Add selected weights for D4P_KITTI (2 files)

selected/D4P_KITTI/config/01__coarse_config.json

@@ -0,0 +1,173 @@
+{
+  "dataset": "K360",
+  "base_path": "<REMOTE_PATH>",
+  "batch_size": 64,
+  "epochs": 20,
+  "learning_rate": 0.0005,
+  "lr_scheduler": "step",
+  "lr_step": 7,
+  "lr_gamma": 0.4,
+  "seed": 0,
+  "shuffle": false,
+  "num_workers": 0,
+  "folder_name": "kitti_bb0_coarse_seed0",
+  "checkpoint_root": "<REMOTE_PATH>",
+  "log_dir": null,
+  "resume": null,
+  "limit_batches": null,
+  "limit_val": null,
+  "device": "cuda",
+  "use_features": [
+    "class",
+    "color",
+    "position",
+    "num"
+  ],
+  "class_embed": false,
+  "color_embed": false,
+  "pointnet_path": "<REMOTE_PATH>",
+  "pointnet_layers": 3,
+  "pointnet_variation": 0,
+  "pointnet_numpoints": 256,
+  "pointnet_freeze": false,
+  "pointnet_features": 2,
+  "num_stats_path": "<REMOTE_PATH>",
+  "no_pc_augment": true,
+  "object_size": 28,
+  "pad_size": 16,
+  "num_mentioned": 6,
+  "coarse_embed_dim": 256,
+  "mfam_heads": 4,
+  "bilstm_stride": 2,
+  "bilstm_layers": 1,
+  "ste_word_layers": 1,
+  "ste_inter_layers": 1,
+  "ste_heads": 4,
+  "prior_layers": 2,
+  "prior_heads": 8,
+  "dropout": 0.1,
+  "fine_embed_dim": 128,
+  "fine_text_layers": 1,
+  "cra_heads": 4,
+  "cra_steps": 6,
+  "pmc_prob": 0.5,
+  "pmc_threshold": 0.4,
+  "offset_lambda": 5.0,
+  "warmup_epochs": 3,
+  "regressor_cell": "all",
+  "regressor_learn": "center",
+  "regressor_eval": "center",
+  "t5_path": "<REMOTE_PATH>",
+  "clip_path": null,
+  "ste_stage1_path": "<REMOTE_PATH>",
+  "clip_dim": 512,
+  "max_text_len": 32,
+  "hungging_model": null,
+  "fixed_embedding": false,
+  "ranking_loss": "contrastive",
+  "temperature": 0.1,
+  "top_k": [
+    1,
+    3,
+    5,
+    10,
+    25
+  ],
+  "select_k": 5,
+  "allow_test": false,
+  "implementation": "Des4Pos paper reimplementation (not authors' code)",
+  "known_classes": [
+    "box",
+    "bridge",
+    "building",
+    "fence",
+    "garage",
+    "guard rail",
+    "lamp",
+    "pad",
+    "parking",
+    "pole",
+    "road",
+    "sidewalk",
+    "smallpole",
+    "stop",
+    "terrain",
+    "traffic light",
+    "traffic sign",
+    "trash bin",
+    "tunnel",
+    "vegetation",
+    "vending machine",
+    "wall"
+  ],
+  "params_total": 16354430,
+  "params_trainable": 5315936,
+  "source_sha256": {
+    "models/mfam.py": "8d94ee028e060244b001a9f0846a6682ab6940e953495296d68661f08df54edd",
+    "models/bilstm.py": "e01cf99125706a8ad20d44bf6af8a7d20527aa558f0bf32de946b54636b241e9",
+    "models/ste.py": "404d4ac3ee83003aa46d1be8fe348da6bea6ea90f999482f09e70173c5a86ed1",
+    "models/cra.py": "34cad68bdc24daa04c599965918c10f19c9d1db47c64798c9e82763b57eb7bbf",
+    "models/coarse.py": "b599e9b8ea40768a0dbc6b91d66cce6272c7e6ffd2e9729469a335d08cc981a2",
+    "models/fine.py": "3c6e586156ff88929affd388054975bd719c37eb3b7754823f57cf4216b949fc",
+    "models/object_encoder.py": "0639381998bf538a3c7582f87c0b5de77d891d2174f6fc3bb6e67644d82815e2",
+    "models/pointcloud/pointnet2.py": "f9e9a35f9c92678464b909657790ab859083477bd69429c987dd954aabc3e9c7",
+    "models/mlp.py": "825cc99737a7757d3294f85cd4a35a40a3485ba2adfb9e895d706a90e49dfaa1",
+    "training/args.py": "70fd3dff84ed19a8fc753625e7417640881e5e54dbacdb0887470b941aa6d7cf",
+    "training/losses.py": "0d83b8e37ce9a3136c69cceaaa258402a7274d89a5dd75569f9ce091a3f39173",
+    "training/coarse.py": "80709fa07ca9c5beb67599fae10e17893301d36867be06cffff017d8a192973d",
+    "training/fine.py": "43a279146b5ab61b9c89845d7f961b1e4fd49c4e206a00f1d12dfbf705ec8420",
