keras-hub-nightly 0.22.0.dev202505290412__py3-none-any.whl → 0.22.0.dev202505310408__py3-none-any.whl

keras_hub/src/utils/transformers/convert_qwen3.py

@@ -0,0 +1,145 @@
+import numpy as np
+
+from keras_hub.src.models.qwen3.qwen3_backbone import Qwen3Backbone
+from keras_hub.src.utils.preset_utils import load_json
+
+backbone_cls = Qwen3Backbone
+
+
+def convert_backbone_config(transformers_config):
+    return {
+        "vocabulary_size": transformers_config["vocab_size"],
+        "head_dim": transformers_config["head_dim"],
+        "hidden_dim": transformers_config["hidden_size"],
+        "num_layers": transformers_config["num_hidden_layers"],
+        "num_query_heads": transformers_config["num_attention_heads"],
+        "num_key_value_heads": transformers_config["num_key_value_heads"],
+        "intermediate_dim": transformers_config["intermediate_size"],
+        "layer_norm_epsilon": transformers_config["rms_norm_eps"],
+        "rope_max_wavelength": transformers_config["rope_theta"],
+        "sliding_window_size": transformers_config["sliding_window"]
+        if transformers_config["use_sliding_window"]
+        else None,
+        "tie_word_embeddings": transformers_config["tie_word_embeddings"],
+    }
+
+
+def convert_weights(backbone, loader, transformers_config):
+    loader.port_weight(
+        keras_variable=backbone.get_layer("token_embedding").embeddings,
+        hf_weight_key="model.embed_tokens.weight",
+    )
+    if not backbone.tie_word_embeddings:
+        loader.port_weight(
+            keras_variable=backbone.get_layer(
+                "token_embedding"
+            ).reverse_embeddings,
+            hf_weight_key="lm_head.weight",
+            # rearrange_pattern="b a -> a b",
+            hook_fn=lambda hf_tensor, _: np.transpose(hf_tensor, axes=(1, 0)),
+        )
+
+    def transpose_and_reshape(x, shape):
+        return np.reshape(np.transpose(x), shape)
+
+    for i in range(backbone.num_layers):
+        decoder_layer = backbone.get_layer(f"transformer_layer_{i}")
+
+        # Input layernorm
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layernorm.scale,
+            hf_weight_key=f"model.layers.{i}.input_layernorm.weight",
+        )
+
+        # Attention layers
+
+        ## Query
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layer._query_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.self_attn.q_proj.weight",
+            hook_fn=transpose_and_reshape,
+        )
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layer._query_dense_layer_norm.scale,
+            hf_weight_key=f"model.layers.{i}.self_attn.q_norm.weight",
+        )
+        ## Key
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layer._key_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.self_attn.k_proj.weight",
+            hook_fn=transpose_and_reshape,
+        )
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layer._key_dense_layer_norm.scale,
+            hf_weight_key=f"model.layers.{i}.self_attn.k_norm.weight",
+        )
+        ## Value
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layer._value_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.self_attn.v_proj.weight",
+            hook_fn=transpose_and_reshape,
+        )
+        ## Output
+        loader.port_weight(
+            keras_variable=decoder_layer._self_attention_layer._output_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.self_attn.o_proj.weight",
+            # rearrange_patterns="c (a b) -> a b c",
+            # rearrange_dims={"a": backbone.num_query_heads},
+            hook_fn=transpose_and_reshape,
+        )
+
+        # MLP layers
+        loader.port_weight(
+            keras_variable=decoder_layer._feedforward_intermediate_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.mlp.up_proj.weight",
+            # rearrange_patterns="b a -> a b",
+            hook_fn=lambda hf_tensor, _: np.transpose(hf_tensor, axes=(1, 0)),
+        )
+        loader.port_weight(
+            keras_variable=decoder_layer._feedforward_output_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.mlp.down_proj.weight",
+            # rearrange_patterns="b a -> a b",
+            hook_fn=lambda hf_tensor, _: np.transpose(hf_tensor, axes=(1, 0)),
+        )
+        loader.port_weight(
+            keras_variable=decoder_layer._feedforward_gate_dense.kernel,
+            hf_weight_key=f"model.layers.{i}.mlp.gate_proj.weight",
+            # rearrange_patterns="b a -> a b",
+            hook_fn=lambda hf_tensor, _: np.transpose(hf_tensor, axes=(1, 0)),
+        )
+
+        # Feedforward layernorm
+        loader.port_weight(
+            keras_variable=decoder_layer._feedforward_layernorm.scale,
+            hf_weight_key=f"model.layers.{i}.post_attention_layernorm.weight",
+        )
+
+    # Final normalization layer
+    loader.port_weight(
+        keras_variable=backbone.get_layer("sequence_output_layernorm").scale,
+        hf_weight_key="model.norm.weight",
+    )
+
+    return backbone
+
+
+def convert_tokenizer(cls, preset, **kwargs):
+    tokenizer_config = load_json(preset, "tokenizer.json")
+    vocab = tokenizer_config["model"]["vocab"]
+    merges = tokenizer_config["model"]["merges"]
+    merges = [" ".join(item) for item in merges]
+
+    # Load all special tokens with the exception of "reserved" ones.
+    special_tokens = set()
+    for token in tokenizer_config["added_tokens"]:
+        if not token["content"].startswith("<|reserved_special_token_"):
+            vocab[token["content"]] = token["id"]
+            special_tokens.add(token["content"])
+
+    kwargs.update(
+        {
+            "unsplittable_tokens": list(special_tokens),
+        }
+    )
+
+    return cls(vocabulary=vocab, merges=merges, **kwargs)

