Remove image_size

preprocessor_config.json

@@ -1,53 +1,51 @@
 {
-		
-	"cache_dir":"/tmp",
-        "downstream_fusion":false,
-        "downstream_fusion_layers":1,
-        "downstream_fusion_method":"elmo",
-        "drop_rate":0.1,
-        "freeze_RoBERTa":false,
-        "freeze_ViT":false,
-        "freeze_layer_count_roberta":false,
-        "freeze_layer_count_vit":false,
-        "head_hidden_scale":2,
-        "hidden_size":768,
-        "image_size":288,
-        "input_text_embed_size":768,
-        "link_tower_shared":false,
-        "link_tower_type":"add",
-        "log_dir":"log_dir",
-        "loss_names":{"contras": 0,
-                "irtr": 0,
-                "itm": 0,
-                "mlm": 0,
-                "mpp": 0,
-                "nlvr2": 0,
-                "snli": 0,
-                "vcr": 0,
-                "vcr_qar": 0,
-                "vqa": 1},
-        "max_text_len":50,
-        "mlp_ratio":4,
-        "model_type":"bridgetower",
-        "num_heads":12,
-        "num_layers":6,
-        "num_nodes":1,
-        "only_load_cross_modal_from_meter":false,
-        "patch_size":16,
-        "resolution_before":224,
-        "stop_gradient":false,
-        "task_head_layers":2,
-        "test_only":false,
-        "tokenizer":"roberta-base",
-        "unfreeze_RoBERTa_attention":false,
-        "unfreeze_RoBERTa_embeddings":false,
-        "unfreeze_RoBERTa_encoder":false,
-        "unfreeze_RoBERTa_layernorm":false,
-        "unfreeze_ViT_attention":false,
-        "unfreeze_ViT_layernorm":false,
-        "vit":"ViT-B/16",
-        "vit_layernorm_init_from_vit":false,
-        "vit_layernorm_shared":true,
-        "vit_remove_last":false,
-        "vocab_size":50265
+    "downstream_fusion":false,
+    "downstream_fusion_layers":1,
+    "downstream_fusion_method":"elmo",
+    "drop_rate":0.1,
+    "freeze_RoBERTa":false,
+    "freeze_ViT":false,
+    "freeze_layer_count_roberta":false,
+    "freeze_layer_count_vit":false,
+    "head_hidden_scale":2,
+    "hidden_size":768,
+    "input_text_embed_size":768,
+    "link_tower_shared":false,
+    "link_tower_type":"add",
+    "log_dir":"log_dir",
+    "loss_names":{"contras": 0,
+            "irtr": 0,
+            "itm": 0,
+            "mlm": 0,
+            "mpp": 0,
+            "nlvr2": 0,
+            "snli": 0,
+            "vcr": 0,
+            "vcr_qar": 0,
+            "vqa": 1},
+    "max_text_len":50,
+    "mlp_ratio":4,
+    "model_type":"bridgetower",
+    "num_heads":12,
+    "num_layers":6,
+    "num_nodes":1,
+    "only_load_cross_modal_from_meter":false,
+    "patch_size":16,
+    "resolution_before":224,
+    "stop_gradient":false,
+    "size":288,
+    "task_head_layers":2,
+    "test_only":false,
+    "tokenizer":"roberta-base",
+    "unfreeze_RoBERTa_attention":false,
+    "unfreeze_RoBERTa_embeddings":false,
+    "unfreeze_RoBERTa_encoder":false,
+    "unfreeze_RoBERTa_layernorm":false,
+    "unfreeze_ViT_attention":false,
+    "unfreeze_ViT_layernorm":false,
+    "vit":"ViT-B/16",
+    "vit_layernorm_init_from_vit":false,
+    "vit_layernorm_shared":true,
+    "vit_remove_last":false,
+    "vocab_size":50265
 }