Training in progress, step 500

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "microsoft/trocr-small-printed",
+  "_name_or_path": "microsoft/trocr-base-printed",
   "architectures": [
     "VisionEncoderDecoderModel"
   ],
   "decoder": {
     "_name_or_path": "",
     "activation_dropout": 0.0,
-    "activation_function": "relu",
+    "activation_function": "gelu",
     "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": 384,
-    "d_model": 256,
-    "decoder_attention_heads": 8,
-    "decoder_ffn_dim": 1024,
+    "cross_attention_hidden_size": 768,
+    "d_model": 1024,
+    "decoder_attention_heads": 16,
+    "decoder_ffn_dim": 4096,
     "decoder_layerdrop": 0.0,
-    "decoder_layers": 6,
+    "decoder_layers": 12,
     "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": true,
+    "scale_embedding": false,
     "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": false,
+    "tie_word_embeddings": true,
     "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": 64044
+    "vocab_size": 50265
   },
   "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": 384,
+    "hidden_size": 768,
     "id2label": {
       "0": "LABEL_0",
       "1": "LABEL_1"
     },
     "image_size": 384,
     "initializer_range": 0.02,
-    "intermediate_size": 1536,
+    "intermediate_size": 3072,
     "is_decoder": false,
     "is_encoder_decoder": false,
     "label2id": {
@@ -126,9 +126,9 @@
     "length_penalty": 1.0,
     "max_length": 20,
     "min_length": 0,
-    "model_type": "deit",
+    "model_type": "vit",
     "no_repeat_ngram_size": 0,
-    "num_attention_heads": 6,
+    "num_attention_heads": 12,
     "num_beam_groups": 1,
     "num_beams": 1,
     "num_channels": 3,
@@ -142,7 +142,7 @@
     "prefix": null,
     "problem_type": null,
     "pruned_heads": {},
-    "qkv_bias": true,
+    "qkv_bias": false,
     "remove_invalid_values": false,
     "repetition_penalty": 1.0,
     "return_dict": true,
@@ -173,5 +173,5 @@
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.37.2",
-  "vocab_size": 64044
+  "vocab_size": 50265
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab21396825da2c05749b63138470d584cf1894d2665aa6b4f03f11600181b618
-size 246430696
+oid sha256:2176485c60464ed43e6ea87b2e20820e09416310cd571900bb589d6f3937b986
+size 1335747032

preprocessor_config.json

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

runs/Apr23_13-15-41_devai/events.out.tfevents.1713878159.devai.1671410.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72db56a933486d3bfff8ddc1a20e65ddc44df3a9fd7a000f3435c0ddec4ff52d
-size 39543
+oid sha256:c5cdb07707403489099a4810e401f0825c9c2636e14e8ff8bba26480fab3f645
+size 41805

runs/Apr24_03-06-19_devai/events.out.tfevents.1713927982.devai.1955363.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b86a60f8b2e13e4f3b61f6910a2177584eaa25d29568353e1f04be01757212f7
+size 8280

runs/Apr24_03-08-45_devai/events.out.tfevents.1713928128.devai.1955753.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:988dde1dcf181fd41a349a5b70f3b23a8fdc9b3c3ea6178d08e84bde839ea9d9
+size 8280

runs/Apr24_03-09-43_devai/events.out.tfevents.1713928186.devai.1956042.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2db3aee3021e02146133b642849a1c63179d71da9430b603f41cf91d46a2b221
+size 8280

runs/Apr24_03-10-37_devai/events.out.tfevents.1713928240.devai.1956278.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5494f78c2644f382361a82460fb63c9b003fac718332557bdc7078216eaa2230
+size 9682

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88bc0f5e73c45520b0025aa402e4a09038a5f619366f21dbb1b7509f3faac844
+oid sha256:6c8841bf546d6ddbda5c5ba1380cbd0a9b788a099db5487d6e92d7489a1f0fc3
 size 4856