keras_hub/src/utils/transformers/preset_loader.py

@@ -6,6 +6,7 @@ from keras_hub.src.utils.preset_utils import jax_memory_cleanup
 from keras_hub.src.utils.transformers import convert_albert
 from keras_hub.src.utils.transformers import convert_bart
 from keras_hub.src.utils.transformers import convert_bert
+from keras_hub.src.utils.transformers import convert_deit
 from keras_hub.src.utils.transformers import convert_distilbert
 from keras_hub.src.utils.transformers import convert_gemma
 from keras_hub.src.utils.transformers import convert_gpt2
@@ -14,6 +15,7 @@ from keras_hub.src.utils.transformers import convert_mistral
 from keras_hub.src.utils.transformers import convert_mixtral
 from keras_hub.src.utils.transformers import convert_pali_gemma
 from keras_hub.src.utils.transformers import convert_qwen
+from keras_hub.src.utils.transformers import convert_qwen3
 from keras_hub.src.utils.transformers import convert_qwen_moe
 from keras_hub.src.utils.transformers import convert_vit
 from keras_hub.src.utils.transformers.safetensor_utils import SafetensorLoader
@@ -29,6 +31,8 @@ class TransformersPresetLoader(PresetLoader):
             self.converter = convert_bart
         elif model_type == "bert":
             self.converter = convert_bert
+        elif model_type == "deit":
+            self.converter = convert_deit
         elif model_type == "distilbert":
             self.converter = convert_distilbert
         elif model_type == "gemma" or model_type == "gemma2":
@@ -50,6 +54,8 @@ class TransformersPresetLoader(PresetLoader):
             self.converter = convert_mixtral
         elif model_type == "qwen2_moe":
             self.converter = convert_qwen_moe
+        elif model_type == "qwen3":
+            self.converter = convert_qwen3
         else:
             raise ValueError(
                 "KerasHub has no converter for huggingface/transformers models "
@@ -79,7 +85,7 @@ class TransformersPresetLoader(PresetLoader):
                 cls, load_weights, load_task_weights, **kwargs
             )
         # Support loading the classification head for classifier models.
-        if architecture == "ViTForImageClassification":
+        if "ForImageClassification" in architecture:
             kwargs["num_classes"] = len(self.config["id2label"])
         task = super().load_task(cls, load_weights, load_task_weights, **kwargs)
         if load_task_weights:

keras_hub/src/version.py

@@ -1,7 +1,7 @@
 from keras_hub.src.api_export import keras_hub_export
 
 # Unique source of truth for the version number.
-__version__ = "0.22.0.dev202505290412"
+__version__ = "0.22.0.dev202505310408"
 
