Upload folder using huggingface_hub

.ipynb_checkpoints/model_index-checkpoint.json

@@ -0,0 +1,24 @@
+{
+  "_class_name": "CogVideoXImageToVideo4DPipeline",
+  "_diffusers_version": "0.33.0.dev0",
+  "scheduler": [
+    "diffusers",
+    "CogVideoXDDIMScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "T5EncoderModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "T5Tokenizer"
+  ],
+  "transformer": [
+    "diffusers",
+    "CogVideoXTransformer4DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLCogVideoX"
+  ]
+}

camera_embs/0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d29c274d32b0692f5bbac360d85141b110ffe2554f899a4691f09a4b72be45a
+size 8295422

camera_embs/1.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2672b6fae035043dda0a58dddc759f2c27fb7f45d45ecff0d712f99c515e85c
+size 8295422

camera_embs/10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69a5825f8543b873cdcab777810c1f5bd965afc89ea6d204edb8771422917aae
+size 8295491

camera_embs/11.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:482684c1201746cfa36fccdc44108441498f03652f31aa1221d99c3f3148ce8b
+size 8295491

camera_embs/12.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdf7e1158b5eeec7b0408adaa6ef252f99cca5b125ed859d0caec079372e3801
+size 8295491

camera_embs/13.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:115e92c8eac909fd6a435042857ecada2a753b5da14202e20a972ba699da4579
+size 8295491

camera_embs/14.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9cad6a3a2ee3083f18133a9abbadbec86e71c0def94e10533b2f17004dfcb26
+size 8295491

camera_embs/15.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaddf6a42d682242b33b43ea59adee5c9d5073cbc7734b78bef16122b5aaba60
+size 8295491

camera_embs/16.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb10275e0ae0be34abb48d945de700e6d74ece66389b2b2150ccc0a4755f8572
+size 8295491

camera_embs/17.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa253f6009c5ea6914c667d0535fd9ea0b58384fc35a046a0d04949466c23e6d
+size 8295491

camera_embs/18.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22922711f62d00119c946d0bd0431cdd5418e1e3c8120609594e3390359335bd
+size 8295491

camera_embs/19.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ecae280ffa16cd50e73fe21eec617b7435b89f3e854ab6371882b11d93ef51b
+size 8295491

camera_embs/2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1dbc1c8fad3c25f6681f7373ad88b91cb02502487da4b87bd894361e69e0efb
+size 8295422

camera_embs/20.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2de5fed811434faf3afeaee86e8f852d812be49e7048dd76995924517d284d36
+size 8295491

camera_embs/3.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74a3aadda727351fe36321a1267ab8733979613add618e257aef15b4d65ad49c
+size 8295422

camera_embs/4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a465788437b99b39c0b70f4b452af7fd55454520bd6251e9558a00af2612ad1
+size 8295422

camera_embs/5.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a23654d384f92dd38e4e8de8c273f905630909e3543409709e35bff234cf074
+size 8295422

camera_embs/6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f477d4185aa3466f950714c68f2b56862184dde1dc0c26dcd2197110fdfd59f9
+size 8295422

camera_embs/7.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b771768114de78b71d0f732ace435e36fa41a5e98cb0cec442d893d1cd51c3d6
+size 8295422

camera_embs/8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1aa1fe6126cbbe0f4f8357ef0d8b264bbe2484c5141ce080cc1faaf0035e1010
+size 8295422

camera_embs/9.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dfaca72ef5b5ecb804b5f3f0892df29b224c1791c29320974d17f2b3d48438d
+size 8295422

text_encoder/.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,31 @@
+{
+  "architectures": [
+    "T5EncoderModel"
+  ],
+  "classifier_dropout": 0.0,
+  "d_ff": 10240,
+  "d_kv": 64,
+  "d_model": 4096,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "gated-gelu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "num_decoder_layers": 24,
+  "num_heads": 64,
+  "num_layers": 24,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.2",
+  "use_cache": true,
+  "vocab_size": 32128
+}

transformer/.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,32 @@
+{
+  "_class_name": "CogVideoXTransformer4DModel",
+  "_diffusers_version": "0.33.0.dev0",
+  "activation_fn": "gelu-approximate",
+  "attention_bias": true,
+  "attention_head_dim": 64,
+  "dropout": 0.0,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 32,
+  "max_text_seq_length": 226,
+  "norm_elementwise_affine": true,
+  "norm_eps": 1e-05,
+  "num_attention_heads": 48,
+  "num_layers": 42,
+  "ofs_embed_dim": null,
+  "out_channels": 16,
+  "patch_bias": true,
+  "patch_size": 2,
+  "patch_size_t": null,
+  "sample_frames": 49,
+  "sample_height": 60,
+  "sample_width": 90,
+  "spatial_interpolation_scale": 1.875,
+  "temporal_compression_ratio": 4,
+  "temporal_interpolation_scale": 1.0,
+  "text_embed_dim": 4096,
+  "time_embed_dim": 512,
+  "timestep_activation_fn": "silu",
+  "use_learned_positional_embeddings": true,
+  "use_rotary_positional_embeddings": true
+}