 313600

preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "crop_size": {
+    "height": 30,
+    "width": 30
+  },
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "BlipImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 30,
+    "width": 30
+  }
+}

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "</s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "errors": "replace",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "processor_class": "Blip2Processor",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "</s>"
+}