 
 @keras_hub_export("keras_hub.version")

{keras_hub_nightly-0.22.0.dev202505290412.dist-info → keras_hub_nightly-0.22.0.dev202505310408.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: keras-hub-nightly
-Version: 0.22.0.dev202505290412
+Version: 0.22.0.dev202505310408
 Summary: Pretrained models for Keras.
 Author-email: Keras team <keras-users@googlegroups.com>
 License-Expression: Apache-2.0

{keras_hub_nightly-0.22.0.dev202505290412.dist-info → keras_hub_nightly-0.22.0.dev202505310408.dist-info}/RECORD

@@ -1,11 +1,11 @@
 keras_hub/__init__.py,sha256=bJbUZkqwhZvTb1Tqx1fbkq6mzBYiEyq-Hin3oQIkhdE,558
-keras_hub/layers/__init__.py,sha256=gnvT-GuASB1hZwY4zrRkLs5yohSQu9Pp1SHDxsWPLY8,5081
+keras_hub/layers/__init__.py,sha256=YQ4bW0_mI39Jqj2yoc8xcnynqoaXV2FBjHJviA9Ffas,5190
 keras_hub/metrics/__init__.py,sha256=KYalsMPBnfwim9BdGHFfJ5WxUKFXOQ1QoKIMT_0lwlM,439
-keras_hub/models/__init__.py,sha256=itSzodVUeuX6HQnmsSXY0Wv-5Htbu397410R-SFW_4I,26411
+keras_hub/models/__init__.py,sha256=7MhCw7S-uIPcko-R6g5a-Jy1idKe7BwlI836PfekhHc,27076
 keras_hub/samplers/__init__.py,sha256=aFQIkiqbZpi8vjrPp2MVII4QUfE-eQjra5fMeHsoy7k,886
 keras_hub/src/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 keras_hub/src/api_export.py,sha256=9pQZK27JObxWZ96QPLBp1OBsjWigh1iuV6RglPGMRk0,1499
-keras_hub/src/version.py,sha256=DDvaRSyKJcjRMYdIJIroiLLIbnEZPXF5mlsR_VQNowQ,222
+keras_hub/src/version.py,sha256=A_oYO8DhCB-uOrecxZt2B7NMyEpt94fhLGZT7-dbdBg,222
 keras_hub/src/layers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 keras_hub/src/layers/modeling/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 keras_hub/src/layers/modeling/alibi_bias.py,sha256=1XBTHI52L_iJDhN_w5ydu_iMhCuTgQAxEPwcLA6BPuk,4411
@@ -135,6 +135,13 @@ keras_hub/src/models/deeplab_v3/deeplab_v3_image_segmeter_preprocessor.py,sha256
 keras_hub/src/models/deeplab_v3/deeplab_v3_layers.py,sha256=mz9nG55gdXSTDE96AXgeTCwUFB95DIpTuqrvWIt5Lco,7840
 keras_hub/src/models/deeplab_v3/deeplab_v3_presets.py,sha256=ZKYY8A7mV2QvwXwjDUd9xAbVHo58-Hgj_IqNUbuyCIU,625
 keras_hub/src/models/deeplab_v3/deeplab_v3_segmenter.py,sha256=pubi30sPJKLOpz9fRQff2FZt_53KBvwf2uyaJ5YL7J8,3726
+keras_hub/src/models/deit/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+keras_hub/src/models/deit/deit_backbone.py,sha256=R5pBOqe8vcvD8VaRnsy_zIRIz6BLnUbkTeKUOoGNHPA,5942
+keras_hub/src/models/deit/deit_image_classifier.py,sha256=pUS2638yBAxEBxcJoHyLABsgjCWv_Y0Mj_8u0YgDPdI,5758