+    "training/ste_stage1.py": "a060508974075365f1dff05b2a16d503d0565b497a346e2fe35440c506c4ae52",
+    "evaluation/decode.py": "76f1230e125e483bffd9da8aa679e67bedc14dcb8c237c0f8433afad2ea295e1"
+  },
+  "t5_config": "{'vocab_size': 32128, 'd_model': 1024, 'd_kv': 64, 'd_ff': 4096, 'num_layers': 24, 'num_decoder_layers': 24, 'num_heads': 16, 'relative_attention_num_buckets': 32, 'relative_attention_max_distance': 128, 'dropout_rate': 0.1, 'classifier_dropout': 0.0, 'layer_norm_epsilon': 1e-06, 'initializer_factor': 1.0, 'feed_forward_proj': 'relu', 'use_cache': True, 'dense_act_fn': 'relu', 'is_gated_act': False, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': None, 'architectures': ['T5ForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 0, 'eos_token_id': 1, 'sep_token_id': None, 'decoder_start_token_id': 0, 'task_specific_params': {'summarization': {'early_stopping': True, 'length_penalty': 2.0, 'max_length': 200, 'min_length': 30, 'no_repeat_ngram_size': 3, 'num_beams': 4, 'prefix': 'summarize: '}, 'translation_en_to_de': {'early_stopping': True, 'max_length': 300, 'num_beams': 4, 'prefix': 'translate English to German: '}, 'translation_en_to_fr': {'early_stopping': True, 'max_length': 300, 'num_beams': 4, 'prefix': 'translate English to French: '}, 'translation_en_to_ro': {'early_stopping': True, 'max_length': 300, 'num_beams': 4, 'prefix': 'translate English to Romanian: '}}, 'problem_type': None, '_name_or_path': '<REMOTE_PATH>', '_attn_implementation_autoset': True, 'transformers_version': '4.46.3', 'model_type': 't5', 'n_positions': 512, 'output_past': True}",
+  "n_train_queries": 28807,
+  "n_val_queries": 3187,
+  "n_train_cells": 11259,
+  "n_val_cells": 1434,
+  "num_stats": {
+    "counts_sha256": null,
+    "domain": "kitti360pose",
+    "fitted_before_fixedpoints": true,
+    "max_count": null,
+    "mean": 1826.6844940968194,
+    "min_count": null,
+    "n_excluded": null,
+    "n_objects": null,
+    "n_seen": null,
+    "pinned_zero_variance": false,
+    "provenance": "official_constants",
+    "source": "nuozimiaowu/Des4Pos models/object_encoder.py (public KITTI360Pose recipe) hard-coded num_mean/num_std; registered unchanged for the KITTI reference run - not re-fitted on VAL/TEST",
+    "split": "train",
+    "std": 2516.8905096993817,
+    "std_ddof": 0,
+    "version": 1
+  },
+  "num_stats_fingerprint": "6a915ff83f4ea7d6defa59bcdaccc584020f9b69ef9f628481fb0366dbf3e7de",
+  "num_stats_source": "<REMOTE_PATH>",
+  "object_encoder_factory": {
+    "pointnet_head_classes": 22,
+    "pointnet_head_colors": 8,
+    "target_num_classes": 22,
+    "target_num_colors": 8,
+    "num_stats": {
+      "counts_sha256": null,
+      "domain": "kitti360pose",
+      "fitted_before_fixedpoints": true,
+      "max_count": null,
+      "mean": 1826.6844940968194,
+      "min_count": null,
+      "n_excluded": null,
+      "n_objects": null,
+      "n_seen": null,
+      "pinned_zero_variance": false,
+      "provenance": "official_constants",
+      "source": "nuozimiaowu/Des4Pos models/object_encoder.py (public KITTI360Pose recipe) hard-coded num_mean/num_std; registered unchanged for the KITTI reference run - not re-fitted on VAL/TEST",
+      "split": "train",
+      "std": 2516.8905096993817,
+      "std_ddof": 0,
+      "version": 1
+    },
+    "num_stats_fingerprint": "6a915ff83f4ea7d6defa59bcdaccc584020f9b69ef9f628481fb0366dbf3e7de",
+    "classifier_outputs_consumed": false
+  }
+}

selected/D4P_KITTI/scar_head/scar23_s0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d875d4e654bd929b15cbaaada9529254e9ac536780675887514f320c1562f40
+size 28188