LADP_Florence-40e / config.json
arjunanand13's picture
Create config.json
42c212c verified
raw
history blame
1.94 kB
{
"_name_or_path": "microsoft/Florence-2-base-ft",
"architectures": ["Florence2ForConditionalGeneration"],
"auto_map": {
"AutoConfig": "configuration_florence2.Florence2Config",
"AutoModelForCausalLM": "modeling_florence2.Florence2ForConditionalGeneration"
},
"bos_token_id": 2,
"eos_token_id": 1,
"ignore_index": -100,
"is_encoder_decoder": true,
"model_type": "florence2",
"pad_token_id": 0,
"projection_dim": 768,
"vocab_size": 51289,
"torch_dtype": "float32", // Keep fine-tuned dtype if trained in float32
"text_config": {
"activation_dropout": 0.1,
"activation_function": "gelu",
"attention_dropout": 0.1,
"bos_token_id": 2,
"decoder_attention_heads": 12,
"decoder_ffn_dim": 3072,
"decoder_layers": 6,
"d_model": 768,
"encoder_attention_heads": 12,
"encoder_ffn_dim": 3072,
"encoder_layers": 6,
"eos_token_id": 1,
"forced_bos_token_id": 2,
"forced_eos_token_id": 1,
"num_beams": 3,
"no_repeat_ngram_size": 3,
"dropout": 0.1,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1,
"LABEL_2": 2
},
"max_position_embeddings": 1024,
"is_encoder_decoder": true,
"pad_token_id": 0
},
"vision_config": {
"model_type": "davit",
"drop_path_rate": 0.1,
"patch_size": [7, 3, 3, 3],
"patch_stride": [4, 2, 2, 2],
"patch_padding": [3, 1, 1, 1],
"patch_prenorm": [false, true, true, true],
"dim_embed": [128, 256, 512, 1024],
"num_heads": [4, 8, 16, 32],
"num_groups": [4, 8, 16, 32],
"depths": [1, 1, 9, 1],
"window_size": 12,
"projection_dim": 768,
"visual_temporal_embedding": {
"type": "COSINE",
"max_temporal_embeddings": 100
},
"image_pos_embed": {
"type": "learned_abs_2d",
"max_pos_embeddings": 50
},
"image_feature_source": ["spatial_avg_pool", "temporal_avg_pool"]
},
"transformers_version": "4.41.0.dev0"
}