Training in progress, step 1000

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "microsoft/trocr-base-handwritten",
+  "_name_or_path": "microsoft/trocr-small-stage1",
   "architectures": [
     "VisionEncoderDecoderModel"
   ],
   "decoder": {
     "_name_or_path": "",
     "activation_dropout": 0.0,
-    "activation_function": "gelu",
+    "activation_function": "relu",
     "add_cross_attention": true,
     "architectures": null,
     "attention_dropout": 0.0,
@@ -15,12 +15,12 @@
     "bos_token_id": 0,
     "chunk_size_feed_forward": 0,
     "classifier_dropout": 0.0,
-    "cross_attention_hidden_size": 768,
-    "d_model": 1024,
-    "decoder_attention_heads": 16,
-    "decoder_ffn_dim": 4096,
+    "cross_attention_hidden_size": 384,
+    "d_model": 256,
+    "decoder_attention_heads": 8,
+    "decoder_ffn_dim": 1024,
     "decoder_layerdrop": 0.0,
-    "decoder_layers": 12,
+    "decoder_layers": 6,
     "decoder_start_token_id": 2,
     "diversity_penalty": 0.0,
     "do_sample": false,
@@ -64,14 +64,14 @@
     "repetition_penalty": 1.0,
     "return_dict": true,
     "return_dict_in_generate": false,
-    "scale_embedding": false,
+    "scale_embedding": true,
     "sep_token_id": null,
     "suppress_tokens": null,
     "task_specific_params": null,
     "temperature": 1.0,
     "tf_legacy_loss": false,
     "tie_encoder_decoder": false,
-    "tie_word_embeddings": true,
+    "tie_word_embeddings": false,
     "tokenizer_class": null,
     "top_k": 50,
     "top_p": 1.0,
@@ -81,7 +81,7 @@
     "use_bfloat16": false,
     "use_cache": false,
     "use_learned_position_embeddings": true,
-    "vocab_size": 50265
+    "vocab_size": 64044
   },
   "decoder_start_token_id": 0,
   "early_stopping": true,
@@ -108,14 +108,14 @@
     "forced_eos_token_id": null,
     "hidden_act": "gelu",
     "hidden_dropout_prob": 0.0,
-    "hidden_size": 768,
+    "hidden_size": 384,
     "id2label": {
       "0": "LABEL_0",
       "1": "LABEL_1"
     },
     "image_size": 384,
     "initializer_range": 0.02,
-    "intermediate_size": 3072,
+    "intermediate_size": 1536,
     "is_decoder": false,
     "is_encoder_decoder": false,
     "label2id": {
@@ -126,9 +126,9 @@
     "length_penalty": 1.0,
     "max_length": 20,
     "min_length": 0,
-    "model_type": "vit",
+    "model_type": "deit",
     "no_repeat_ngram_size": 0,
-    "num_attention_heads": 12,
+    "num_attention_heads": 6,
     "num_beam_groups": 1,
     "num_beams": 1,
     "num_channels": 3,
@@ -142,7 +142,7 @@
     "prefix": null,
     "problem_type": null,
     "pruned_heads": {},
-    "qkv_bias": false,
+    "qkv_bias": true,
     "remove_invalid_values": false,
     "repetition_penalty": 1.0,
     "return_dict": true,
@@ -170,9 +170,8 @@
   "no_repeat_ngram_size": 3,
   "num_beams": 4,
   "pad_token_id": 1,
-  "processor_class": "TrOCRProcessor",
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.44.2",
-  "vocab_size": 50265
+  "vocab_size": 64044
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0738ecfb4d730a4a7c78d45065de43c2bda86533a37cb76f41a51f426cb18823
-size 1335747032
+oid sha256:9463f99a60406b70fd45948d2c81b5a8814e65a7c2a706578327a2aa4805eb8c
+size 246430696

preprocessor_config.json

@@ -1,4 +1,9 @@
 {
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": false,
   "do_normalize": true,
   "do_rescale": true,
   "do_resize": true,
@@ -7,13 +12,13 @@
     0.5,
     0.5
   ],
-  "image_processor_type": "ViTImageProcessor",
+  "image_processor_type": "DeiTImageProcessor",
   "image_std": [
     0.5,
     0.5,
     0.5
   ],
-  "resample": 2,
+  "resample": 3,
   "rescale_factor": 0.00392156862745098,
   "size": {
     "height": 384,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b87537a4c40bb212ce66b4f591484fbb6a149bc9b1b59349066cc5b0a6ecd600
+oid sha256:6dd15a9a7ba5a4eebc5265a10d9ea1e6271df6fe100f065d0639401e31bac86a
 size 5304