End of training

README.md

@@ -1,8 +1,7 @@
 ---
-base_model: microsoft/trocr-small-stage1
+base_model: microsoft/trocr-base-stage1
 tags:
-- trocr
-- image-to-text
+- generated_from_trainer
 model-index:
 - name: trocr-IAM
   results: []
@@ -13,10 +12,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # trocr-IAM
 
-This model is a fine-tuned version of [microsoft/trocr-small-stage1](https://huggingface.co/microsoft/trocr-small-stage1) on an unknown dataset.
+This model is a fine-tuned version of [microsoft/trocr-base-stage1](https://huggingface.co/microsoft/trocr-base-stage1) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 6.9227
-- Cer: 0.8191
+- Loss: 0.9888
+- Cer: 0.1284
 
 ## Model description
 
@@ -36,7 +35,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 16
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -47,7 +46,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Cer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 7.0738        | 0.79  | 100  | 6.9227          | 0.8191 |
+| 1.7207        | 0.2   | 50   | 1.5524          | 0.2112 |
+| 0.7594        | 0.4   | 100  | 0.9888          | 0.1284 |
 
 
 ### Framework versions
@@ -55,4 +55,4 @@ The following hyperparameters were used during training:
 - Transformers 4.35.2
 - Pytorch 2.1.0+cu121
 - Datasets 2.16.0
-- Tokenizers 0.15.0
+- Tokenizers 0.15.0

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "microsoft/trocr-small-stage1",
+  "_name_or_path": "microsoft/trocr-base-stage1",
   "architectures": [
     "VisionEncoderDecoderModel"
   ],
@@ -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,
@@ -43,10 +43,10 @@
       "LABEL_0": 0,
       "LABEL_1": 1
     },
-    "layernorm_embedding": true,
+    "layernorm_embedding": false,
     "length_penalty": 1.0,
     "max_length": 20,
-    "max_position_embeddings": 512,
+    "max_position_embeddings": 1024,
     "min_length": 0,
     "model_type": "trocr",
     "no_repeat_ngram_size": 0,
@@ -80,8 +80,8 @@
     "typical_p": 1.0,
     "use_bfloat16": false,
     "use_cache": false,
-    "use_learned_position_embeddings": true,
-    "vocab_size": 64044
+    "use_learned_position_embeddings": false,
+    "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": "4.35.2",
-  "vocab_size": 64044
+  "vocab_size": 50265
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4fa644f6b5ad795584c617bdbb348a8aaedcd16fd89465cf8dac887c1c471d1
-size 246430696
+oid sha256:4e7f883ff584596682f50f2bfe0d85276883b8173d4eb7567e628b1ff9e19e68
+size 1539518812

runs/Dec30_09-31-11_56e0cc75e629/events.out.tfevents.1703928680.56e0cc75e629.18680.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b38b5ea1f00feeb9b65638ab4eea843635aa4f739db801143adf767b8e589cb5
+size 17123

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8e7729de998ceff24acd971e78c5e1a7129bb20dd9312b1c5860bbc885139ee
+oid sha256:1bf3ee739dd2a91fa6b4784a4db8c78960318c061662a4ae67ebbdf66f768b5d
 size 4728