+keras_hub/src/models/deit/deit_image_classifier_preprocessor.py,sha256=s5pTcsUjlt1oIXFWIu-9gf2-sBesAyrjJIYmFOB96Xs,514
+keras_hub/src/models/deit/deit_image_converter.py,sha256=wEGCLHS_i4wF9WA4m7uUXcHNbwf6TYgvPoM6C_t0rpM,330
+keras_hub/src/models/deit/deit_layers.py,sha256=A80-UTHEUV8g5rEG-fr8OQpGe3HeoYlYwpoDCtq71ZU,17278
+keras_hub/src/models/deit/deit_presets.py,sha256=0c2jm2DIznOr6ciQoLM6QYopQTLiMx4jONGLaXvtt6g,1778
 keras_hub/src/models/densenet/__init__.py,sha256=r7StyamnWeeZxOk9r4ZYNbS_YVhu9YGPyXhNxljvdPg,269
 keras_hub/src/models/densenet/densenet_backbone.py,sha256=f2nfsXyXQert2aYHq-L-JZtp8inq1fs1K47rzZQ9nTI,6744
 keras_hub/src/models/densenet/densenet_image_classifier.py,sha256=ye-Ix3oU42pfsDoh-h1PG4di1kzldO0ZO7Nj304p_X4,544
@@ -259,7 +266,7 @@ keras_hub/src/models/mixtral/mixtral_causal_lm.py,sha256=JA1t6xTeaYX_fNo9ftRyvzd
 keras_hub/src/models/mixtral/mixtral_causal_lm_preprocessor.py,sha256=q2qXa9QAUWBvOWv9DeNvwsBNXSORJAbQFoQsWQ7e8V8,3079
 keras_hub/src/models/mixtral/mixtral_decoder.py,sha256=CvOjhTxPnGQ_HNknZXRI6Cx1kpuHG99_TiOh-mNcsDw,18190
 keras_hub/src/models/mixtral/mixtral_layer_norm.py,sha256=zfbDKZEb45FTwP0zQd7WPPp8tuiGoSNfS-DRYWkZyWw,1031
-keras_hub/src/models/mixtral/mixtral_presets.py,sha256=AteLrYXyVjooz_DHLnBA1OMlZS6LMu7Y7gGUWddn6go,856
+keras_hub/src/models/mixtral/mixtral_presets.py,sha256=pi5hHcwVSqr7ytf4dSnU_ew_t7NYw7EsZrmklQDqDVo,852
 keras_hub/src/models/mixtral/mixtral_tokenizer.py,sha256=Kc233k879QMyX164X_CzWbqpnqEkKWNqa648guTGkBk,661
 keras_hub/src/models/mobilenet/__init__.py,sha256=hxkNGGj_iAMu62iooUDEPA818sNOIgjG7pXMLEMOsAE,275
 keras_hub/src/models/mobilenet/mobilenet_backbone.py,sha256=aZBSFeLUObYYoi3od9DI1KfgPCqh5GHTcAI8Y2ZHShA,29536
@@ -311,8 +318,14 @@ keras_hub/src/models/qwen/qwen_causal_lm.py,sha256=_f-UHaKHp0ncxknpkpEJiW3jlng3E
 keras_hub/src/models/qwen/qwen_causal_lm_preprocessor.py,sha256=Va-4TLJD3ycEnkS41rF3dVj4_6K0j-gxLTrREFRcyr0,609
 keras_hub/src/models/qwen/qwen_decoder.py,sha256=utmAvZlU7_nP-6pjGPDinK4JaMzsQSwOARG0ote-jAg,11771
 keras_hub/src/models/qwen/qwen_layernorm.py,sha256=DS35r3qd6g5ocL7Nhf_vNzLLMo1aI9VCSmL64dgNOYI,924
-keras_hub/src/models/qwen/qwen_presets.py,sha256=DpRplWNwktM4KDgIP495PTUBJxQE_mS6KQSK5LGWOyc,1998
+keras_hub/src/models/qwen/qwen_presets.py,sha256=1FkKV6M3yqJz4EP1xa7bEvfIQ721xXT-_ikjWX0xvww,1992
 keras_hub/src/models/qwen/qwen_tokenizer.py,sha256=LCv3IyiDDHqVnM9N3lf5-BE3iwicIh0nKS1hjoPw9lE,1532
