Training in progress, epoch 0

config.json

@@ -22,7 +22,6 @@
     "init_std": 0.02,
     "is_decoder": true,
     "is_encoder_decoder": false,
-    "max_length": 768,
     "max_position_embeddings": 1536,
     "model_type": "mbart",
     "num_hidden_layers": 12,
@@ -46,8 +45,8 @@
     "hidden_dropout_prob": 0.0,
     "hidden_size": 1024,
     "image_size": [
-      1280,
-      960
+      2560,
+      1920
     ],
     "initializer_range": 0.02,
     "layer_norm_eps": 1e-05,
@@ -68,7 +67,12 @@
     "use_absolute_embeddings": false,
     "window_size": 10
   },
+  "image_size": [
+    1280,
+    960
+  ],
   "is_encoder_decoder": true,
+  "max_length": null,
   "model_type": "vision-encoder-decoder",
   "pad_token_id": 1,
   "tie_word_embeddings": false,

generation_config.json

@@ -1,6 +1,7 @@
 {
   "_from_model_config": true,
   "bos_token_id": 0,
+  "decoder_start_token_id": 57579,
   "eos_token_id": 2,
   "forced_eos_token_id": 2,
   "max_length": 768,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce7144ea32365e525cb5957fc711931968c3b48aa93be7521cb3bb96e90bbcc3
+oid sha256:0460e2eaca30dc1b69b31018a5996c88c4f4f5fb4b5124b51bb0703a368c0795
 size 809296024