update model card README.md

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

README.md

@@ -0,0 +1,68 @@
+---
+license: apache-2.0
+tags:
+- vision
+- depth-estimation
+- generated_from_trainer
+model-index:
+- name: glpn-kitti-finetuned-diode
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# glpn-kitti-finetuned-diode
+
+This model is a fine-tuned version of [vinvino02/glpn-kitti](https://huggingface.co/vinvino02/glpn-kitti) on the diode-subset dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5845
+- Rmse: 0.6175
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Rmse   |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| No log        | 1.0   | 10   | 0.8001          | 0.8455 |
+| 0.8187        | 2.0   | 20   | 0.7558          | 0.7907 |
+| 0.8187        | 3.0   | 30   | 0.7391          | 0.7379 |
+| 0.7618        | 4.0   | 40   | 0.6937          | 0.6895 |
+| 0.7618        | 5.0   | 50   | 0.6954          | nan    |
+| 0.6917        | 6.0   | 60   | 0.6834          | nan    |
+| 0.6917        | 7.0   | 70   | 0.6719          | nan    |
+| 0.6625        | 8.0   | 80   | 0.6634          | nan    |
+| 0.6625        | 9.0   | 90   | 0.6592          | nan    |
+| 0.6553        | 10.0  | 100  | 0.6579          | nan    |
+
+
+### Framework versions
+
+- Transformers 4.24.0
+- Pytorch 1.12.1+cu113
+- Tokenizers 0.13.2

config.json

@@ -0,0 +1,70 @@
+{
+  "_name_or_path": "vinvino02/glpn-kitti",
+  "architectures": [
+    "GLPNForDepthEstimation"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "classifier_dropout_prob": 0.1,
+  "decoder_hidden_size": 64,
+  "depths": [
+    3,
+    8,
+    27,
+    3
+  ],
+  "downsampling_rates": [
+    1,
+    4,
+    8,
+    16
+  ],
+  "drop_path_rate": 0.1,
+  "head_in_index": -1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_sizes": [
+    64,
+    128,
+    320,
+    512
+  ],
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "layer_norm_eps": 1e-06,
+  "max_depth": 10,
+  "mlp_ratios": [
+    4,
+    4,
+    4,
+    4
+  ],
+  "model_type": "glpn",
+  "num_attention_heads": [
+    1,
+    2,
+    5,
+    8
+  ],
+  "num_channels": 3,
+  "num_encoder_blocks": 4,
+  "patch_sizes": [
+    7,
+    3,
+    3,
+    3
+  ],
+  "sr_ratios": [
+    8,
+    4,
+    2,
+    1
+  ],
+  "strides": [
+    4,
+    2,
+    2,
+    2
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.24.0"
+}

preprocessor_config.json

@@ -0,0 +1,7 @@
+{
+  "do_rescale": true,
+  "do_resize": true,
+  "feature_extractor_type": "GLPNImageProcessor",
+  "resample": 2,
+  "size_divisor": 32
+}