+keras_hub/src/models/qwen3/qwen3_attention.py,sha256=sewLjli290XvJ1efGZJEAYqUZfRll7cmhu0258s4C48,13042
+keras_hub/src/models/qwen3/qwen3_backbone.py,sha256=Ylpk_rRWWRxy8irlAPjJU-YrxYGpo8c9lSEO1zZl4gU,7456
+keras_hub/src/models/qwen3/qwen3_causal_lm_preprocessor.py,sha256=H4g-bgvuhAUnDwjJovydK16Kes38ZFZWPvflrgHqZis,458
+keras_hub/src/models/qwen3/qwen3_decoder.py,sha256=68s9jQj53zFmXE4-SGXKYHu546fXOyi9LUbnKk-HGYY,11595
+keras_hub/src/models/qwen3/qwen3_layernorm.py,sha256=EJxjf7Pr6ufPQnNeuYQxkExzPjPk4PQxqMsoBeSEkDo,1073
+keras_hub/src/models/qwen3/qwen3_tokenizer.py,sha256=LmPtg0vprMchDvYfTj8m5PraXI2QS3-YgdIIpIm5iAs,1448
 keras_hub/src/models/qwen_moe/__init__.py,sha256=5D8GUmVDsJs0J4sVZHcXOLkZf12U96l-WtwyVee4lu8,267
 keras_hub/src/models/qwen_moe/qwen_moe_attention.py,sha256=pE79_iHUm2LGkoWL6zMJw_pNfzIvmyq3yJaiq47W2TY,13242
 keras_hub/src/models/qwen_moe/qwen_moe_backbone.py,sha256=nrfELvIvRLmrgKrUNXci2CrecmeI6bWzJj7HH-RcWJA,15341
@@ -320,7 +333,7 @@ keras_hub/src/models/qwen_moe/qwen_moe_causal_lm.py,sha256=MeP60v7GcN_SmH5_ULRpq
 keras_hub/src/models/qwen_moe/qwen_moe_causal_lm_preprocessor.py,sha256=uKaXRrJs02vkVudjdehzJPp0B84tPMkxNHlp166kceE,589
 keras_hub/src/models/qwen_moe/qwen_moe_decoder.py,sha256=kmUjLpYTbJQ3J_31qWhLOd0Dg2_9cl_JX_zM8ZMH1Qo,23130
 keras_hub/src/models/qwen_moe/qwen_moe_layernorm.py,sha256=DbkWJo7U0-cwdZwHPeAnFznYwtao6o0fjpoDJ9UWnpc,927
-keras_hub/src/models/qwen_moe/qwen_moe_presets.py,sha256=uKrA9xLV3P3jtYUUsqdhKq_HPkB4lXmOYseB1wXTZnI,457
+keras_hub/src/models/qwen_moe/qwen_moe_presets.py,sha256=LhOA3Ow-z3cNTan4AOrtyCXS58EgfvO_gtqiZt5cUQc,455
 keras_hub/src/models/qwen_moe/qwen_moe_tokenizer.py,sha256=2c3X8jNGO0q0UL5NtUqSgHWLqhyJGi2ohNcTeOGhd84,1407
 keras_hub/src/models/resnet/__init__.py,sha256=C5UqlQ6apm8WSp1bnrxB6Bi3BGaknxRQs-r3b2wpaGA,257
 keras_hub/src/models/resnet/resnet_backbone.py,sha256=Q7nlqcTXZzjqd0e-DsjHC4ok58yOX7qxseotym3uZpM,31276
@@ -488,6 +501,7 @@ keras_hub/src/utils/transformers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRk
 keras_hub/src/utils/transformers/convert_albert.py,sha256=VdKclZpCxtDWq3UbUUQZf4fR9DJK_JYZ73B4O_G9skg,7695
 keras_hub/src/utils/transformers/convert_bart.py,sha256=Tk4h9Md9rwN5wjQbGIVrC7qzDpF8kI8qm-FKL8HlUok,14411
 keras_hub/src/utils/transformers/convert_bert.py,sha256=4gQqXCJzC9QWdLPDUAq741K8t_kjPIET050YjUnLeDA,5977
+keras_hub/src/utils/transformers/convert_deit.py,sha256=ubcqYzMlhWTCE2S_TsXICCMmqjN9RsQPaw_70vArnjo,5306
 keras_hub/src/utils/transformers/convert_distilbert.py,sha256=SlfIRhSRk5c1ir2HGiDPiXa5XdOId_DbcnZO9lbwyZ8,6498
 keras_hub/src/utils/transformers/convert_gemma.py,sha256=ElCgwBpSN5Q7rV5PJawTsoytPzs5ZjuwoY60YAe8y_A,6533
 keras_hub/src/utils/transformers/convert_gpt2.py,sha256=HCeHN_-GiQJRxLCM9OCJJ1watPVpIBF8ujS8pGbBOWc,5703
@@ -496,13 +510,14 @@ keras_hub/src/utils/transformers/convert_mistral.py,sha256=kVhN9h1ZFVhwkNW8p3wnS
 keras_hub/src/utils/transformers/convert_mixtral.py,sha256=PxeCY8Xe7U_caICugwOCEjuSZ51ZUtmef6rUxh-Wt54,5508
 keras_hub/src/utils/transformers/convert_pali_gemma.py,sha256=B1leeDw96Yvu81hYumf66hIid07k5NLqoeWAJgPnaLs,10649
 keras_hub/src/utils/transformers/convert_qwen.py,sha256=WUxMAEFVqRs7TRw7QU5TH3_ev4yf02R1xFVliMvTQqg,5886
+keras_hub/src/utils/transformers/convert_qwen3.py,sha256=LIormvCMWPq6X9Wo2eNbADjtFZ0nI7tFGZFBxmo4GKw,5700
 keras_hub/src/utils/transformers/convert_qwen_moe.py,sha256=a7R28aln-PdAcNuKAXdrtzvslho2Co6GypChxLMKPpc,10618
 keras_hub/src/utils/transformers/convert_vit.py,sha256=9SUZ9utNJhW_5cj3acMn9cRy47u2eIcDsrhmzj77o9k,5187
-keras_hub/src/utils/transformers/preset_loader.py,sha256=1nfS5xVsl-JROGXJXltTqV1fQdcUlZbGGcbf-n79pXM,4225
+keras_hub/src/utils/transformers/preset_loader.py,sha256=K5FzDAtCuXS9rmZc0Zj7UCwbz5J9_pf7ozWov1qRAfg,4495
 keras_hub/src/utils/transformers/safetensor_utils.py,sha256=CYUHyA4y-B61r7NDnCsFb4t_UmSwZ1k9L-8gzEd6KRg,3339
 keras_hub/tokenizers/__init__.py,sha256=uMjjm0mzUkRb0e4Ac_JK8aJ9cKGUi5UqmzWoWAFJprE,4164
 keras_hub/utils/__init__.py,sha256=jXPqVGBpJr_PpYmqD8aDG-fRMlxH-ulqCR2SZMn288Y,646
-keras_hub_nightly-0.22.0.dev202505290412.dist-info/METADATA,sha256=W4vT73-ho1j4QwQv59qS5xF4i6bWH5k7tHiUJ7-_y4k,7393
-keras_hub_nightly-0.22.0.dev202505290412.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-keras_hub_nightly-0.22.0.dev202505290412.dist-info/top_level.txt,sha256=N4J6piIWBKa38A4uV-CnIopnOEf8mHAbkNXafXm_CuA,10
-keras_hub_nightly-0.22.0.dev202505290412.dist-info/RECORD,,
+keras_hub_nightly-0.22.0.dev202505310408.dist-info/METADATA,sha256=v4Rvzln90tKecsbiwiU29ZFrct9xpLCV10RQDme4-DI,7393
+keras_hub_nightly-0.22.0.dev202505310408.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+keras_hub_nightly-0.22.0.dev202505310408.dist-info/top_level.txt,sha256=N4J6piIWBKa38A4uV-CnIopnOEf8mHAbkNXafXm_CuA,10
+keras_hub_nightly-0.22.0.dev202505310408.dist-info/RECORD,,